diff --git a/.github/workflows/build-package.yaml b/.github/workflows/build-package.yaml new file mode 100644 index 000000000..7f4fee53e --- /dev/null +++ b/.github/workflows/build-package.yaml @@ -0,0 +1,34 @@ +name: Build and upload python package + +on: + release: + types: [created] + +jobs: + deploy: + runs-on: ubuntu-latest + + steps: + - uses: actions/checkout@v4 + - name: Set up Python + uses: actions/setup-python@v4 + with: + python-version: '3.9' + cache: 'pip' + - name: Install dependencies + run: | + python -m pip install --upgrade pip + pip install -r requirements.txt + pip install -e. + pip install setuptools wheel twine + - name: Set package version + run: | + export VERSION="${GITHUB_REF#refs/tags/v}" + sed -i "s/version=.*/version=\"${VERSION}\",/" setup.py + - name: Build and publish + env: + TWINE_USERNAME: __token__ + TWINE_PASSWORD: ${{ secrets.PYPI_API_TOKEN }} + run: | + python setup.py bdist_wheel sdist + twine upload dist/* \ No newline at end of file diff --git a/.github/workflows/unittest.yaml b/.github/workflows/unittest.yaml index fd56c42fb..87ccbf144 100644 --- a/.github/workflows/unittest.yaml +++ b/.github/workflows/unittest.yaml @@ -50,6 +50,7 @@ jobs: run: | export ALLOW_OPENAI_API_CALL=0 echo "${{ secrets.METAGPT_KEY_YAML }}" | base64 -d > config/key.yaml + mkdir -p ~/.metagpt && echo "${{ secrets.METAGPT_CONFIG2_YAML }}" | base64 -d > ~/.metagpt/config2.yaml pytest tests/ --doctest-modules --cov=./metagpt/ --cov-report=xml:cov.xml --cov-report=html:htmlcov --durations=20 | tee unittest.txt - name: Show coverage report run: | diff --git a/.gitignore b/.gitignore index a69b3b1c2..4752806d7 100644 --- a/.gitignore +++ b/.gitignore @@ -176,5 +176,6 @@ htmlcov.* cov.xml *.dot *.pkl +*.faiss *-structure.csv *-structure.json diff --git a/README.md b/README.md index 9c88c92a1..90c586068 100644 --- a/README.md +++ b/README.md @@ -6,16 +6,16 @@ # MetaGPT: The Multi-Agent Framework

-Assign different roles to GPTs to form a collaborative software entity for complex tasks. +Assign different roles to GPTs to form a collaborative entity for complex tasks.

CN doc EN doc JA doc -Discord Follow License: MIT roadmap +Discord Follow Twitter Follow

@@ -25,19 +25,31 @@ # MetaGPT: The Multi-Agent Framework Hugging Face

+## News +🚀 Jan. 16, 2024: Our paper [MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework +](https://arxiv.org/abs/2308.00352) accepted for oral presentation **(top 1.2%)** at ICLR 2024, **ranking #1** in the LLM-based Agent category. + +🚀 Jan. 03, 2024: [v0.6.0](https://github.com/geekan/MetaGPT/releases/tag/v0.6.0) released, new features include serialization, upgraded OpenAI package and supported multiple LLM, provided [minimal example for debate](https://github.com/geekan/MetaGPT/blob/main/examples/debate_simple.py) etc. + +🚀 Dec. 15, 2023: [v0.5.0](https://github.com/geekan/MetaGPT/releases/tag/v0.5.0) released, introducing some experimental features such as **incremental development**, **multilingual**, **multiple programming languages**, etc. + +🔥 Nov. 08, 2023: MetaGPT is selected into [Open100: Top 100 Open Source achievements](https://www.benchcouncil.org/evaluation/opencs/annual.html). + +🔥 Sep. 01, 2023: MetaGPT tops GitHub Trending Monthly for the **17th time** in August 2023. + +🌟 Jun. 30, 2023: MetaGPT is now open source. + +🌟 Apr. 24, 2023: First line of MetaGPT code committed. + +## Software Company as Multi-Agent System + 1. MetaGPT takes a **one line requirement** as input and outputs **user stories / competitive analysis / requirements / data structures / APIs / documents, etc.** 2. Internally, MetaGPT includes **product managers / architects / project managers / engineers.** It provides the entire process of a **software company along with carefully orchestrated SOPs.** 1. `Code = SOP(Team)` is the core philosophy. We materialize SOP and apply it to teams composed of LLMs. ![A software company consists of LLM-based roles](docs/resources/software_company_cd.jpeg) -

Software Company Multi-Role Schematic (Gradually Implementing)

- -## News -🚀 Jan 03: Here comes [v0.6.0](https://github.com/geekan/MetaGPT/releases/tag/v0.6.0)! In this version, we added serialization and deserialization of important objects and enabled breakpoint recovery. We upgraded OpenAI package to v1.6.0 and supported Gemini, ZhipuAI, Ollama, OpenLLM, etc. Moreover, we provided extremely simple examples where you need only 7 lines to implement a general election [debate](https://github.com/geekan/MetaGPT/blob/main/examples/debate_simple.py). Check out more details [here](https://github.com/geekan/MetaGPT/releases/tag/v0.6.0)! - - -🚀 Dec 15: [v0.5.0](https://github.com/geekan/MetaGPT/releases/tag/v0.5.0) is released! We introduced **incremental development**, facilitating agents to build up larger projects on top of their previous efforts or existing codebase. We also launched a whole collection of important features, including **multilingual support** (experimental), multiple **programming languages support** (experimental), **incremental development** (experimental), CLI support, pip support, enhanced code review, documentation mechanism, and optimized messaging mechanism! +

Software Company Multi-Agent Schematic (Gradually Implementing)

## Install diff --git a/config/config.yaml b/config/config.yaml deleted file mode 100644 index d8fab693e..000000000 --- a/config/config.yaml +++ /dev/null @@ -1,149 +0,0 @@ -# DO NOT MODIFY THIS FILE, create a new key.yaml, define OPENAI_API_KEY. -# The configuration of key.yaml has a higher priority and will not enter git - -#### Project Path Setting -# WORKSPACE_PATH: "Path for placing output files" - -#### if OpenAI -## The official OPENAI_BASE_URL is https://api.openai.com/v1 -## If the official OPENAI_BASE_URL is not available, we recommend using the [openai-forward](https://github.com/beidongjiedeguang/openai-forward). -## Or, you can configure OPENAI_PROXY to access official OPENAI_BASE_URL. -OPENAI_BASE_URL: "https://api.openai.com/v1" -#OPENAI_PROXY: "http://127.0.0.1:8118" -#OPENAI_API_KEY: "YOUR_API_KEY" # set the value to sk-xxx if you host the openai interface for open llm model -OPENAI_API_MODEL: "gpt-4-1106-preview" -MAX_TOKENS: 4096 -RPM: 10 -TIMEOUT: 60 # Timeout for llm invocation -#DEFAULT_PROVIDER: openai - -#### if Spark -#SPARK_APPID : "YOUR_APPID" -#SPARK_API_SECRET : "YOUR_APISecret" -#SPARK_API_KEY : "YOUR_APIKey" -#DOMAIN : "generalv2" -#SPARK_URL : "ws://spark-api.xf-yun.com/v2.1/chat" - -#### if Anthropic -#ANTHROPIC_API_KEY: "YOUR_API_KEY" - -#### if AZURE, check https://github.com/openai/openai-cookbook/blob/main/examples/azure/chat.ipynb -#OPENAI_API_TYPE: "azure" -#OPENAI_BASE_URL: "YOUR_AZURE_ENDPOINT" -#OPENAI_API_KEY: "YOUR_AZURE_API_KEY" -#OPENAI_API_VERSION: "YOUR_AZURE_API_VERSION" -#DEPLOYMENT_NAME: "YOUR_DEPLOYMENT_NAME" - -#### if zhipuai from `https://open.bigmodel.cn`. You can set here or export API_KEY="YOUR_API_KEY" -# ZHIPUAI_API_KEY: "YOUR_API_KEY" - -#### if Google Gemini from `https://ai.google.dev/` and API_KEY from `https://makersuite.google.com/app/apikey`. -#### You can set here or export GOOGLE_API_KEY="YOUR_API_KEY" -# GEMINI_API_KEY: "YOUR_API_KEY" - -#### if use self-host open llm model with openai-compatible interface -#OPEN_LLM_API_BASE: "http://127.0.0.1:8000/v1" -#OPEN_LLM_API_MODEL: "llama2-13b" -# -##### if use Fireworks api -#FIREWORKS_API_KEY: "YOUR_API_KEY" -#FIREWORKS_API_BASE: "https://api.fireworks.ai/inference/v1" -#FIREWORKS_API_MODEL: "YOUR_LLM_MODEL" # example, accounts/fireworks/models/llama-v2-13b-chat - -#### if use self-host open llm model by ollama -# OLLAMA_API_BASE: http://127.0.0.1:11434/api -# OLLAMA_API_MODEL: llama2 - -#### for Search - -## Supported values: serpapi/google/serper/ddg -#SEARCH_ENGINE: serpapi - -## Visit https://serpapi.com/ to get key. -#SERPAPI_API_KEY: "YOUR_API_KEY" - -## Visit https://console.cloud.google.com/apis/credentials to get key. -#GOOGLE_API_KEY: "YOUR_API_KEY" -## Visit https://programmablesearchengine.google.com/controlpanel/create to get id. -#GOOGLE_CSE_ID: "YOUR_CSE_ID" - -## Visit https://serper.dev/ to get key. -#SERPER_API_KEY: "YOUR_API_KEY" - -#### for web access - -## Supported values: playwright/selenium -#WEB_BROWSER_ENGINE: playwright - -## Supported values: chromium/firefox/webkit, visit https://playwright.dev/python/docs/api/class-browsertype -##PLAYWRIGHT_BROWSER_TYPE: chromium - -## Supported values: chrome/firefox/edge/ie, visit https://www.selenium.dev/documentation/webdriver/browsers/ -# SELENIUM_BROWSER_TYPE: chrome - -#### for TTS - -#AZURE_TTS_SUBSCRIPTION_KEY: "YOUR_API_KEY" -#AZURE_TTS_REGION: "eastus" - -#### for OPENAI VISION - -#OPENAI_VISION_MODEL: "YOUR_VISION_MODEL_NAME" -#VISION_MAX_TOKENS: 4096 - -#### for Stable Diffusion -## Use SD service, based on https://github.com/AUTOMATIC1111/stable-diffusion-webui -#SD_URL: "YOUR_SD_URL" -#SD_T2I_API: "/sdapi/v1/txt2img" - -#### for Execution -#LONG_TERM_MEMORY: false - -#### for Mermaid CLI -## If you installed mmdc (Mermaid CLI) only for metagpt then enable the following configuration. -#PUPPETEER_CONFIG: "./config/puppeteer-config.json" -#MMDC: "./node_modules/.bin/mmdc" - - -### for calc_usage -# CALC_USAGE: false - -### for Research -# MODEL_FOR_RESEARCHER_SUMMARY: gpt-3.5-turbo -# MODEL_FOR_RESEARCHER_REPORT: gpt-3.5-turbo-16k - -### choose the engine for mermaid conversion, -# default is nodejs, you can change it to playwright,pyppeteer or ink -# MERMAID_ENGINE: nodejs - -### browser path for pyppeteer engine, support Chrome, Chromium,MS Edge -#PYPPETEER_EXECUTABLE_PATH: "/usr/bin/google-chrome-stable" - -### for repair non-openai LLM's output when parse json-text if PROMPT_FORMAT=json -### due to non-openai LLM's output will not always follow the instruction, so here activate a post-process -### repair operation on the content extracted from LLM's raw output. Warning, it improves the result but not fix all cases. -# REPAIR_LLM_OUTPUT: false - -# PROMPT_FORMAT: json #json or markdown - -### Agent configurations -# RAISE_NOT_CONFIG_ERROR: true # "true" if the LLM key is not configured, throw a NotConfiguredException, else "false". -# WORKSPACE_PATH_WITH_UID: false # "true" if using `{workspace}/{uid}` as the workspace path; "false" use `{workspace}`. - -### Meta Models -#METAGPT_TEXT_TO_IMAGE_MODEL: MODEL_URL - -### S3 config -#S3_ACCESS_KEY: "YOUR_S3_ACCESS_KEY" -#S3_SECRET_KEY: "YOUR_S3_SECRET_KEY" -#S3_ENDPOINT_URL: "YOUR_S3_ENDPOINT_URL" -#S3_SECURE: true # true/false -#S3_BUCKET: "YOUR_S3_BUCKET" - -### Redis config -#REDIS_HOST: "YOUR_REDIS_HOST" -#REDIS_PORT: "YOUR_REDIS_PORT" -#REDIS_PASSWORD: "YOUR_REDIS_PASSWORD" -#REDIS_DB: "YOUR_REDIS_DB_INDEX, str, 0-based" - -# DISABLE_LLM_PROVIDER_CHECK: false diff --git a/config/config2.yaml b/config/config2.yaml new file mode 100644 index 000000000..5e7f34809 --- /dev/null +++ b/config/config2.yaml @@ -0,0 +1,3 @@ +llm: + api_key: "YOUR_API_KEY" + model: "gpt-3.5-turbo-1106" \ No newline at end of file diff --git a/config/config2.yaml.example b/config/config2.yaml.example new file mode 100644 index 000000000..35575e5a5 --- /dev/null +++ b/config/config2.yaml.example @@ -0,0 +1,42 @@ +llm: + api_type: "openai" + base_url: "YOUR_BASE_URL" + api_key: "YOUR_API_KEY" + model: "gpt-3.5-turbo-1106" # or gpt-4-1106-preview + +proxy: "YOUR_PROXY" + +search: + api_type: "google" + api_key: "YOUR_API_KEY" + cse_id: "YOUR_CSE_ID" + +mermaid: + engine: "pyppeteer" + path: "/Applications/Google Chrome.app" + +redis: + host: "YOUR_HOST" + port: 32582 + password: "YOUR_PASSWORD" + db: "0" + +s3: + access_key: "YOUR_ACCESS_KEY" + secret_key: "YOUR_SECRET_KEY" + endpoint: "YOUR_ENDPOINT" + secure: false + bucket: "test" + + +AZURE_TTS_SUBSCRIPTION_KEY: "YOUR_SUBSCRIPTION_KEY" +AZURE_TTS_REGION: "eastus" + +IFLYTEK_APP_ID: "YOUR_APP_ID" +IFLYTEK_API_KEY: "YOUR_API_KEY" +IFLYTEK_API_SECRET: "YOUR_API_SECRET" + +METAGPT_TEXT_TO_IMAGE_MODEL_URL: "YOUR_MODEL_URL" + +PYPPETEER_EXECUTABLE_PATH: "/Applications/Google Chrome.app" + diff --git a/docs/ROADMAP.md b/docs/ROADMAP.md index d3f7ea408..4bb530bf2 100644 --- a/docs/ROADMAP.md +++ b/docs/ROADMAP.md @@ -9,24 +9,22 @@ ### Short-term Objective 1. Become the multi-agent framework with the highest ROI. 2. Support fully automatic implementation of medium-sized projects (around 2000 lines of code). -3. Implement most identified tasks, reaching version 0.5. +3. Implement most identified tasks, reaching version 1.0. ### Tasks -To reach version v0.5, approximately 70% of the following tasks need to be completed. - 1. Usability 1. ~~Release v0.01 pip package to try to solve issues like npm installation (though not necessarily successfully)~~ (v0.3.0) - 2. Support for overall save and recovery of software companies + 2. ~~Support for overall save and recovery of software companies~~ (v0.6.0) 3. ~~Support human confirmation and modification during the process~~ (v0.3.0) New: Support human confirmation and modification with fewer constrainsts and a more user-friendly interface 4. Support process caching: Consider carefully whether to add server caching mechanism 5. ~~Resolve occasional failure to follow instruction under current prompts, causing code parsing errors, through stricter system prompts~~ (v0.4.0, with function call) 6. Write documentation, describing the current features and usage at all levels (ongoing, continuously adding contents to [documentation site](https://docs.deepwisdom.ai/main/en/guide/get_started/introduction.html)) 7. ~~Support Docker~~ 2. Features - 1. Support a more standard and stable parser (need to analyze the format that the current LLM is better at) - 2. ~~Establish a separate output queue, differentiated from the message queue~~ - 3. Attempt to atomize all role work, but this may significantly increase token overhead + 1. ~~Support a more standard and stable parser (need to analyze the format that the current LLM is better at)~~ (v0.5.0) + 2. ~~Establish a separate output queue, differentiated from the message queue~~ (v0.5.0) + 3. ~~Attempt to atomize all role work, but this may significantly increase token overhead~~ (v0.5.0) 4. Complete the design and implementation of module breakdown 5. Support various modes of memory: clearly distinguish between long-term and short-term memory 6. Perfect the test role, and carry out necessary interactions with humans @@ -43,10 +41,10 @@ ### Tasks 4. Actions 1. ~~Implementation: Search~~ (v0.2.1) 2. Implementation: Knowledge search, supporting 10+ data formats - 3. Implementation: Data EDA (expected v0.6.0) - 4. Implementation: Review - 5. ~~Implementation~~: Add Document (v0.5.0) - 6. ~~Implementation~~: Delete Document (v0.5.0) + 3. Implementation: Data EDA (expected v0.7.0) + 4. Implementation: Review & Revise (expected v0.7.0) + 5. ~~Implementation: Add Document~~ (v0.5.0) + 6. ~~Implementation: Delete Document~~ (v0.5.0) 7. Implementation: Self-training 8. ~~Implementation: DebugError~~ (v0.2.1) 9. Implementation: Generate reliable unit tests based on YAPI @@ -64,15 +62,14 @@ ### Tasks 3. ~~Support Playwright apis~~ 7. Roles 1. Perfect the action pool/skill pool for each role - 2. Red Book blogger - 3. E-commerce seller - 4. Data analyst (expected v0.6.0) - 5. News observer - 6. ~~Institutional researcher~~ (v0.2.1) + 2. E-commerce seller + 3. Data analyst (expected v0.7.0) + 4. News observer + 5. ~~Institutional researcher~~ (v0.2.1) 8. Evaluation 1. Support an evaluation on a game dataset (experimentation done with game agents) 2. Reproduce papers, implement full skill acquisition for a single game role, achieving SOTA results (experimentation done with game agents) - 3. Support an evaluation on a math dataset (expected v0.6.0) + 3. Support an evaluation on a math dataset (expected v0.7.0) 4. Reproduce papers, achieving SOTA results for current mathematical problem solving process 9. LLM 1. Support Claude underlying API @@ -80,7 +77,7 @@ ### Tasks 3. Support streaming version of all APIs 4. ~~Make gpt-3.5-turbo available (HARD)~~ 10. Other - 1. Clean up existing unused code - 2. Unify all code styles and establish contribution standards - 3. Multi-language support - 4. Multi-programming-language support + 1. ~~Clean up existing unused code~~ + 2. ~~Unify all code styles and establish contribution standards~~ + 3. ~~Multi-language support~~ + 4. ~~Multi-programming-language support~~ diff --git a/examples/agent_creator.py b/examples/agent_creator.py index 340dfafa4..bd58840ce 100644 --- a/examples/agent_creator.py +++ b/examples/agent_creator.py @@ -6,7 +6,7 @@ Author: garylin2099 import re from metagpt.actions import Action -from metagpt.config import CONFIG +from metagpt.config2 import config from metagpt.const import METAGPT_ROOT from metagpt.logs import logger from metagpt.roles import Role @@ -48,8 +48,8 @@ class CreateAgent(Action): pattern = r"```python(.*)```" match = re.search(pattern, rsp, re.DOTALL) code_text = match.group(1) if match else "" - CONFIG.workspace_path.mkdir(parents=True, exist_ok=True) - new_file = CONFIG.workspace_path / "agent_created_agent.py" + config.workspace.path.mkdir(parents=True, exist_ok=True) + new_file = config.workspace.path / "agent_created_agent.py" new_file.write_text(code_text) return code_text @@ -61,7 +61,7 @@ class AgentCreator(Role): def __init__(self, **kwargs): super().__init__(**kwargs) - self._init_actions([CreateAgent]) + self.set_actions([CreateAgent]) async def _act(self) -> Message: logger.info(f"{self._setting}: to do {self.rc.todo}({self.rc.todo.name})") diff --git a/examples/build_customized_agent.py b/examples/build_customized_agent.py index 6c3219efc..cfe264b47 100644 --- a/examples/build_customized_agent.py +++ b/examples/build_customized_agent.py @@ -57,7 +57,7 @@ class SimpleCoder(Role): def __init__(self, **kwargs): super().__init__(**kwargs) - self._init_actions([SimpleWriteCode]) + self.set_actions([SimpleWriteCode]) async def _act(self) -> Message: logger.info(f"{self._setting}: to do {self.rc.todo}({self.rc.todo.name})") @@ -76,7 +76,7 @@ class RunnableCoder(Role): def __init__(self, **kwargs): super().__init__(**kwargs) - self._init_actions([SimpleWriteCode, SimpleRunCode]) + self.set_actions([SimpleWriteCode, SimpleRunCode]) self._set_react_mode(react_mode=RoleReactMode.BY_ORDER.value) async def _act(self) -> Message: diff --git a/examples/build_customized_multi_agents.py b/examples/build_customized_multi_agents.py index 73278c08c..296323cea 100644 --- a/examples/build_customized_multi_agents.py +++ b/examples/build_customized_multi_agents.py @@ -46,7 +46,7 @@ class SimpleCoder(Role): def __init__(self, **kwargs): super().__init__(**kwargs) self._watch([UserRequirement]) - self._init_actions([SimpleWriteCode]) + self.set_actions([SimpleWriteCode]) class SimpleWriteTest(Action): @@ -75,7 +75,7 @@ class SimpleTester(Role): def __init__(self, **kwargs): super().__init__(**kwargs) - self._init_actions([SimpleWriteTest]) + self.set_actions([SimpleWriteTest]) # self._watch([SimpleWriteCode]) self._watch([SimpleWriteCode, SimpleWriteReview]) # feel free to try this too @@ -114,7 +114,7 @@ class SimpleReviewer(Role): def __init__(self, **kwargs): super().__init__(**kwargs) - self._init_actions([SimpleWriteReview]) + self.set_actions([SimpleWriteReview]) self._watch([SimpleWriteTest]) diff --git a/examples/crawl_webpage.py b/examples/crawl_webpage.py new file mode 100644 index 000000000..35413d2ff --- /dev/null +++ b/examples/crawl_webpage.py @@ -0,0 +1,22 @@ +# -*- encoding: utf-8 -*- +""" +@Date : 2024/01/24 15:11:27 +@Author : orange-crow +@File : crawl_webpage.py +""" + +from metagpt.roles.code_interpreter import CodeInterpreter + + +async def main(): + prompt = """Get data from `paperlist` table in https://papercopilot.com/statistics/iclr-statistics/iclr-2024-statistics/, + and save it to a csv file. paper title must include `multiagent` or `large language model`. *notice: print key data*""" + ci = CodeInterpreter(goal=prompt, use_tools=True) + + await ci.run(prompt) + + +if __name__ == "__main__": + import asyncio + + asyncio.run(main()) diff --git a/examples/debate.py b/examples/debate.py index eb0a09839..72ab8796d 100644 --- a/examples/debate.py +++ b/examples/debate.py @@ -49,7 +49,7 @@ class Debator(Role): def __init__(self, **data: Any): super().__init__(**data) - self._init_actions([SpeakAloud]) + self.set_actions([SpeakAloud]) self._watch([UserRequirement, SpeakAloud]) async def _observe(self) -> int: diff --git a/examples/debate_simple.py b/examples/debate_simple.py index aa95c5b85..869e02a0e 100644 --- a/examples/debate_simple.py +++ b/examples/debate_simple.py @@ -13,7 +13,9 @@ from metagpt.roles import Role from metagpt.team import Team action1 = Action(name="AlexSay", instruction="Express your opinion with emotion and don't repeat it") +action1.llm.model = "gpt-4-1106-preview" action2 = Action(name="BobSay", instruction="Express your opinion with emotion and don't repeat it") +action2.llm.model = "gpt-3.5-turbo-1106" alex = Role(name="Alex", profile="Democratic candidate", goal="Win the election", actions=[action1], watch=[action2]) bob = Role(name="Bob", profile="Republican candidate", goal="Win the election", actions=[action2], watch=[action1]) env = Environment(desc="US election live broadcast") diff --git a/examples/example.faiss b/examples/example.faiss deleted file mode 100644 index 580946190..000000000 Binary files a/examples/example.faiss and /dev/null differ diff --git a/examples/example.pkl b/examples/example.pkl deleted file mode 100644 index f706fd803..000000000 Binary files a/examples/example.pkl and /dev/null differ diff --git a/examples/llm_hello_world.py b/examples/llm_hello_world.py index 76be1cc90..219a303c8 100644 --- a/examples/llm_hello_world.py +++ b/examples/llm_hello_world.py @@ -23,6 +23,10 @@ async def main(): # streaming mode, much slower await llm.acompletion_text(hello_msg, stream=True) + # check completion if exist to test llm complete functions + if hasattr(llm, "completion"): + logger.info(llm.completion(hello_msg)) + if __name__ == "__main__": asyncio.run(main()) diff --git a/examples/search_kb.py b/examples/search_kb.py index 0e0e0ffd0..995720cc1 100644 --- a/examples/search_kb.py +++ b/examples/search_kb.py @@ -8,7 +8,7 @@ import asyncio from langchain.embeddings import OpenAIEmbeddings -from metagpt.config import CONFIG +from metagpt.config2 import config from metagpt.const import DATA_PATH, EXAMPLE_PATH from metagpt.document_store import FaissStore from metagpt.logs import logger @@ -16,7 +16,8 @@ from metagpt.roles import Sales def get_store(): - embedding = OpenAIEmbeddings(openai_api_key=CONFIG.openai_api_key, openai_api_base=CONFIG.openai_base_url) + llm = config.get_openai_llm() + embedding = OpenAIEmbeddings(openai_api_key=llm.api_key, openai_api_base=llm.base_url) return FaissStore(DATA_PATH / "example.json", embedding=embedding) diff --git a/metagpt/actions/action.py b/metagpt/actions/action.py index b586bcc22..1b93213f7 100644 --- a/metagpt/actions/action.py +++ b/metagpt/actions/action.py @@ -10,41 +10,67 @@ from __future__ import annotations from typing import Optional, Union -from pydantic import ConfigDict, Field, model_validator +from pydantic import BaseModel, ConfigDict, Field, model_validator from metagpt.actions.action_node import ActionNode -from metagpt.llm import LLM -from metagpt.provider.base_llm import BaseLLM +from metagpt.context_mixin import ContextMixin from metagpt.schema import ( + CodePlanAndChangeContext, CodeSummarizeContext, CodingContext, RunCodeContext, SerializationMixin, TestingContext, ) +from metagpt.utils.project_repo import ProjectRepo -class Action(SerializationMixin, is_polymorphic_base=True): - model_config = ConfigDict(arbitrary_types_allowed=True, exclude=["llm"]) +class Action(SerializationMixin, ContextMixin, BaseModel): + model_config = ConfigDict(arbitrary_types_allowed=True) name: str = "" - llm: BaseLLM = Field(default_factory=LLM, exclude=True) - context: Union[dict, CodingContext, CodeSummarizeContext, TestingContext, RunCodeContext, str, None] = "" + i_context: Union[ + dict, CodingContext, CodeSummarizeContext, TestingContext, RunCodeContext, CodePlanAndChangeContext, str, None + ] = "" prefix: str = "" # aask*时会加上prefix，作为system_message desc: str = "" # for skill manager node: ActionNode = Field(default=None, exclude=True) + @property + def repo(self) -> ProjectRepo: + if not self.context.repo: + self.context.repo = ProjectRepo(self.context.git_repo) + return self.context.repo + + @property + def prompt_schema(self): + return self.config.prompt_schema + + @property + def project_name(self): + return self.config.project_name + + @project_name.setter + def project_name(self, value): + self.config.project_name = value + + @property + def project_path(self): + return self.config.project_path + @model_validator(mode="before") + @classmethod def set_name_if_empty(cls, values): if "name" not in values or not values["name"]: values["name"] = cls.__name__ return values @model_validator(mode="before") + @classmethod def _init_with_instruction(cls, values): if "instruction" in values: name = values["name"] - i = values["instruction"] + i = values.pop("instruction") values["node"] = ActionNode(key=name, expected_type=str, instruction=i, example="", schema="raw") return values diff --git a/metagpt/actions/action_node.py b/metagpt/actions/action_node.py index 6c65b33ef..162ab90eb 100644 --- a/metagpt/actions/action_node.py +++ b/metagpt/actions/action_node.py @@ -9,16 +9,30 @@ NOTE: You should use typing.List instead of list to do type annotation. Because we can use typing to extract the type of the node, but we cannot use built-in list to extract. """ import json -from typing import Any, Dict, List, Optional, Tuple, Type +from enum import Enum +from typing import Any, Dict, List, Optional, Tuple, Type, Union -from pydantic import BaseModel, create_model, model_validator +from pydantic import BaseModel, Field, create_model, model_validator from tenacity import retry, stop_after_attempt, wait_random_exponential -from metagpt.config import CONFIG +from metagpt.actions.action_outcls_registry import register_action_outcls from metagpt.llm import BaseLLM from metagpt.logs import logger from metagpt.provider.postprocess.llm_output_postprocess import llm_output_postprocess from metagpt.utils.common import OutputParser, general_after_log +from metagpt.utils.human_interaction import HumanInteraction + + +class ReviewMode(Enum): + HUMAN = "human" + AUTO = "auto" + + +class ReviseMode(Enum): + HUMAN = "human" # human revise + HUMAN_REVIEW = "human_review" # human-review and auto-revise + AUTO = "auto" # auto-review and auto-revise + TAG = "CONTENT" @@ -45,6 +59,58 @@ SIMPLE_TEMPLATE = """ Follow instructions of nodes, generate output and make sure it follows the format example. """ +REVIEW_TEMPLATE = """ +## context +Compare the key's value of nodes_output and the corresponding requirements one by one. If a key's value that does not match the requirement is found, provide the comment content on how to modify it. No output is required for matching keys. + +### nodes_output +{nodes_output} + +----- + +## format example +[{tag}] +{{ + "key1": "comment1", + "key2": "comment2", + "keyn": "commentn" +}} +[/{tag}] + +## nodes: ": # " +- key1: # the first key name of mismatch key +- key2: # the second key name of mismatch key +- keyn: # the last key name of mismatch key + +## constraint +{constraint} + +## action +Follow format example's {prompt_schema} format, generate output and make sure it follows the format example. +""" + +REVISE_TEMPLATE = """ +## context +change the nodes_output key's value to meet its comment and no need to add extra comment. + +### nodes_output +{nodes_output} + +----- + +## format example +{example} + +## nodes: ": # " +{instruction} + +## constraint +{constraint} + +## action +Follow format example's {prompt_schema} format, generate output and make sure it follows the format example. +""" + def dict_to_markdown(d, prefix="- ", kv_sep="\n", postfix="\n"): markdown_str = "" @@ -105,6 +171,9 @@ class ActionNode: """增加子ActionNode""" self.children[node.key] = node + def get_child(self, key: str) -> Union["ActionNode", None]: + return self.children.get(key, None) + def add_children(self, nodes: List["ActionNode"]): """批量增加子ActionNode""" for node in nodes: @@ -117,11 +186,27 @@ class ActionNode: obj.add_children(nodes) return obj - def get_children_mapping(self, exclude=None) -> Dict[str, Tuple[Type, Any]]: + def get_children_mapping_old(self, exclude=None) -> Dict[str, Tuple[Type, Any]]: """获得子ActionNode的字典，以key索引""" exclude = exclude or [] return {k: (v.expected_type, ...) for k, v in self.children.items() if k not in exclude} + def get_children_mapping(self, exclude=None) -> Dict[str, Tuple[Type, Any]]: + """获得子ActionNode的字典，以key索引，支持多级结构""" + exclude = exclude or [] + mapping = {} + + def _get_mapping(node: "ActionNode", prefix: str = ""): + for key, child in node.children.items(): + if key in exclude: + continue + full_key = f"{prefix}{key}" + mapping[full_key] = (child.expected_type, ...) + _get_mapping(child, prefix=f"{full_key}.") + + _get_mapping(self) + return mapping + def get_self_mapping(self) -> Dict[str, Tuple[Type, Any]]: """get self key: type mapping""" return {self.key: (self.expected_type, ...)} @@ -133,6 +218,7 @@ class ActionNode: return {} if exclude and self.key in exclude else self.get_self_mapping() @classmethod + @register_action_outcls def create_model_class(cls, class_name: str, mapping: Dict[str, Tuple[Type, Any]]): """基于pydantic v1的模型动态生成，用来检验结果类型正确性""" @@ -152,6 +238,11 @@ class ActionNode: new_class = create_model(class_name, __validators__=validators, **mapping) return new_class + def create_class(self, mode: str = "auto", class_name: str = None, exclude=None): + class_name = class_name if class_name else f"{self.key}_AN" + mapping = self.get_mapping(mode=mode, exclude=exclude) + return self.create_model_class(class_name, mapping) + def create_children_class(self, exclude=None): """使用object内有的字段直接生成model_class""" class_name = f"{self.key}_AN" @@ -186,6 +277,25 @@ class ActionNode: return node_dict + def update_instruct_content(self, incre_data: dict[str, Any]): + assert self.instruct_content + origin_sc_dict = self.instruct_content.model_dump() + origin_sc_dict.update(incre_data) + output_class = self.create_class() + self.instruct_content = output_class(**origin_sc_dict) + + def keys(self, mode: str = "auto") -> list: + if mode == "children" or (mode == "auto" and self.children): + keys = [] + else: + keys = [self.key] + if mode == "root": + return keys + + for _, child_node in self.children.items(): + keys.append(child_node.key) + return keys + def compile_to(self, i: Dict, schema, kv_sep) -> str: if schema == "json": return json.dumps(i, indent=4) @@ -262,7 +372,7 @@ class ActionNode: output_data_mapping: dict, system_msgs: Optional[list[str]] = None, schema="markdown", # compatible to original format - timeout=CONFIG.timeout, + timeout=3, ) -> (str, BaseModel): """Use ActionOutput to wrap the output of aask""" content = await self.llm.aask(prompt, system_msgs, timeout=timeout) @@ -294,7 +404,7 @@ class ActionNode: def set_context(self, context): self.set_recursive("context", context) - async def simple_fill(self, schema, mode, timeout=CONFIG.timeout, exclude=None): + async def simple_fill(self, schema, mode, timeout=3, exclude=None): prompt = self.compile(context=self.context, schema=schema, mode=mode, exclude=exclude) if schema != "raw": @@ -309,7 +419,7 @@ class ActionNode: return self - async def fill(self, context, llm, schema="json", mode="auto", strgy="simple", timeout=CONFIG.timeout, exclude=[]): + async def fill(self, context, llm, schema="json", mode="auto", strgy="simple", timeout=3, exclude=[]): """Fill the node(s) with mode. :param context: Everything we should know when filling node. @@ -343,7 +453,241 @@ class ActionNode: if exclude and i.key in exclude: continue child = await i.simple_fill(schema=schema, mode=mode, timeout=timeout, exclude=exclude) - tmp.update(child.instruct_content.dict()) + tmp.update(child.instruct_content.model_dump()) cls = self.create_children_class() self.instruct_content = cls(**tmp) return self + + async def human_review(self) -> dict[str, str]: + review_comments = HumanInteraction().interact_with_instruct_content( + instruct_content=self.instruct_content, interact_type="review" + ) + + return review_comments + + def _makeup_nodes_output_with_req(self) -> dict[str, str]: + instruct_content_dict = self.instruct_content.model_dump() + nodes_output = {} + for key, value in instruct_content_dict.items(): + child = self.get_child(key) + nodes_output[key] = {"value": value, "requirement": child.instruction if child else self.instruction} + return nodes_output + + async def auto_review(self, template: str = REVIEW_TEMPLATE) -> dict[str, str]: + """use key's output value and its instruction to review the modification comment""" + nodes_output = self._makeup_nodes_output_with_req() + """nodes_output format: + { + "key": {"value": "output value", "requirement": "key instruction"} + } + """ + if not nodes_output: + return dict() + + prompt = template.format( + nodes_output=json.dumps(nodes_output, ensure_ascii=False), + tag=TAG, + constraint=FORMAT_CONSTRAINT, + prompt_schema="json", + ) + + content = await self.llm.aask(prompt) + # Extract the dict of mismatch key and its comment. Due to the mismatch keys are unknown, here use the keys + # of ActionNode to judge if exist in `content` and then follow the `data_mapping` method to create model class. + keys = self.keys() + include_keys = [] + for key in keys: + if f'"{key}":' in content: + include_keys.append(key) + if not include_keys: + return dict() + + exclude_keys = list(set(keys).difference(include_keys)) + output_class_name = f"{self.key}_AN_REVIEW" + output_class = self.create_class(class_name=output_class_name, exclude=exclude_keys) + parsed_data = llm_output_postprocess( + output=content, schema=output_class.model_json_schema(), req_key=f"[/{TAG}]" + ) + instruct_content = output_class(**parsed_data) + return instruct_content.model_dump() + + async def simple_review(self, review_mode: ReviewMode = ReviewMode.AUTO): + # generate review comments + if review_mode == ReviewMode.HUMAN: + review_comments = await self.human_review() + else: + review_comments = await self.auto_review() + + if not review_comments: + logger.warning("There are no review comments") + return review_comments + + async def review(self, strgy: str = "simple", review_mode: ReviewMode = ReviewMode.AUTO): + """only give the review comment of each exist and mismatch key + + :param strgy: simple/complex + - simple: run only once + - complex: run each node + """ + if not hasattr(self, "llm"): + raise RuntimeError("use `review` after `fill`") + assert review_mode in ReviewMode + assert self.instruct_content, 'review only support with `schema != "raw"`' + + if strgy == "simple": + review_comments = await self.simple_review(review_mode) + elif strgy == "complex": + # review each child node one-by-one + review_comments = {} + for _, child in self.children.items(): + child_review_comment = await child.simple_review(review_mode) + review_comments.update(child_review_comment) + + return review_comments + + async def human_revise(self) -> dict[str, str]: + review_contents = HumanInteraction().interact_with_instruct_content( + instruct_content=self.instruct_content, mapping=self.get_mapping(mode="auto"), interact_type="revise" + ) + # re-fill the ActionNode + self.update_instruct_content(review_contents) + return review_contents + + def _makeup_nodes_output_with_comment(self, review_comments: dict[str, str]) -> dict[str, str]: + instruct_content_dict = self.instruct_content.model_dump() + nodes_output = {} + for key, value in instruct_content_dict.items(): + if key in review_comments: + nodes_output[key] = {"value": value, "comment": review_comments[key]} + return nodes_output + + async def auto_revise( + self, revise_mode: ReviseMode = ReviseMode.AUTO, template: str = REVISE_TEMPLATE + ) -> dict[str, str]: + """revise the value of incorrect keys""" + # generate review comments + if revise_mode == ReviseMode.AUTO: + review_comments: dict = await self.auto_review() + elif revise_mode == ReviseMode.HUMAN_REVIEW: + review_comments: dict = await self.human_review() + + include_keys = list(review_comments.keys()) + + # generate revise content, two-steps + # step1, find the needed revise keys from review comments to makeup prompt template + nodes_output = self._makeup_nodes_output_with_comment(review_comments) + keys = self.keys() + exclude_keys = list(set(keys).difference(include_keys)) + example = self.compile_example(schema="json", mode="auto", tag=TAG, exclude=exclude_keys) + instruction = self.compile_instruction(schema="markdown", mode="auto", exclude=exclude_keys) + + prompt = template.format( + nodes_output=json.dumps(nodes_output, ensure_ascii=False), + example=example, + instruction=instruction, + constraint=FORMAT_CONSTRAINT, + prompt_schema="json", + ) + + # step2, use `_aask_v1` to get revise structure result + output_mapping = self.get_mapping(mode="auto", exclude=exclude_keys) + output_class_name = f"{self.key}_AN_REVISE" + content, scontent = await self._aask_v1( + prompt=prompt, output_class_name=output_class_name, output_data_mapping=output_mapping, schema="json" + ) + + # re-fill the ActionNode + sc_dict = scontent.model_dump() + self.update_instruct_content(sc_dict) + return sc_dict + + async def simple_revise(self, revise_mode: ReviseMode = ReviseMode.AUTO) -> dict[str, str]: + if revise_mode == ReviseMode.HUMAN: + revise_contents = await self.human_revise() + else: + revise_contents = await self.auto_revise(revise_mode) + + return revise_contents + + async def revise(self, strgy: str = "simple", revise_mode: ReviseMode = ReviseMode.AUTO) -> dict[str, str]: + """revise the content of ActionNode and update the instruct_content + + :param strgy: simple/complex + - simple: run only once + - complex: run each node + """ + if not hasattr(self, "llm"): + raise RuntimeError("use `revise` after `fill`") + assert revise_mode in ReviseMode + assert self.instruct_content, 'revise only support with `schema != "raw"`' + + if strgy == "simple": + revise_contents = await self.simple_revise(revise_mode) + elif strgy == "complex": + # revise each child node one-by-one + revise_contents = {} + for _, child in self.children.items(): + child_revise_content = await child.simple_revise(revise_mode) + revise_contents.update(child_revise_content) + self.update_instruct_content(revise_contents) + + return revise_contents + + @classmethod + def from_pydantic(cls, model: Type[BaseModel], key: str = None): + """ + Creates an ActionNode tree from a Pydantic model. + + Args: + model (Type[BaseModel]): The Pydantic model to convert. + + Returns: + ActionNode: The root node of the created ActionNode tree. + """ + key = key or model.__name__ + root_node = cls(key=model.__name__, expected_type=Type[model], instruction="", example="") + + for field_name, field_model in model.model_fields.items(): + # Extracting field details + expected_type = field_model.annotation + instruction = field_model.description or "" + example = field_model.default + + # Check if the field is a Pydantic model itself. + # Use isinstance to avoid typing.List, typing.Dict, etc. (they are instances of type, not subclasses) + if isinstance(expected_type, type) and issubclass(expected_type, BaseModel): + # Recursively process the nested model + child_node = cls.from_pydantic(expected_type, key=field_name) + else: + child_node = cls(key=field_name, expected_type=expected_type, instruction=instruction, example=example) + + root_node.add_child(child_node) + + return root_node + + +class ToolUse(BaseModel): + tool_name: str = Field(default="a", description="tool name", examples=[]) + + +class Task(BaseModel): + task_id: int = Field(default="1", description="task id", examples=[1, 2, 3]) + name: str = Field(default="Get data from ...", description="task name", examples=[]) + dependent_task_ids: List[int] = Field(default=[], description="dependent task ids", examples=[1, 2, 3]) + tool: ToolUse = Field(default=ToolUse(), description="tool use", examples=[]) + + +class Tasks(BaseModel): + tasks: List[Task] = Field(default=[], description="tasks", examples=[]) + + +if __name__ == "__main__": + node = ActionNode.from_pydantic(Tasks) + print("Tasks") + print(Tasks.model_json_schema()) + print("Task") + print(Task.model_json_schema()) + print(node) + prompt = node.compile(context="") + node.create_children_class() + print(prompt) diff --git a/metagpt/actions/action_outcls_registry.py b/metagpt/actions/action_outcls_registry.py new file mode 100644 index 000000000..6baa4cea9 --- /dev/null +++ b/metagpt/actions/action_outcls_registry.py @@ -0,0 +1,42 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +# @Desc : registry to store Dynamic Model from ActionNode.create_model_class to keep it as same Class +# with same class name and mapping + +from functools import wraps + +action_outcls_registry = dict() + + +def register_action_outcls(func): + """ + Due to `create_model` return different Class even they have same class name and mapping. + In order to do a comparison, use outcls_id to identify same Class with same class name and field definition + """ + + @wraps(func) + def decorater(*args, **kwargs): + """ + arr example + [, 'test', {'field': (str, Ellipsis)}] + """ + arr = list(args) + list(kwargs.values()) + """ + outcls_id example + "_test_{'field': (str, Ellipsis)}" + """ + for idx, item in enumerate(arr): + if isinstance(item, dict): + arr[idx] = dict(sorted(item.items())) + outcls_id = "_".join([str(i) for i in arr]) + # eliminate typing influence + outcls_id = outcls_id.replace("typing.List", "list").replace("typing.Dict", "dict") + + if outcls_id in action_outcls_registry: + return action_outcls_registry[outcls_id] + + out_cls = func(*args, **kwargs) + action_outcls_registry[outcls_id] = out_cls + return out_cls + + return decorater diff --git a/metagpt/actions/debug_error.py b/metagpt/actions/debug_error.py index 34f784072..5ed31bed8 100644 --- a/metagpt/actions/debug_error.py +++ b/metagpt/actions/debug_error.py @@ -13,12 +13,9 @@ import re from pydantic import Field from metagpt.actions.action import Action -from metagpt.config import CONFIG -from metagpt.const import TEST_CODES_FILE_REPO, TEST_OUTPUTS_FILE_REPO from metagpt.logs import logger from metagpt.schema import RunCodeContext, RunCodeResult from metagpt.utils.common import CodeParser -from metagpt.utils.file_repository import FileRepository PROMPT_TEMPLATE = """ NOTICE @@ -49,13 +46,10 @@ Now you should start rewriting the code: class DebugError(Action): - name: str = "DebugError" - context: RunCodeContext = Field(default_factory=RunCodeContext) + i_context: RunCodeContext = Field(default_factory=RunCodeContext) async def run(self, *args, **kwargs) -> str: - output_doc = await FileRepository.get_file( - filename=self.context.output_filename, relative_path=TEST_OUTPUTS_FILE_REPO - ) + output_doc = await self.repo.test_outputs.get(filename=self.i_context.output_filename) if not output_doc: return "" output_detail = RunCodeResult.loads(output_doc.content) @@ -64,15 +58,13 @@ class DebugError(Action): if matches: return "" - logger.info(f"Debug and rewrite {self.context.test_filename}") - code_doc = await FileRepository.get_file( - filename=self.context.code_filename, relative_path=CONFIG.src_workspace + logger.info(f"Debug and rewrite {self.i_context.test_filename}") + code_doc = await self.repo.with_src_path(self.context.src_workspace).srcs.get( + filename=self.i_context.code_filename ) if not code_doc: return "" - test_doc = await FileRepository.get_file( - filename=self.context.test_filename, relative_path=TEST_CODES_FILE_REPO - ) + test_doc = await self.repo.tests.get(filename=self.i_context.test_filename) if not test_doc: return "" prompt = PROMPT_TEMPLATE.format(code=code_doc.content, test_code=test_doc.content, logs=output_detail.stderr) diff --git a/metagpt/actions/design_api.py b/metagpt/actions/design_api.py index 2574550e4..cb6013538 100644 --- a/metagpt/actions/design_api.py +++ b/metagpt/actions/design_api.py @@ -14,18 +14,17 @@ from pathlib import Path from typing import Optional from metagpt.actions import Action, ActionOutput -from metagpt.actions.design_api_an import DESIGN_API_NODE -from metagpt.config import CONFIG -from metagpt.const import ( - DATA_API_DESIGN_FILE_REPO, - PRDS_FILE_REPO, - SEQ_FLOW_FILE_REPO, - SYSTEM_DESIGN_FILE_REPO, - SYSTEM_DESIGN_PDF_FILE_REPO, +from metagpt.actions.design_api_an import ( + DATA_STRUCTURES_AND_INTERFACES, + DESIGN_API_NODE, + PROGRAM_CALL_FLOW, + REFINED_DATA_STRUCTURES_AND_INTERFACES, + REFINED_DESIGN_NODE, + REFINED_PROGRAM_CALL_FLOW, ) +from metagpt.const import DATA_API_DESIGN_FILE_REPO, SEQ_FLOW_FILE_REPO from metagpt.logs import logger from metagpt.schema import Document, Documents, Message -from metagpt.utils.file_repository import FileRepository from metagpt.utils.mermaid import mermaid_to_file NEW_REQ_TEMPLATE = """ @@ -39,36 +38,30 @@ NEW_REQ_TEMPLATE = """ class WriteDesign(Action): name: str = "" - context: Optional[str] = None + i_context: Optional[str] = None desc: str = ( "Based on the PRD, think about the system design, and design the corresponding APIs, " "data structures, library tables, processes, and paths. Please provide your design, feedback " "clearly and in detail." ) - async def run(self, with_messages: Message, schema: str = CONFIG.prompt_schema): - # Use `git status` to identify which PRD documents have been modified in the `docs/prds` directory. - prds_file_repo = CONFIG.git_repo.new_file_repository(PRDS_FILE_REPO) - changed_prds = prds_file_repo.changed_files + async def run(self, with_messages: Message, schema: str = None): + # Use `git status` to identify which PRD documents have been modified in the `docs/prd` directory. + changed_prds = self.repo.docs.prd.changed_files # Use `git status` to identify which design documents in the `docs/system_designs` directory have undergone # changes. - system_design_file_repo = CONFIG.git_repo.new_file_repository(SYSTEM_DESIGN_FILE_REPO) - changed_system_designs = system_design_file_repo.changed_files + changed_system_designs = self.repo.docs.system_design.changed_files # For those PRDs and design documents that have undergone changes, regenerate the design content. changed_files = Documents() for filename in changed_prds.keys(): - doc = await self._update_system_design( - filename=filename, prds_file_repo=prds_file_repo, system_design_file_repo=system_design_file_repo - ) + doc = await self._update_system_design(filename=filename) changed_files.docs[filename] = doc for filename in changed_system_designs.keys(): if filename in changed_files.docs: continue - doc = await self._update_system_design( - filename=filename, prds_file_repo=prds_file_repo, system_design_file_repo=system_design_file_repo - ) + doc = await self._update_system_design(filename=filename) changed_files.docs[filename] = doc if not changed_files.docs: logger.info("Nothing has changed.") @@ -76,61 +69,52 @@ class WriteDesign(Action): # leaving room for global optimization in subsequent steps. return ActionOutput(content=changed_files.model_dump_json(), instruct_content=changed_files) - async def _new_system_design(self, context, schema=CONFIG.prompt_schema): - node = await DESIGN_API_NODE.fill(context=context, llm=self.llm, schema=schema) + async def _new_system_design(self, context): + node = await DESIGN_API_NODE.fill(context=context, llm=self.llm) return node - async def _merge(self, prd_doc, system_design_doc, schema=CONFIG.prompt_schema): + async def _merge(self, prd_doc, system_design_doc): context = NEW_REQ_TEMPLATE.format(old_design=system_design_doc.content, context=prd_doc.content) - node = await DESIGN_API_NODE.fill(context=context, llm=self.llm, schema=schema) + node = await REFINED_DESIGN_NODE.fill(context=context, llm=self.llm) system_design_doc.content = node.instruct_content.model_dump_json() return system_design_doc - async def _update_system_design(self, filename, prds_file_repo, system_design_file_repo) -> Document: - prd = await prds_file_repo.get(filename) - old_system_design_doc = await system_design_file_repo.get(filename) + async def _update_system_design(self, filename) -> Document: + prd = await self.repo.docs.prd.get(filename) + old_system_design_doc = await self.repo.docs.system_design.get(filename) if not old_system_design_doc: system_design = await self._new_system_design(context=prd.content) - doc = Document( - root_path=SYSTEM_DESIGN_FILE_REPO, + doc = await self.repo.docs.system_design.save( filename=filename, content=system_design.instruct_content.model_dump_json(), + dependencies={prd.root_relative_path}, ) else: doc = await self._merge(prd_doc=prd, system_design_doc=old_system_design_doc) - await system_design_file_repo.save( - filename=filename, content=doc.content, dependencies={prd.root_relative_path} - ) + await self.repo.docs.system_design.save_doc(doc=doc, dependencies={prd.root_relative_path}) await self._save_data_api_design(doc) await self._save_seq_flow(doc) - await self._save_pdf(doc) + await self.repo.resources.system_design.save_pdf(doc=doc) return doc - @staticmethod - async def _save_data_api_design(design_doc): + async def _save_data_api_design(self, design_doc): m = json.loads(design_doc.content) - data_api_design = m.get("Data structures and interfaces") + data_api_design = m.get(DATA_STRUCTURES_AND_INTERFACES.key) or m.get(REFINED_DATA_STRUCTURES_AND_INTERFACES.key) if not data_api_design: return - pathname = CONFIG.git_repo.workdir / DATA_API_DESIGN_FILE_REPO / Path(design_doc.filename).with_suffix("") - await WriteDesign._save_mermaid_file(data_api_design, pathname) + pathname = self.repo.workdir / DATA_API_DESIGN_FILE_REPO / Path(design_doc.filename).with_suffix("") + await self._save_mermaid_file(data_api_design, pathname) logger.info(f"Save class view to {str(pathname)}") - @staticmethod - async def _save_seq_flow(design_doc): + async def _save_seq_flow(self, design_doc): m = json.loads(design_doc.content) - seq_flow = m.get("Program call flow") + seq_flow = m.get(PROGRAM_CALL_FLOW.key) or m.get(REFINED_PROGRAM_CALL_FLOW.key) if not seq_flow: return - pathname = CONFIG.git_repo.workdir / Path(SEQ_FLOW_FILE_REPO) / Path(design_doc.filename).with_suffix("") - await WriteDesign._save_mermaid_file(seq_flow, pathname) + pathname = self.repo.workdir / Path(SEQ_FLOW_FILE_REPO) / Path(design_doc.filename).with_suffix("") + await self._save_mermaid_file(seq_flow, pathname) logger.info(f"Saving sequence flow to {str(pathname)}") - @staticmethod - async def _save_pdf(design_doc): - await FileRepository.save_as(doc=design_doc, with_suffix=".md", relative_path=SYSTEM_DESIGN_PDF_FILE_REPO) - - @staticmethod - async def _save_mermaid_file(data: str, pathname: Path): + async def _save_mermaid_file(self, data: str, pathname: Path): pathname.parent.mkdir(parents=True, exist_ok=True) - await mermaid_to_file(data, pathname) + await mermaid_to_file(self.config.mermaid_engine, data, pathname) diff --git a/metagpt/actions/design_api_an.py b/metagpt/actions/design_api_an.py index 3737203cf..35b50ef8f 100644 --- a/metagpt/actions/design_api_an.py +++ b/metagpt/actions/design_api_an.py @@ -8,6 +8,7 @@ from typing import List from metagpt.actions.action_node import ActionNode +from metagpt.logs import logger from metagpt.utils.mermaid import MMC1, MMC2 IMPLEMENTATION_APPROACH = ActionNode( @@ -17,6 +18,15 @@ IMPLEMENTATION_APPROACH = ActionNode( example="We will ...", ) +REFINED_IMPLEMENTATION_APPROACH = ActionNode( + key="Refined Implementation Approach", + expected_type=str, + instruction="Update and extend the original implementation approach to reflect the evolving challenges and " + "requirements due to incremental development. Outline the steps involved in the implementation process with the " + "detailed strategies.", + example="We will refine ...", +) + PROJECT_NAME = ActionNode( key="Project name", expected_type=str, instruction="The project name with underline", example="game_2048" ) @@ -28,6 +38,14 @@ FILE_LIST = ActionNode( example=["main.py", "game.py"], ) +REFINED_FILE_LIST = ActionNode( + key="Refined File list", + expected_type=List[str], + instruction="Update and expand the original file list including only relative paths. Up to 2 files can be added." + "Ensure that the refined file list reflects the evolving structure of the project.", + example=["main.py", "game.py", "new_feature.py"], +) + DATA_STRUCTURES_AND_INTERFACES = ActionNode( key="Data structures and interfaces", expected_type=str, @@ -37,6 +55,16 @@ DATA_STRUCTURES_AND_INTERFACES = ActionNode( example=MMC1, ) +REFINED_DATA_STRUCTURES_AND_INTERFACES = ActionNode( + key="Refined Data structures and interfaces", + expected_type=str, + instruction="Update and extend the existing mermaid classDiagram code syntax to incorporate new classes, " + "methods (including __init__), and functions with precise type annotations. Delineate additional " + "relationships between classes, ensuring clarity and adherence to PEP8 standards." + "Retain content that is not related to incremental development but important for consistency and clarity.", + example=MMC1, +) + PROGRAM_CALL_FLOW = ActionNode( key="Program call flow", expected_type=str, @@ -45,6 +73,16 @@ PROGRAM_CALL_FLOW = ActionNode( example=MMC2, ) +REFINED_PROGRAM_CALL_FLOW = ActionNode( + key="Refined Program call flow", + expected_type=str, + instruction="Extend the existing sequenceDiagram code syntax with detailed information, accurately covering the" + "CRUD and initialization of each object. Ensure correct syntax usage and reflect the incremental changes introduced" + "in the classes and API defined above. " + "Retain content that is not related to incremental development but important for consistency and clarity.", + example=MMC2, +) + ANYTHING_UNCLEAR = ActionNode( key="Anything UNCLEAR", expected_type=str, @@ -61,4 +99,24 @@ NODES = [ ANYTHING_UNCLEAR, ] +REFINED_NODES = [ + REFINED_IMPLEMENTATION_APPROACH, + REFINED_FILE_LIST, + REFINED_DATA_STRUCTURES_AND_INTERFACES, + REFINED_PROGRAM_CALL_FLOW, + ANYTHING_UNCLEAR, +] + DESIGN_API_NODE = ActionNode.from_children("DesignAPI", NODES) +REFINED_DESIGN_NODE = ActionNode.from_children("RefinedDesignAPI", REFINED_NODES) + + +def main(): + prompt = DESIGN_API_NODE.compile(context="") + logger.info(prompt) + prompt = REFINED_DESIGN_NODE.compile(context="") + logger.info(prompt) + + +if __name__ == "__main__": + main() diff --git a/metagpt/actions/design_api_review.py b/metagpt/actions/design_api_review.py index fb1b92d85..ccd01a4c3 100644 --- a/metagpt/actions/design_api_review.py +++ b/metagpt/actions/design_api_review.py @@ -13,7 +13,7 @@ from metagpt.actions.action import Action class DesignReview(Action): name: str = "DesignReview" - context: Optional[str] = None + i_context: Optional[str] = None async def run(self, prd, api_design): prompt = ( diff --git a/metagpt/actions/execute_code.py b/metagpt/actions/execute_code.py index 6591f479f..6a4a9abb8 100644 --- a/metagpt/actions/execute_code.py +++ b/metagpt/actions/execute_code.py @@ -123,7 +123,10 @@ class ExecutePyCode(ExecuteCode, Action): return parsed_output for i, output in enumerate(outputs): - if output["output_type"] == "stream": + if output["output_type"] == "stream" and not any( + tag in output["text"] + for tag in ["| INFO | metagpt", "| ERROR | metagpt", "| WARNING | metagpt"] + ): parsed_output += output["text"] elif output["output_type"] == "display_data": if "image/png" in output["data"]: diff --git a/metagpt/actions/execute_task.py b/metagpt/actions/execute_task.py index 4ae4ee17b..1cc3bd699 100644 --- a/metagpt/actions/execute_task.py +++ b/metagpt/actions/execute_task.py @@ -13,7 +13,7 @@ from metagpt.schema import Message class ExecuteTask(Action): name: str = "ExecuteTask" - context: list[Message] = [] + i_context: list[Message] = [] async def run(self, *args, **kwargs): pass diff --git a/metagpt/actions/generate_questions.py b/metagpt/actions/generate_questions.py index 8573708f2..c96a37649 100644 --- a/metagpt/actions/generate_questions.py +++ b/metagpt/actions/generate_questions.py @@ -1,8 +1,6 @@ #!/usr/bin/env python # -*- coding: utf-8 -*- """ -@Time : 2023/9/12 17:45 -@Author : fisherdeng @File : generate_questions.py """ from metagpt.actions import Action @@ -23,5 +21,5 @@ class GenerateQuestions(Action): name: str = "GenerateQuestions" - async def run(self, context): + async def run(self, context) -> ActionNode: return await QUESTIONS.fill(context=context, llm=self.llm) diff --git a/metagpt/actions/invoice_ocr.py b/metagpt/actions/invoice_ocr.py index 36570097a..7cf71a8ff 100644 --- a/metagpt/actions/invoice_ocr.py +++ b/metagpt/actions/invoice_ocr.py @@ -16,17 +16,14 @@ from typing import Optional import pandas as pd from paddleocr import PaddleOCR -from pydantic import Field from metagpt.actions import Action from metagpt.const import INVOICE_OCR_TABLE_PATH -from metagpt.llm import LLM from metagpt.logs import logger from metagpt.prompts.invoice_ocr import ( EXTRACT_OCR_MAIN_INFO_PROMPT, REPLY_OCR_QUESTION_PROMPT, ) -from metagpt.provider.base_llm import BaseLLM from metagpt.utils.common import OutputParser from metagpt.utils.file import File @@ -41,7 +38,7 @@ class InvoiceOCR(Action): """ name: str = "InvoiceOCR" - context: Optional[str] = None + i_context: Optional[str] = None @staticmethod async def _check_file_type(file_path: Path) -> str: @@ -132,8 +129,7 @@ class GenerateTable(Action): """ name: str = "GenerateTable" - context: Optional[str] = None - llm: BaseLLM = Field(default_factory=LLM) + i_context: Optional[str] = None language: str = "ch" async def run(self, ocr_results: list, filename: str, *args, **kwargs) -> dict[str, str]: @@ -176,9 +172,6 @@ class ReplyQuestion(Action): """ - name: str = "ReplyQuestion" - context: Optional[str] = None - llm: BaseLLM = Field(default_factory=LLM) language: str = "ch" async def run(self, query: str, ocr_result: list, *args, **kwargs) -> str: diff --git a/metagpt/actions/ml_da_action.py b/metagpt/actions/ml_action.py similarity index 52% rename from metagpt/actions/ml_da_action.py rename to metagpt/actions/ml_action.py index d4e77773f..a61233e5a 100644 --- a/metagpt/actions/ml_da_action.py +++ b/metagpt/actions/ml_action.py @@ -1,28 +1,64 @@ import json +from typing import List, Tuple from metagpt.actions import Action -from metagpt.prompts.ml_engineer import PRINT_DATA_COLUMNS, UPDATE_DATA_COLUMNS -from metagpt.schema import Plan +from metagpt.actions.write_analysis_code import WriteCodeWithTools +from metagpt.prompts.ml_action import ( + GENERATE_CODE_PROMPT, + ML_TOOL_USAGE_PROMPT, + PRINT_DATA_COLUMNS, + UPDATE_DATA_COLUMNS, +) +from metagpt.prompts.write_analysis_code import CODE_GENERATOR_WITH_TOOLS +from metagpt.schema import Message, Plan from metagpt.utils.common import CodeParser, create_func_config, remove_comments -class SummarizeAnalysis(Action): - PROMPT_TEMPLATE: str = """ - # Context - {context} - # Summary - Output a 30-word summary on analysis tool and modeling algorithms you have used, and the corresponding result. Make sure to announce the complete path to your test prediction file. Your summary: - """ +class WriteCodeWithToolsML(WriteCodeWithTools): + async def run( + self, + context: List[Message], + plan: Plan = None, + column_info: str = "", + **kwargs, + ) -> Tuple[List[Message], str]: + # prepare tool schemas and tool-type-specific instruction + tool_schemas, tool_type_usage_prompt = await self._prepare_tools(plan=plan) - async def run(self, conmpleted_plan: Plan) -> str: - tasks = json.dumps( - [task.dict() for task in conmpleted_plan.tasks], - indent=4, - ensure_ascii=False, - ) # all tasks finished, return all task outputs - prompt = self.PROMPT_TEMPLATE.format(context=tasks) - summary = await self._aask(prompt) - return summary + # ML-specific variables to be used in prompt + code_steps = plan.current_task.code_steps + finished_tasks = plan.get_finished_tasks() + code_context = [remove_comments(task.code) for task in finished_tasks] + code_context = "\n\n".join(code_context) + + # prepare prompt depending on tool availability & LLM call + if tool_schemas: + prompt = ML_TOOL_USAGE_PROMPT.format( + user_requirement=plan.goal, + history_code=code_context, + current_task=plan.current_task.instruction, + column_info=column_info, + tool_type_usage_prompt=tool_type_usage_prompt, + code_steps=code_steps, + tool_schemas=tool_schemas, + ) + + else: + prompt = GENERATE_CODE_PROMPT.format( + user_requirement=plan.goal, + history_code=code_context, + current_task=plan.current_task.instruction, + column_info=column_info, + tool_type_usage_prompt=tool_type_usage_prompt, + code_steps=code_steps, + ) + tool_config = create_func_config(CODE_GENERATOR_WITH_TOOLS) + rsp = await self.llm.aask_code(prompt, **tool_config) + + # Extra output to be used for potential debugging + context = [Message(content=prompt, role="user")] + + return context, rsp class Reflect(Action): diff --git a/metagpt/actions/prepare_documents.py b/metagpt/actions/prepare_documents.py index 5c5798d95..ab069dc11 100644 --- a/metagpt/actions/prepare_documents.py +++ b/metagpt/actions/prepare_documents.py @@ -12,39 +12,41 @@ from pathlib import Path from typing import Optional from metagpt.actions import Action, ActionOutput -from metagpt.config import CONFIG -from metagpt.const import DOCS_FILE_REPO, REQUIREMENT_FILENAME -from metagpt.schema import Document +from metagpt.const import REQUIREMENT_FILENAME from metagpt.utils.file_repository import FileRepository from metagpt.utils.git_repository import GitRepository +from metagpt.utils.project_repo import ProjectRepo class PrepareDocuments(Action): """PrepareDocuments Action: initialize project folder and add new requirements to docs/requirements.txt.""" name: str = "PrepareDocuments" - context: Optional[str] = None + i_context: Optional[str] = None + + @property + def config(self): + return self.context.config def _init_repo(self): """Initialize the Git environment.""" - if not CONFIG.project_path: - name = CONFIG.project_name or FileRepository.new_filename() - path = Path(CONFIG.workspace_path) / name + if not self.config.project_path: + name = self.config.project_name or FileRepository.new_filename() + path = Path(self.config.workspace.path) / name else: - path = Path(CONFIG.project_path) - if path.exists() and not CONFIG.inc: + path = Path(self.config.project_path) + if path.exists() and not self.config.inc: shutil.rmtree(path) - CONFIG.project_path = path - CONFIG.git_repo = GitRepository(local_path=path, auto_init=True) + self.config.project_path = path + self.context.git_repo = GitRepository(local_path=path, auto_init=True) + self.context.repo = ProjectRepo(self.context.git_repo) async def run(self, with_messages, **kwargs): """Create and initialize the workspace folder, initialize the Git environment.""" self._init_repo() # Write the newly added requirements from the main parameter idea to `docs/requirement.txt`. - doc = Document(root_path=DOCS_FILE_REPO, filename=REQUIREMENT_FILENAME, content=with_messages[0].content) - await FileRepository.save_file(filename=REQUIREMENT_FILENAME, content=doc.content, relative_path=DOCS_FILE_REPO) - + doc = await self.repo.docs.save(filename=REQUIREMENT_FILENAME, content=with_messages[0].content) # Send a Message notification to the WritePRD action, instructing it to process requirements using - # `docs/requirement.txt` and `docs/prds/`. + # `docs/requirement.txt` and `docs/prd/`. return ActionOutput(content=doc.content, instruct_content=doc) diff --git a/metagpt/actions/project_management.py b/metagpt/actions/project_management.py index e40c2034b..67a614d6f 100644 --- a/metagpt/actions/project_management.py +++ b/metagpt/actions/project_management.py @@ -13,23 +13,16 @@ import json from typing import Optional -from metagpt.actions import ActionOutput from metagpt.actions.action import Action -from metagpt.actions.project_management_an import PM_NODE -from metagpt.config import CONFIG -from metagpt.const import ( - PACKAGE_REQUIREMENTS_FILENAME, - SYSTEM_DESIGN_FILE_REPO, - TASK_FILE_REPO, - TASK_PDF_FILE_REPO, -) +from metagpt.actions.action_output import ActionOutput +from metagpt.actions.project_management_an import PM_NODE, REFINED_PM_NODE +from metagpt.const import PACKAGE_REQUIREMENTS_FILENAME from metagpt.logs import logger from metagpt.schema import Document, Documents -from metagpt.utils.file_repository import FileRepository NEW_REQ_TEMPLATE = """ ### Legacy Content -{old_tasks} +{old_task} ### New Requirements {context} @@ -38,30 +31,23 @@ NEW_REQ_TEMPLATE = """ class WriteTasks(Action): name: str = "CreateTasks" - context: Optional[str] = None + i_context: Optional[str] = None - async def run(self, with_messages, schema=CONFIG.prompt_schema): - system_design_file_repo = CONFIG.git_repo.new_file_repository(SYSTEM_DESIGN_FILE_REPO) - changed_system_designs = system_design_file_repo.changed_files - - tasks_file_repo = CONFIG.git_repo.new_file_repository(TASK_FILE_REPO) - changed_tasks = tasks_file_repo.changed_files + async def run(self, with_messages): + changed_system_designs = self.repo.docs.system_design.changed_files + changed_tasks = self.repo.docs.task.changed_files change_files = Documents() # Rewrite the system designs that have undergone changes based on the git head diff under # `docs/system_designs/`. for filename in changed_system_designs: - task_doc = await self._update_tasks( - filename=filename, system_design_file_repo=system_design_file_repo, tasks_file_repo=tasks_file_repo - ) + task_doc = await self._update_tasks(filename=filename) change_files.docs[filename] = task_doc # Rewrite the task files that have undergone changes based on the git head diff under `docs/tasks/`. for filename in changed_tasks: if filename in change_files.docs: continue - task_doc = await self._update_tasks( - filename=filename, system_design_file_repo=system_design_file_repo, tasks_file_repo=tasks_file_repo - ) + task_doc = await self._update_tasks(filename=filename) change_files.docs[filename] = task_doc if not change_files.docs: @@ -70,39 +56,36 @@ class WriteTasks(Action): # global optimization in subsequent steps. return ActionOutput(content=change_files.model_dump_json(), instruct_content=change_files) - async def _update_tasks(self, filename, system_design_file_repo, tasks_file_repo): - system_design_doc = await system_design_file_repo.get(filename) - task_doc = await tasks_file_repo.get(filename) + async def _update_tasks(self, filename): + system_design_doc = await self.repo.docs.system_design.get(filename) + task_doc = await self.repo.docs.task.get(filename) if task_doc: task_doc = await self._merge(system_design_doc=system_design_doc, task_doc=task_doc) + await self.repo.docs.task.save_doc(doc=task_doc, dependencies={system_design_doc.root_relative_path}) else: rsp = await self._run_new_tasks(context=system_design_doc.content) - task_doc = Document( - root_path=TASK_FILE_REPO, filename=filename, content=rsp.instruct_content.model_dump_json() + task_doc = await self.repo.docs.task.save( + filename=filename, + content=rsp.instruct_content.model_dump_json(), + dependencies={system_design_doc.root_relative_path}, ) - await tasks_file_repo.save( - filename=filename, content=task_doc.content, dependencies={system_design_doc.root_relative_path} - ) await self._update_requirements(task_doc) - await self._save_pdf(task_doc=task_doc) return task_doc - async def _run_new_tasks(self, context, schema=CONFIG.prompt_schema): - node = await PM_NODE.fill(context, self.llm, schema) + async def _run_new_tasks(self, context): + node = await PM_NODE.fill(context, self.llm, schema=self.prompt_schema) return node - async def _merge(self, system_design_doc, task_doc, schema=CONFIG.prompt_schema) -> Document: - context = NEW_REQ_TEMPLATE.format(context=system_design_doc.content, old_tasks=task_doc.content) - node = await PM_NODE.fill(context, self.llm, schema) + async def _merge(self, system_design_doc, task_doc) -> Document: + context = NEW_REQ_TEMPLATE.format(context=system_design_doc.content, old_task=task_doc.content) + node = await REFINED_PM_NODE.fill(context, self.llm, schema=self.prompt_schema) task_doc.content = node.instruct_content.model_dump_json() return task_doc - @staticmethod - async def _update_requirements(doc): + async def _update_requirements(self, doc): m = json.loads(doc.content) - packages = set(m.get("Required Python third-party packages", set())) - file_repo = CONFIG.git_repo.new_file_repository() - requirement_doc = await file_repo.get(filename=PACKAGE_REQUIREMENTS_FILENAME) + packages = set(m.get("Required Python packages", set())) + requirement_doc = await self.repo.get(filename=PACKAGE_REQUIREMENTS_FILENAME) if not requirement_doc: requirement_doc = Document(filename=PACKAGE_REQUIREMENTS_FILENAME, root_path=".", content="") lines = requirement_doc.content.splitlines() @@ -110,8 +93,4 @@ class WriteTasks(Action): if pkg == "": continue packages.add(pkg) - await file_repo.save(PACKAGE_REQUIREMENTS_FILENAME, content="\n".join(packages)) - - @staticmethod - async def _save_pdf(task_doc): - await FileRepository.save_as(doc=task_doc, with_suffix=".md", relative_path=TASK_PDF_FILE_REPO) + await self.repo.save(filename=PACKAGE_REQUIREMENTS_FILENAME, content="\n".join(packages)) diff --git a/metagpt/actions/project_management_an.py b/metagpt/actions/project_management_an.py index 215a67202..379a23384 100644 --- a/metagpt/actions/project_management_an.py +++ b/metagpt/actions/project_management_an.py @@ -35,6 +35,20 @@ LOGIC_ANALYSIS = ActionNode( ], ) +REFINED_LOGIC_ANALYSIS = ActionNode( + key="Refined Logic Analysis", + expected_type=List[List[str]], + instruction="Review and refine the logic analysis by merging the Legacy Content and Incremental Content. " + "Provide a comprehensive list of files with classes/methods/functions to be implemented or modified incrementally. " + "Include dependency analysis, consider potential impacts on existing code, and document necessary imports.", + example=[ + ["game.py", "Contains Game class and ... functions"], + ["main.py", "Contains main function, from game import Game"], + ["new_feature.py", "Introduces NewFeature class and related functions"], + ["utils.py", "Modifies existing utility functions to support incremental changes"], + ], +) + TASK_LIST = ActionNode( key="Task list", expected_type=List[str], @@ -42,6 +56,15 @@ TASK_LIST = ActionNode( example=["game.py", "main.py"], ) +REFINED_TASK_LIST = ActionNode( + key="Refined Task list", + expected_type=List[str], + instruction="Review and refine the combined task list after the merger of Legacy Content and Incremental Content, " + "and consistent with Refined File List. Ensure that tasks are organized in a logical and prioritized order, " + "considering dependencies for a streamlined and efficient development process. ", + example=["new_feature.py", "utils", "game.py", "main.py"], +) + FULL_API_SPEC = ActionNode( key="Full API spec", expected_type=str, @@ -54,9 +77,19 @@ SHARED_KNOWLEDGE = ActionNode( key="Shared Knowledge", expected_type=str, instruction="Detail any shared knowledge, like common utility functions or configuration variables.", - example="'game.py' contains functions shared across the project.", + example="`game.py` contains functions shared across the project.", ) +REFINED_SHARED_KNOWLEDGE = ActionNode( + key="Refined Shared Knowledge", + expected_type=str, + instruction="Update and expand shared knowledge to reflect any new elements introduced. This includes common " + "utility functions, configuration variables for team collaboration. Retain content that is not related to " + "incremental development but important for consistency and clarity.", + example="`new_module.py` enhances shared utility functions for improved code reusability and collaboration.", +) + + ANYTHING_UNCLEAR_PM = ActionNode( key="Anything UNCLEAR", expected_type=str, @@ -74,13 +107,25 @@ NODES = [ ANYTHING_UNCLEAR_PM, ] +REFINED_NODES = [ + REQUIRED_PYTHON_PACKAGES, + REQUIRED_OTHER_LANGUAGE_PACKAGES, + REFINED_LOGIC_ANALYSIS, + REFINED_TASK_LIST, + FULL_API_SPEC, + REFINED_SHARED_KNOWLEDGE, + ANYTHING_UNCLEAR_PM, +] PM_NODE = ActionNode.from_children("PM_NODE", NODES) +REFINED_PM_NODE = ActionNode.from_children("REFINED_PM_NODE", REFINED_NODES) def main(): prompt = PM_NODE.compile(context="") logger.info(prompt) + prompt = REFINED_PM_NODE.compile(context="") + logger.info(prompt) if __name__ == "__main__": diff --git a/metagpt/actions/rebuild_class_view.py b/metagpt/actions/rebuild_class_view.py index 5128b9fee..2140ad874 100644 --- a/metagpt/actions/rebuild_class_view.py +++ b/metagpt/actions/rebuild_class_view.py @@ -12,7 +12,7 @@ from pathlib import Path import aiofiles from metagpt.actions import Action -from metagpt.config import CONFIG +from metagpt.config2 import config from metagpt.const import ( AGGREGATION, COMPOSITION, @@ -29,16 +29,16 @@ from metagpt.utils.graph_repository import GraphKeyword, GraphRepository class RebuildClassView(Action): - async def run(self, with_messages=None, format=CONFIG.prompt_schema): - graph_repo_pathname = CONFIG.git_repo.workdir / GRAPH_REPO_FILE_REPO / CONFIG.git_repo.workdir.name + async def run(self, with_messages=None, format=config.prompt_schema): + graph_repo_pathname = self.context.git_repo.workdir / GRAPH_REPO_FILE_REPO / self.context.git_repo.workdir.name graph_db = await DiGraphRepository.load_from(str(graph_repo_pathname.with_suffix(".json"))) - repo_parser = RepoParser(base_directory=Path(self.context)) + repo_parser = RepoParser(base_directory=Path(self.i_context)) # use pylint - class_views, relationship_views, package_root = await repo_parser.rebuild_class_views(path=Path(self.context)) + class_views, relationship_views, package_root = await repo_parser.rebuild_class_views(path=Path(self.i_context)) await GraphRepository.update_graph_db_with_class_views(graph_db, class_views) await GraphRepository.update_graph_db_with_class_relationship_views(graph_db, relationship_views) # use ast - direction, diff_path = self._diff_path(path_root=Path(self.context).resolve(), package_root=package_root) + direction, diff_path = self._diff_path(path_root=Path(self.i_context).resolve(), package_root=package_root) symbols = repo_parser.generate_symbols() for file_info in symbols: # Align to the same root directory in accordance with `class_views`. @@ -48,9 +48,9 @@ class RebuildClassView(Action): await graph_db.save() async def _create_mermaid_class_views(self, graph_db): - path = Path(CONFIG.git_repo.workdir) / DATA_API_DESIGN_FILE_REPO + path = Path(self.context.git_repo.workdir) / DATA_API_DESIGN_FILE_REPO path.mkdir(parents=True, exist_ok=True) - pathname = path / CONFIG.git_repo.workdir.name + pathname = path / self.context.git_repo.workdir.name async with aiofiles.open(str(pathname.with_suffix(".mmd")), mode="w", encoding="utf-8") as writer: content = "classDiagram\n" logger.debug(content) diff --git a/metagpt/actions/rebuild_sequence_view.py b/metagpt/actions/rebuild_sequence_view.py index 865050c93..777dde8ce 100644 --- a/metagpt/actions/rebuild_sequence_view.py +++ b/metagpt/actions/rebuild_sequence_view.py @@ -12,7 +12,7 @@ from pathlib import Path from typing import List from metagpt.actions import Action -from metagpt.config import CONFIG +from metagpt.config2 import config from metagpt.const import GRAPH_REPO_FILE_REPO from metagpt.logs import logger from metagpt.utils.common import aread, list_files @@ -21,8 +21,8 @@ from metagpt.utils.graph_repository import GraphKeyword class RebuildSequenceView(Action): - async def run(self, with_messages=None, format=CONFIG.prompt_schema): - graph_repo_pathname = CONFIG.git_repo.workdir / GRAPH_REPO_FILE_REPO / CONFIG.git_repo.workdir.name + async def run(self, with_messages=None, format=config.prompt_schema): + graph_repo_pathname = self.context.git_repo.workdir / GRAPH_REPO_FILE_REPO / self.context.git_repo.workdir.name graph_db = await DiGraphRepository.load_from(str(graph_repo_pathname.with_suffix(".json"))) entries = await RebuildSequenceView._search_main_entry(graph_db) for entry in entries: @@ -41,7 +41,9 @@ class RebuildSequenceView(Action): async def _rebuild_sequence_view(self, entry, graph_db): filename = entry.subject.split(":", 1)[0] - src_filename = RebuildSequenceView._get_full_filename(root=self.context, pathname=filename) + src_filename = RebuildSequenceView._get_full_filename(root=self.i_context, pathname=filename) + if not src_filename: + return content = await aread(filename=src_filename, encoding="utf-8") content = f"```python\n{content}\n```\n\n---\nTranslate the code above into Mermaid Sequence Diagram." data = await self.llm.aask( diff --git a/metagpt/actions/research.py b/metagpt/actions/research.py index 90b08cb6a..2755628c9 100644 --- a/metagpt/actions/research.py +++ b/metagpt/actions/research.py @@ -8,10 +8,8 @@ from typing import Callable, Optional, Union from pydantic import Field, parse_obj_as from metagpt.actions import Action -from metagpt.config import CONFIG -from metagpt.llm import LLM +from metagpt.config2 import config from metagpt.logs import logger -from metagpt.provider.base_llm import BaseLLM from metagpt.tools.search_engine import SearchEngine from metagpt.tools.web_browser_engine import WebBrowserEngine, WebBrowserEngineType from metagpt.utils.common import OutputParser @@ -81,7 +79,7 @@ class CollectLinks(Action): """Action class to collect links from a search engine.""" name: str = "CollectLinks" - context: Optional[str] = None + i_context: Optional[str] = None desc: str = "Collect links from a search engine." search_engine: SearchEngine = Field(default_factory=SearchEngine) @@ -129,8 +127,8 @@ class CollectLinks(Action): if len(remove) == 0: break - model_name = CONFIG.get_model_name(CONFIG.get_default_llm_provider_enum()) - prompt = reduce_message_length(gen_msg(), model_name, system_text, CONFIG.max_tokens_rsp) + model_name = config.get_openai_llm().model + prompt = reduce_message_length(gen_msg(), model_name, system_text, 4096) logger.debug(prompt) queries = await self._aask(prompt, [system_text]) try: @@ -177,19 +175,16 @@ class WebBrowseAndSummarize(Action): """Action class to explore the web and provide summaries of articles and webpages.""" name: str = "WebBrowseAndSummarize" - context: Optional[str] = None - llm: BaseLLM = Field(default_factory=LLM) + i_context: Optional[str] = None desc: str = "Explore the web and provide summaries of articles and webpages." browse_func: Union[Callable[[list[str]], None], None] = None - web_browser_engine: Optional[WebBrowserEngine] = None + web_browser_engine: Optional[WebBrowserEngine] = WebBrowserEngineType.PLAYWRIGHT def __init__(self, **kwargs): super().__init__(**kwargs) - if CONFIG.model_for_researcher_summary: - self.llm.model = CONFIG.model_for_researcher_summary self.web_browser_engine = WebBrowserEngine( - engine=WebBrowserEngineType.CUSTOM if self.browse_func else None, + engine=WebBrowserEngineType.CUSTOM if self.browse_func else WebBrowserEngineType.PLAYWRIGHT, run_func=self.browse_func, ) @@ -220,9 +215,7 @@ class WebBrowseAndSummarize(Action): for u, content in zip([url, *urls], contents): content = content.inner_text chunk_summaries = [] - for prompt in generate_prompt_chunk( - content, prompt_template, self.llm.model, system_text, CONFIG.max_tokens_rsp - ): + for prompt in generate_prompt_chunk(content, prompt_template, self.llm.model, system_text, 4096): logger.debug(prompt) summary = await self._aask(prompt, [system_text]) if summary == "Not relevant.": @@ -247,14 +240,8 @@ class WebBrowseAndSummarize(Action): class ConductResearch(Action): """Action class to conduct research and generate a research report.""" - name: str = "ConductResearch" - context: Optional[str] = None - llm: BaseLLM = Field(default_factory=LLM) - def __init__(self, **kwargs): super().__init__(**kwargs) - if CONFIG.model_for_researcher_report: - self.llm.model = CONFIG.model_for_researcher_report async def run( self, diff --git a/metagpt/actions/run_code.py b/metagpt/actions/run_code.py index 30b06f1a6..3b84cc9f2 100644 --- a/metagpt/actions/run_code.py +++ b/metagpt/actions/run_code.py @@ -16,12 +16,12 @@ class. """ import subprocess +from pathlib import Path from typing import Tuple from pydantic import Field from metagpt.actions.action import Action -from metagpt.config import CONFIG from metagpt.logs import logger from metagpt.schema import RunCodeContext, RunCodeResult from metagpt.utils.exceptions import handle_exception @@ -48,7 +48,7 @@ WRITE ONLY ONE WORD, Engineer OR QaEngineer OR NoOne, IN THIS SECTION. You should fill in necessary instruction, status, send to, and finally return all content between the --- segment line. """ -CONTEXT = """ +TEMPLATE_CONTEXT = """ ## Development Code File Name {code_file_name} ## Development Code @@ -77,7 +77,7 @@ standard errors: class RunCode(Action): name: str = "RunCode" - context: RunCodeContext = Field(default_factory=RunCodeContext) + i_context: RunCodeContext = Field(default_factory=RunCodeContext) @classmethod async def run_text(cls, code) -> Tuple[str, str]: @@ -89,13 +89,12 @@ class RunCode(Action): return "", str(e) return namespace.get("result", ""), "" - @classmethod - async def run_script(cls, working_directory, additional_python_paths=[], command=[]) -> Tuple[str, str]: + async def run_script(self, working_directory, additional_python_paths=[], command=[]) -> Tuple[str, str]: working_directory = str(working_directory) additional_python_paths = [str(path) for path in additional_python_paths] # Copy the current environment variables - env = CONFIG.new_environ() + env = self.context.new_environ() # Modify the PYTHONPATH environment variable additional_python_paths = [working_directory] + additional_python_paths @@ -119,25 +118,25 @@ class RunCode(Action): return stdout.decode("utf-8"), stderr.decode("utf-8") async def run(self, *args, **kwargs) -> RunCodeResult: - logger.info(f"Running {' '.join(self.context.command)}") - if self.context.mode == "script": + logger.info(f"Running {' '.join(self.i_context.command)}") + if self.i_context.mode == "script": outs, errs = await self.run_script( - command=self.context.command, - working_directory=self.context.working_directory, - additional_python_paths=self.context.additional_python_paths, + command=self.i_context.command, + working_directory=self.i_context.working_directory, + additional_python_paths=self.i_context.additional_python_paths, ) - elif self.context.mode == "text": - outs, errs = await self.run_text(code=self.context.code) + elif self.i_context.mode == "text": + outs, errs = await self.run_text(code=self.i_context.code) logger.info(f"{outs=}") logger.info(f"{errs=}") - context = CONTEXT.format( - code=self.context.code, - code_file_name=self.context.code_filename, - test_code=self.context.test_code, - test_file_name=self.context.test_filename, - command=" ".join(self.context.command), + context = TEMPLATE_CONTEXT.format( + code=self.i_context.code, + code_file_name=self.i_context.code_filename, + test_code=self.i_context.test_code, + test_file_name=self.i_context.test_filename, + command=" ".join(self.i_context.command), outs=outs[:500], # outs might be long but they are not important, truncate them to avoid token overflow errs=errs[:10000], # truncate errors to avoid token overflow ) @@ -152,11 +151,23 @@ class RunCode(Action): return subprocess.run(cmd, check=check, cwd=cwd, env=env) @staticmethod - def _install_dependencies(working_directory, env): + def _install_requirements(working_directory, env): + file_path = Path(working_directory) / "requirements.txt" + if not file_path.exists(): + return + if file_path.stat().st_size == 0: + return install_command = ["python", "-m", "pip", "install", "-r", "requirements.txt"] logger.info(" ".join(install_command)) RunCode._install_via_subprocess(install_command, check=True, cwd=working_directory, env=env) + @staticmethod + def _install_pytest(working_directory, env): install_pytest_command = ["python", "-m", "pip", "install", "pytest"] logger.info(" ".join(install_pytest_command)) RunCode._install_via_subprocess(install_pytest_command, check=True, cwd=working_directory, env=env) + + @staticmethod + def _install_dependencies(working_directory, env): + RunCode._install_requirements(working_directory, env) + RunCode._install_pytest(working_directory, env) diff --git a/metagpt/actions/search_and_summarize.py b/metagpt/actions/search_and_summarize.py index d2e361f73..59b35cd58 100644 --- a/metagpt/actions/search_and_summarize.py +++ b/metagpt/actions/search_and_summarize.py @@ -8,10 +8,9 @@ from typing import Any, Optional import pydantic -from pydantic import Field, model_validator +from pydantic import model_validator from metagpt.actions import Action -from metagpt.config import CONFIG, Config from metagpt.logs import logger from metagpt.schema import Message from metagpt.tools import SearchEngineType @@ -103,32 +102,25 @@ You are a member of a professional butler team and will provide helpful suggesti """ -# TOTEST class SearchAndSummarize(Action): name: str = "" content: Optional[str] = None - config: None = Field(default_factory=Config) - engine: Optional[SearchEngineType] = CONFIG.search_engine + engine: Optional[SearchEngineType] = None search_func: Optional[Any] = None search_engine: SearchEngine = None result: str = "" - @model_validator(mode="before") - @classmethod - def validate_engine_and_run_func(cls, values): - engine = values.get("engine") - search_func = values.get("search_func") - config = Config() - - if engine is None: - engine = config.search_engine + @model_validator(mode="after") + def validate_engine_and_run_func(self): + if self.engine is None: + self.engine = self.config.search_engine try: - search_engine = SearchEngine(engine=engine, run_func=search_func) + search_engine = SearchEngine(engine=self.engine, run_func=self.search_func) except pydantic.ValidationError: search_engine = None - values["search_engine"] = search_engine - return values + self.search_engine = search_engine + return self async def run(self, context: list[Message], system_text=SEARCH_AND_SUMMARIZE_SYSTEM) -> str: if self.search_engine is None: diff --git a/metagpt/actions/skill_action.py b/metagpt/actions/skill_action.py index 301cebaab..b68596809 100644 --- a/metagpt/actions/skill_action.py +++ b/metagpt/actions/skill_action.py @@ -29,9 +29,7 @@ class ArgumentsParingAction(Action): @property def prompt(self): - prompt = "You are a function parser. You can convert spoken words into function parameters.\n" - prompt += "\n---\n" - prompt += f"{self.skill.name} function parameters description:\n" + prompt = f"{self.skill.name} function parameters description:\n" for k, v in self.skill.arguments.items(): prompt += f"parameter `{k}`: {v}\n" prompt += "\n---\n" @@ -49,7 +47,10 @@ class ArgumentsParingAction(Action): async def run(self, with_message=None, **kwargs) -> Message: prompt = self.prompt - rsp = await self.llm.aask(msg=prompt, system_msgs=[]) + rsp = await self.llm.aask( + msg=prompt, + system_msgs=["You are a function parser.", "You can convert spoken words into function parameters."], + ) logger.debug(f"SKILL:{prompt}\n, RESULT:{rsp}") self.args = ArgumentsParingAction.parse_arguments(skill_name=self.skill.name, txt=rsp) self.rsp = Message(content=rsp, role="assistant", instruct_content=self.args, cause_by=self) diff --git a/metagpt/actions/summarize_code.py b/metagpt/actions/summarize_code.py index bdad546d7..d21b62f83 100644 --- a/metagpt/actions/summarize_code.py +++ b/metagpt/actions/summarize_code.py @@ -11,11 +11,8 @@ from pydantic import Field from tenacity import retry, stop_after_attempt, wait_random_exponential from metagpt.actions.action import Action -from metagpt.config import CONFIG -from metagpt.const import SYSTEM_DESIGN_FILE_REPO, TASK_FILE_REPO from metagpt.logs import logger from metagpt.schema import CodeSummarizeContext -from metagpt.utils.file_repository import FileRepository PROMPT_TEMPLATE = """ NOTICE @@ -29,9 +26,9 @@ ATTENTION: Use '##' to SPLIT SECTIONS, not '#'. Output format carefully referenc {system_design} ``` ----- -# Tasks +# Task ```text -{tasks} +{task} ``` ----- {code_blocks} @@ -90,10 +87,9 @@ flowchart TB """ -# TOTEST class SummarizeCode(Action): name: str = "SummarizeCode" - context: CodeSummarizeContext = Field(default_factory=CodeSummarizeContext) + i_context: CodeSummarizeContext = Field(default_factory=CodeSummarizeContext) @retry(stop=stop_after_attempt(2), wait=wait_random_exponential(min=1, max=60)) async def summarize_code(self, prompt): @@ -101,20 +97,20 @@ class SummarizeCode(Action): return code_rsp async def run(self): - design_pathname = Path(self.context.design_filename) - design_doc = await FileRepository.get_file(filename=design_pathname.name, relative_path=SYSTEM_DESIGN_FILE_REPO) - task_pathname = Path(self.context.task_filename) - task_doc = await FileRepository.get_file(filename=task_pathname.name, relative_path=TASK_FILE_REPO) - src_file_repo = CONFIG.git_repo.new_file_repository(relative_path=CONFIG.src_workspace) + design_pathname = Path(self.i_context.design_filename) + design_doc = await self.repo.docs.system_design.get(filename=design_pathname.name) + task_pathname = Path(self.i_context.task_filename) + task_doc = await self.repo.docs.task.get(filename=task_pathname.name) + src_file_repo = self.repo.with_src_path(self.context.src_workspace).srcs code_blocks = [] - for filename in self.context.codes_filenames: + for filename in self.i_context.codes_filenames: code_doc = await src_file_repo.get(filename) code_block = f"```python\n{code_doc.content}\n```\n-----" code_blocks.append(code_block) format_example = FORMAT_EXAMPLE prompt = PROMPT_TEMPLATE.format( system_design=design_doc.content, - tasks=task_doc.content, + task=task_doc.content, code_blocks="\n".join(code_blocks), format_example=format_example, ) diff --git a/metagpt/actions/talk_action.py b/metagpt/actions/talk_action.py index dc127109c..0aac1c5a0 100644 --- a/metagpt/actions/talk_action.py +++ b/metagpt/actions/talk_action.py @@ -9,25 +9,31 @@ from typing import Optional from metagpt.actions import Action -from metagpt.config import CONFIG -from metagpt.const import DEFAULT_LANGUAGE +from metagpt.config2 import config from metagpt.logs import logger from metagpt.schema import Message -# TOTEST class TalkAction(Action): - context: str + i_context: str history_summary: str = "" knowledge: str = "" rsp: Optional[Message] = None + @property + def agent_description(self): + return self.context.kwargs.agent_description + + @property + def language(self): + return self.context.kwargs.language or config.language + @property def prompt(self): prompt = "" - if CONFIG.agent_description: + if self.agent_description: prompt = ( - f"You are {CONFIG.agent_description}. Your responses should align with the role-play agreement, " + f"You are {self.agent_description}. Your responses should align with the role-play agreement, " f"maintaining the character's persona and habits. When faced with unrelated questions, playfully " f"decline to answer without revealing your AI nature to preserve the character's image.\n\n" ) @@ -36,10 +42,10 @@ class TalkAction(Action): prompt += ( "If the information is insufficient, you can search in the historical conversation or knowledge above.\n" ) - language = CONFIG.language or DEFAULT_LANGUAGE + language = self.language prompt += ( f"Answer the following questions strictly in {language}, and the answers must follow the Markdown format.\n " - f"{self.context}" + f"{self.i_context}" ) logger.debug(f"PROMPT: {prompt}") return prompt @@ -47,11 +53,11 @@ class TalkAction(Action): @property def prompt_gpt4(self): kvs = { - "{role}": CONFIG.agent_description or "", + "{role}": self.agent_description or "", "{history}": self.history_summary or "", "{knowledge}": self.knowledge or "", - "{language}": CONFIG.language or DEFAULT_LANGUAGE, - "{ask}": self.context, + "{language}": self.language, + "{ask}": self.i_context, } prompt = TalkActionPrompt.FORMATION_LOOSE for k, v in kvs.items(): @@ -68,9 +74,9 @@ class TalkAction(Action): @property def aask_args(self): - language = CONFIG.language or DEFAULT_LANGUAGE + language = self.language system_msgs = [ - f"You are {CONFIG.agent_description}.", + f"You are {self.agent_description}.", "Your responses should align with the role-play agreement, " "maintaining the character's persona and habits. When faced with unrelated questions, playfully " "decline to answer without revealing your AI nature to preserve the character's image.", @@ -82,7 +88,7 @@ class TalkAction(Action): format_msgs.append({"role": "assistant", "content": self.knowledge}) if self.history_summary: format_msgs.append({"role": "assistant", "content": self.history_summary}) - return self.context, format_msgs, system_msgs + return self.i_context, format_msgs, system_msgs async def run(self, with_message=None, **kwargs) -> Message: msg, format_msgs, system_msgs = self.aask_args diff --git a/metagpt/actions/write_analysis_code.py b/metagpt/actions/write_analysis_code.py index c6e504b9e..5cea9fe51 100644 --- a/metagpt/actions/write_analysis_code.py +++ b/metagpt/actions/write_analysis_code.py @@ -4,19 +4,12 @@ @Author : orange-crow @File : write_code_v2.py """ -import re -from pathlib import Path from typing import Dict, List, Tuple, Union -from tenacity import retry, stop_after_attempt, wait_fixed - from metagpt.actions import Action -from metagpt.llm import LLM from metagpt.logs import logger -from metagpt.prompts.ml_engineer import ( +from metagpt.prompts.write_analysis_code import ( CODE_GENERATOR_WITH_TOOLS, - GENERATE_CODE_PROMPT, - ML_TOOL_USAGE_PROMPT, SELECT_FUNCTION_TOOLS, TOOL_RECOMMENDATION_PROMPT, TOOL_USAGE_PROMPT, @@ -24,7 +17,7 @@ from metagpt.prompts.ml_engineer import ( from metagpt.schema import Message, Plan from metagpt.tools import TOOL_REGISTRY from metagpt.tools.tool_registry import validate_tool_names -from metagpt.utils.common import create_func_config, remove_comments +from metagpt.utils.common import create_func_config class BaseWriteAnalysisCode(Action): @@ -84,8 +77,8 @@ class WriteCodeByGenerate(BaseWriteAnalysisCode): ) -> dict: # context.append(Message(content=self.REUSE_CODE_INSTRUCTION, role="user")) prompt = self.process_msg(context, system_msg) - code_content = await self.llm.aask_code(prompt, **kwargs) - return code_content + rsp = await self.llm.aask_code(prompt, **kwargs) + return rsp class WriteCodeWithTools(BaseWriteAnalysisCode): @@ -195,133 +188,3 @@ class WriteCodeWithTools(BaseWriteAnalysisCode): rsp = await self.llm.aask_code(prompt, **tool_config) return rsp - - -class WriteCodeWithToolsML(WriteCodeWithTools): - async def run( - self, - context: List[Message], - plan: Plan = None, - column_info: str = "", - **kwargs, - ) -> Tuple[List[Message], str]: - # prepare tool schemas and tool-type-specific instruction - tool_schemas, tool_type_usage_prompt = await self._prepare_tools(plan=plan) - - # ML-specific variables to be used in prompt - code_steps = plan.current_task.code_steps - finished_tasks = plan.get_finished_tasks() - code_context = [remove_comments(task.code) for task in finished_tasks] - code_context = "\n\n".join(code_context) - - # prepare prompt depending on tool availability & LLM call - if tool_schemas: - prompt = ML_TOOL_USAGE_PROMPT.format( - user_requirement=plan.goal, - history_code=code_context, - current_task=plan.current_task.instruction, - column_info=column_info, - tool_type_usage_prompt=tool_type_usage_prompt, - code_steps=code_steps, - tool_schemas=tool_schemas, - ) - - else: - prompt = GENERATE_CODE_PROMPT.format( - user_requirement=plan.goal, - history_code=code_context, - current_task=plan.current_task.instruction, - column_info=column_info, - tool_type_usage_prompt=tool_type_usage_prompt, - code_steps=code_steps, - ) - tool_config = create_func_config(CODE_GENERATOR_WITH_TOOLS) - rsp = await self.llm.aask_code(prompt, **tool_config) - - # Extra output to be used for potential debugging - context = [Message(content=prompt, role="user")] - - return context, rsp - - -class MakeTools(WriteCodeByGenerate): - DEFAULT_SYSTEM_MSG: str = """Convert any codes provied for you to a very General Function Code startswith `def`.\n - **Notice: - 1. Your code must contain a general function start with `def`. - 2. Refactor your code to get the most efficient implementation for large input data in the shortest amount of time. - 3. Must use Google style for function docstring, and your docstring must be consistent with the code,without missing anything. - 4. Write example code after `if __name__ == '__main__':`by using old varibales in old code, - and make sure it could be execute in the user's machine. - 5. Only use the imported packages** - """ - - def __init__(self, name: str = "", context: list[Message] = None, llm: LLM = None, workspace: str = None): - """ - :param str name: name, defaults to '' - :param list[Message] context: context, defaults to None - :param LLM llm: llm, defaults to None - :param str workspace: tools code saved file path dir, defaults to None - """ - super().__init__(name, context, llm) - self.workspace = workspace or str(Path(__file__).parents[1].joinpath("./tools/functions/libs/udf")) - self.file_suffix: str = ".py" - self.context = [] - - def parse_function_name(self, function_code: str) -> str: - # 定义正则表达式模式 - pattern = r"\bdef\s+([a-zA-Z_]\w*)\s*\(" - # 在代码中搜索匹配的模式 - match = re.search(pattern, function_code) - # 如果找到匹配项，则返回匹配的函数名；否则返回None - if match: - return match.group(1) - else: - return None - - def save(self, tool_code: str) -> None: - func_name = self.parse_function_name(tool_code) - if func_name is None: - raise ValueError(f"No function name found in {tool_code}") - saved_path = Path(self.workspace).joinpath(func_name + self.file_suffix) - logger.info(f"Saved tool_code {func_name} in {str(saved_path)}.") - saved_path.write_text(tool_code, encoding="utf-8") - - @retry(stop=stop_after_attempt(3), wait=wait_fixed(1)) - async def run(self, code: Union[str, List[dict]], code_desc: str = None, **kwargs) -> str: - # 拼接code prompt - code_prompt = f"The following code is about {code_desc}, convert it to be a General Function, {code}" - if not self.context: - self.context = self.process_msg(code_prompt) - else: - self.context.append(self.process_msg(code_prompt)[-1]) - logger.info(f"\n\nAsk to Make tools:\n{'-'*60}\n {self.context[-1]}") - - # 更新kwargs - if "code" in kwargs: - kwargs.pop("code") - if "code_desc" in kwargs: - kwargs.pop("code_desc") - - max_tries, current_try = 3, 0 - while True: - tool_code = await self.llm.aask_code(self.context, **kwargs) - func_name = self.parse_function_name(tool_code["code"]) - current_try += 1 - # make tools failed, add error message to context. - if not func_name: - logger.info(f"\n\nTools Respond\n{'-'*60}\n: {tool_code}") - logger.error(f"No function name found in code, we will retry make tools.\n{tool_code['code']}\n") - self.context.append( - {"role": "user", "content": "We need a general function in above code,but not found function."} - ) - # end make tools - if func_name is not None or current_try >= max_tries: - if current_try >= max_tries: - logger.error( - f"We have tried the maximum number of attempts {max_tries}\ - and still have not created tools successfully, we will skip it." - ) - break - logger.info(f"\n\nTools Respond\n{'-'*60}\n: {tool_code}") - self.save(tool_code["code"]) - return tool_code["code"] diff --git a/metagpt/actions/write_code.py b/metagpt/actions/write_code.py index 7377442b5..0b86ac1bb 100644 --- a/metagpt/actions/write_code.py +++ b/metagpt/actions/write_code.py @@ -21,18 +21,17 @@ from pydantic import Field from tenacity import retry, stop_after_attempt, wait_random_exponential from metagpt.actions.action import Action -from metagpt.config import CONFIG +from metagpt.actions.project_management_an import REFINED_TASK_LIST, TASK_LIST +from metagpt.actions.write_code_plan_and_change_an import REFINED_TEMPLATE from metagpt.const import ( BUGFIX_FILENAME, - CODE_SUMMARIES_FILE_REPO, - DOCS_FILE_REPO, - TASK_FILE_REPO, - TEST_OUTPUTS_FILE_REPO, + CODE_PLAN_AND_CHANGE_FILENAME, + REQUIREMENT_FILENAME, ) from metagpt.logs import logger from metagpt.schema import CodingContext, Document, RunCodeResult from metagpt.utils.common import CodeParser -from metagpt.utils.file_repository import FileRepository +from metagpt.utils.project_repo import ProjectRepo PROMPT_TEMPLATE = """ NOTICE @@ -44,8 +43,8 @@ ATTENTION: Use '##' to SPLIT SECTIONS, not '#'. Output format carefully referenc ## Design {design} -## Tasks -{tasks} +## Task +{task} ## Legacy Code ```Code @@ -87,7 +86,7 @@ ATTENTION: Use '##' to SPLIT SECTIONS, not '#'. Output format carefully referenc class WriteCode(Action): name: str = "WriteCode" - context: Document = Field(default_factory=Document) + i_context: Document = Field(default_factory=Document) @retry(wait=wait_random_exponential(min=1, max=60), stop=stop_after_attempt(6)) async def write_code(self, prompt) -> str: @@ -96,16 +95,15 @@ class WriteCode(Action): return code async def run(self, *args, **kwargs) -> CodingContext: - bug_feedback = await FileRepository.get_file(filename=BUGFIX_FILENAME, relative_path=DOCS_FILE_REPO) - coding_context = CodingContext.loads(self.context.content) - test_doc = await FileRepository.get_file( - filename="test_" + coding_context.filename + ".json", relative_path=TEST_OUTPUTS_FILE_REPO - ) + bug_feedback = await self.repo.docs.get(filename=BUGFIX_FILENAME) + coding_context = CodingContext.loads(self.i_context.content) + test_doc = await self.repo.test_outputs.get(filename="test_" + coding_context.filename + ".json") + code_plan_and_change_doc = await self.repo.docs.code_plan_and_change.get(filename=CODE_PLAN_AND_CHANGE_FILENAME) + code_plan_and_change = code_plan_and_change_doc.content if code_plan_and_change_doc else "" + requirement_doc = await self.repo.docs.get(filename=REQUIREMENT_FILENAME) summary_doc = None if coding_context.design_doc and coding_context.design_doc.filename: - summary_doc = await FileRepository.get_file( - filename=coding_context.design_doc.filename, relative_path=CODE_SUMMARIES_FILE_REPO - ) + summary_doc = await self.repo.docs.code_summary.get(filename=coding_context.design_doc.filename) logs = "" if test_doc: test_detail = RunCodeResult.loads(test_doc.content) @@ -113,42 +111,109 @@ class WriteCode(Action): if bug_feedback: code_context = coding_context.code_doc.content + elif code_plan_and_change: + code_context = await self.get_codes( + coding_context.task_doc, exclude=self.i_context.filename, project_repo=self.repo, use_inc=True + ) else: - code_context = await self.get_codes(coding_context.task_doc, exclude=self.context.filename) + code_context = await self.get_codes( + coding_context.task_doc, + exclude=self.i_context.filename, + project_repo=self.repo.with_src_path(self.context.src_workspace), + ) - prompt = PROMPT_TEMPLATE.format( - design=coding_context.design_doc.content if coding_context.design_doc else "", - tasks=coding_context.task_doc.content if coding_context.task_doc else "", - code=code_context, - logs=logs, - feedback=bug_feedback.content if bug_feedback else "", - filename=self.context.filename, - summary_log=summary_doc.content if summary_doc else "", - ) + if code_plan_and_change: + prompt = REFINED_TEMPLATE.format( + user_requirement=requirement_doc.content if requirement_doc else "", + code_plan_and_change=code_plan_and_change, + design=coding_context.design_doc.content if coding_context.design_doc else "", + task=coding_context.task_doc.content if coding_context.task_doc else "", + code=code_context, + logs=logs, + feedback=bug_feedback.content if bug_feedback else "", + filename=self.i_context.filename, + summary_log=summary_doc.content if summary_doc else "", + ) + else: + prompt = PROMPT_TEMPLATE.format( + design=coding_context.design_doc.content if coding_context.design_doc else "", + task=coding_context.task_doc.content if coding_context.task_doc else "", + code=code_context, + logs=logs, + feedback=bug_feedback.content if bug_feedback else "", + filename=self.i_context.filename, + summary_log=summary_doc.content if summary_doc else "", + ) logger.info(f"Writing {coding_context.filename}..") code = await self.write_code(prompt) if not coding_context.code_doc: # avoid root_path pydantic ValidationError if use WriteCode alone - root_path = CONFIG.src_workspace if CONFIG.src_workspace else "" + root_path = self.context.src_workspace if self.context.src_workspace else "" coding_context.code_doc = Document(filename=coding_context.filename, root_path=str(root_path)) coding_context.code_doc.content = code return coding_context @staticmethod - async def get_codes(task_doc, exclude) -> str: + async def get_codes(task_doc: Document, exclude: str, project_repo: ProjectRepo, use_inc: bool = False) -> str: + """ + Get codes for generating the exclude file in various scenarios. + + Attributes: + task_doc (Document): Document object of the task file. + exclude (str): The file to be generated. Specifies the filename to be excluded from the code snippets. + project_repo (ProjectRepo): ProjectRepo object of the project. + use_inc (bool): Indicates whether the scenario involves incremental development. Defaults to False. + + Returns: + str: Codes for generating the exclude file. + """ if not task_doc: return "" if not task_doc.content: - task_doc.content = FileRepository.get_file(filename=task_doc.filename, relative_path=TASK_FILE_REPO) + task_doc = project_repo.docs.task.get(filename=task_doc.filename) m = json.loads(task_doc.content) - code_filenames = m.get("Task list", []) + code_filenames = m.get(TASK_LIST.key, []) if use_inc else m.get(REFINED_TASK_LIST.key, []) codes = [] - src_file_repo = CONFIG.git_repo.new_file_repository(relative_path=CONFIG.src_workspace) - for filename in code_filenames: - if filename == exclude: - continue - doc = await src_file_repo.get(filename=filename) - if not doc: - continue - codes.append(f"----- {filename}\n" + doc.content) + src_file_repo = project_repo.srcs + + # Incremental development scenario + if use_inc: + src_files = src_file_repo.all_files + # Get the old workspace contained the old codes and old workspace are created in previous CodePlanAndChange + old_file_repo = project_repo.git_repo.new_file_repository(relative_path=project_repo.old_workspace) + old_files = old_file_repo.all_files + # Get the union of the files in the src and old workspaces + union_files_list = list(set(src_files) | set(old_files)) + for filename in union_files_list: + # Exclude the current file from the all code snippets + if filename == exclude: + # If the file is in the old workspace, use the old code + # Exclude unnecessary code to maintain a clean and focused main.py file, ensuring only relevant and + # essential functionality is included for the project’s requirements + if filename in old_files and filename != "main.py": + # Use old code + doc = await old_file_repo.get(filename=filename) + # If the file is in the src workspace, skip it + else: + continue + codes.insert(0, f"-----Now, {filename} to be rewritten\n```{doc.content}```\n=====") + # The code snippets are generated from the src workspace + else: + doc = await src_file_repo.get(filename=filename) + # If the file does not exist in the src workspace, skip it + if not doc: + continue + codes.append(f"----- {filename}\n```{doc.content}```") + + # Normal scenario + else: + for filename in code_filenames: + # Exclude the current file to get the code snippets for generating the current file + if filename == exclude: + continue + doc = await src_file_repo.get(filename=filename) + if not doc: + continue + codes.append(f"----- {filename}\n```{doc.content}```") + return "\n".join(codes) diff --git a/metagpt/actions/write_code_an_draft.py b/metagpt/actions/write_code_an_draft.py index 968c8924b..ce030b0e9 100644 --- a/metagpt/actions/write_code_an_draft.py +++ b/metagpt/actions/write_code_an_draft.py @@ -5,7 +5,7 @@ @File : write_review.py """ import asyncio -from typing import List +from typing import List, Literal from metagpt.actions import Action from metagpt.actions.action_node import ActionNode @@ -21,16 +21,15 @@ REVIEW = ActionNode( ], ) -LGTM = ActionNode( - key="LGTM", - expected_type=str, - instruction="LGTM/LBTM. If the code is fully implemented, " - "give a LGTM (Looks Good To Me), otherwise provide a LBTM (Looks Bad To Me).", +REVIEW_RESULT = ActionNode( + key="ReviewResult", + expected_type=Literal["LGTM", "LBTM"], + instruction="LGTM/LBTM. If the code is fully implemented, " "give a LGTM, otherwise provide a LBTM.", example="LBTM", ) -ACTIONS = ActionNode( - key="Actions", +NEXT_STEPS = ActionNode( + key="NextSteps", expected_type=str, instruction="Based on the code review outcome, suggest actionable steps. This can include code changes, " "refactoring suggestions, or any follow-up tasks.", @@ -69,7 +68,7 @@ WRITE_DRAFT = ActionNode( ) -WRITE_MOVE_FUNCTION = ActionNode( +WRITE_FUNCTION = ActionNode( key="WriteFunction", expected_type=str, instruction="write code for the function not implemented.", @@ -555,8 +554,8 @@ LBTM """ -WRITE_CODE_NODE = ActionNode.from_children("WRITE_REVIEW_NODE", [REVIEW, LGTM, ACTIONS]) -WRITE_MOVE_NODE = ActionNode.from_children("WRITE_MOVE_NODE", [WRITE_DRAFT, WRITE_MOVE_FUNCTION]) +WRITE_CODE_NODE = ActionNode.from_children("WRITE_REVIEW_NODE", [REVIEW, REVIEW_RESULT, NEXT_STEPS]) +WRITE_MOVE_NODE = ActionNode.from_children("WRITE_MOVE_NODE", [WRITE_DRAFT, WRITE_FUNCTION]) CR_FOR_MOVE_FUNCTION_BY_3 = """ @@ -579,8 +578,7 @@ class WriteCodeAN(Action): async def run(self, context): self.llm.system_prompt = "You are an outstanding engineer and can implement any code" - return await WRITE_MOVE_FUNCTION.fill(context=context, llm=self.llm, schema="json") - # return await WRITE_CODE_NODE.fill(context=context, llm=self.llm, schema="markdown") + return await WRITE_MOVE_NODE.fill(context=context, llm=self.llm, schema="json") async def main(): diff --git a/metagpt/actions/write_code_plan_and_change_an.py b/metagpt/actions/write_code_plan_and_change_an.py new file mode 100644 index 000000000..708808050 --- /dev/null +++ b/metagpt/actions/write_code_plan_and_change_an.py @@ -0,0 +1,210 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2023/12/26 +@Author : mannaandpoem +@File : write_code_plan_and_change_an.py +""" +import os + +from pydantic import Field + +from metagpt.actions.action import Action +from metagpt.actions.action_node import ActionNode +from metagpt.schema import CodePlanAndChangeContext + +CODE_PLAN_AND_CHANGE = ActionNode( + key="Code Plan And Change", + expected_type=str, + instruction="Developing comprehensive and step-by-step incremental development plan, and write Incremental " + "Change by making a code draft that how to implement incremental development including detailed steps based on the " + "context. Note: Track incremental changes using mark of '+' or '-' for add/modify/delete code, and conforms to the " + "output format of git diff", + example=""" +1. Plan for calculator.py: Enhance the functionality of `calculator.py` by extending it to incorporate methods for subtraction, multiplication, and division. Additionally, implement robust error handling for the division operation to mitigate potential issues related to division by zero. +```python +class Calculator: + self.result = number1 + number2 + return self.result + +- def sub(self, number1, number2) -> float: ++ def subtract(self, number1: float, number2: float) -> float: ++ ''' ++ Subtracts the second number from the first and returns the result. ++ ++ Args: ++ number1 (float): The number to be subtracted from. ++ number2 (float): The number to subtract. ++ ++ Returns: ++ float: The difference of number1 and number2. ++ ''' ++ self.result = number1 - number2 ++ return self.result ++ + def multiply(self, number1: float, number2: float) -> float: +- pass ++ ''' ++ Multiplies two numbers and returns the result. ++ ++ Args: ++ number1 (float): The first number to multiply. ++ number2 (float): The second number to multiply. ++ ++ Returns: ++ float: The product of number1 and number2. ++ ''' ++ self.result = number1 * number2 ++ return self.result ++ + def divide(self, number1: float, number2: float) -> float: +- pass ++ ''' ++ ValueError: If the second number is zero. ++ ''' ++ if number2 == 0: ++ raise ValueError('Cannot divide by zero') ++ self.result = number1 / number2 ++ return self.result ++ +- def reset_result(self): ++ def clear(self): ++ if self.result != 0.0: ++ print("Result is not zero, clearing...") ++ else: ++ print("Result is already zero, no need to clear.") ++ + self.result = 0.0 +``` + +2. Plan for main.py: Integrate new API endpoints for subtraction, multiplication, and division into the existing codebase of `main.py`. Then, ensure seamless integration with the overall application architecture and maintain consistency with coding standards. +```python +def add_numbers(): + result = calculator.add_numbers(num1, num2) + return jsonify({'result': result}), 200 + +-# TODO: Implement subtraction, multiplication, and division operations ++@app.route('/subtract_numbers', methods=['POST']) ++def subtract_numbers(): ++ data = request.get_json() ++ num1 = data.get('num1', 0) ++ num2 = data.get('num2', 0) ++ result = calculator.subtract_numbers(num1, num2) ++ return jsonify({'result': result}), 200 ++ ++@app.route('/multiply_numbers', methods=['POST']) ++def multiply_numbers(): ++ data = request.get_json() ++ num1 = data.get('num1', 0) ++ num2 = data.get('num2', 0) ++ try: ++ result = calculator.divide_numbers(num1, num2) ++ except ValueError as e: ++ return jsonify({'error': str(e)}), 400 ++ return jsonify({'result': result}), 200 ++ + if __name__ == '__main__': + app.run() +```""", +) + +CODE_PLAN_AND_CHANGE_CONTEXT = """ +## User New Requirements +{requirement} + +## PRD +{prd} + +## Design +{design} + +## Task +{task} + +## Legacy Code +{code} +""" + +REFINED_TEMPLATE = """ +NOTICE +Role: You are a professional engineer; The main goal is to complete incremental development by combining legacy code and plan and Incremental Change, ensuring the integration of new features. + +# Context +## User New Requirements +{user_requirement} + +## Code Plan And Change +{code_plan_and_change} + +## Design +{design} + +## Task +{task} + +## Legacy Code +```Code +{code} +``` + +## Debug logs +```text +{logs} + +{summary_log} +``` + +## Bug Feedback logs +```text +{feedback} +``` + +# Format example +## Code: {filename} +```python +## {filename} +... +``` + +# Instruction: Based on the context, follow "Format example", write or rewrite code. +## Write/Rewrite Code: Only write one file {filename}, write or rewrite complete code using triple quotes based on the following attentions and context. +1. Only One file: do your best to implement THIS ONLY ONE FILE. +2. COMPLETE CODE: Your code will be part of the entire project, so please implement complete, reliable, reusable code snippets. +3. Set default value: If there is any setting, ALWAYS SET A DEFAULT VALUE, ALWAYS USE STRONG TYPE AND EXPLICIT VARIABLE. AVOID circular import. +4. Follow design: YOU MUST FOLLOW "Data structures and interfaces". DONT CHANGE ANY DESIGN. Do not use public member functions that do not exist in your design. +5. Follow Code Plan And Change: If there is any Incremental Change that is marked by the git diff format using '+' and '-' for add/modify/delete code, or Legacy Code files contain "{filename} to be rewritten", you must merge it into the code file according to the plan. +6. CAREFULLY CHECK THAT YOU DONT MISS ANY NECESSARY CLASS/FUNCTION IN THIS FILE. +7. Before using a external variable/module, make sure you import it first. +8. Write out EVERY CODE DETAIL, DON'T LEAVE TODO. +9. Attention: Retain details that are not related to incremental development but are important for maintaining the consistency and clarity of the old code. +""" + +WRITE_CODE_PLAN_AND_CHANGE_NODE = ActionNode.from_children("WriteCodePlanAndChange", [CODE_PLAN_AND_CHANGE]) + + +class WriteCodePlanAndChange(Action): + name: str = "WriteCodePlanAndChange" + i_context: CodePlanAndChangeContext = Field(default_factory=CodePlanAndChangeContext) + + async def run(self, *args, **kwargs): + self.llm.system_prompt = "You are a professional software engineer, your primary responsibility is to " + "meticulously craft comprehensive incremental development plan and deliver detailed incremental change" + prd_doc = await self.repo.docs.prd.get(filename=self.i_context.prd_filename) + design_doc = await self.repo.docs.system_design.get(filename=self.i_context.design_filename) + task_doc = await self.repo.docs.task.get(filename=self.i_context.task_filename) + code_text = await self.get_old_codes() + context = CODE_PLAN_AND_CHANGE_CONTEXT.format( + requirement=self.i_context.requirement, + prd=prd_doc.content, + design=design_doc.content, + task=task_doc.content, + code=code_text, + ) + return await WRITE_CODE_PLAN_AND_CHANGE_NODE.fill(context=context, llm=self.llm, schema="json") + + async def get_old_codes(self) -> str: + self.repo.old_workspace = self.repo.git_repo.workdir / os.path.basename(self.config.project_path) + old_file_repo = self.repo.git_repo.new_file_repository(relative_path=self.repo.old_workspace) + old_codes = await old_file_repo.get_all() + codes = [f"----- {code.filename}\n```{code.content}```" for code in old_codes] + return "\n".join(codes) diff --git a/metagpt/actions/write_code_review.py b/metagpt/actions/write_code_review.py index a8c913573..da636eb36 100644 --- a/metagpt/actions/write_code_review.py +++ b/metagpt/actions/write_code_review.py @@ -13,7 +13,7 @@ from tenacity import retry, stop_after_attempt, wait_random_exponential from metagpt.actions import WriteCode from metagpt.actions.action import Action -from metagpt.config import CONFIG +from metagpt.const import CODE_PLAN_AND_CHANGE_FILENAME, REQUIREMENT_FILENAME from metagpt.logs import logger from metagpt.schema import CodingContext from metagpt.utils.common import CodeParser @@ -120,7 +120,7 @@ REWRITE_CODE_TEMPLATE = """ class WriteCodeReview(Action): name: str = "WriteCodeReview" - context: CodingContext = Field(default_factory=CodingContext) + i_context: CodingContext = Field(default_factory=CodingContext) @retry(wait=wait_random_exponential(min=1, max=60), stop=stop_after_attempt(6)) async def write_code_review_and_rewrite(self, context_prompt, cr_prompt, filename): @@ -136,41 +136,64 @@ class WriteCodeReview(Action): return result, code async def run(self, *args, **kwargs) -> CodingContext: - iterative_code = self.context.code_doc.content - k = CONFIG.code_review_k_times or 1 + iterative_code = self.i_context.code_doc.content + k = self.context.config.code_review_k_times or 1 + for i in range(k): - format_example = FORMAT_EXAMPLE.format(filename=self.context.code_doc.filename) - task_content = self.context.task_doc.content if self.context.task_doc else "" - code_context = await WriteCode.get_codes(self.context.task_doc, exclude=self.context.filename) - context = "\n".join( - [ - "## System Design\n" + str(self.context.design_doc) + "\n", - "## Tasks\n" + task_content + "\n", - "## Code Files\n" + code_context + "\n", - ] + format_example = FORMAT_EXAMPLE.format(filename=self.i_context.code_doc.filename) + task_content = self.i_context.task_doc.content if self.i_context.task_doc else "" + code_context = await WriteCode.get_codes( + self.i_context.task_doc, + exclude=self.i_context.filename, + project_repo=self.repo.with_src_path(self.context.src_workspace), + use_inc=self.config.inc, ) + + if not self.config.inc: + context = "\n".join( + [ + "## System Design\n" + str(self.i_context.design_doc) + "\n", + "## Task\n" + task_content + "\n", + "## Code Files\n" + code_context + "\n", + ] + ) + else: + requirement_doc = await self.repo.docs.get(filename=REQUIREMENT_FILENAME) + code_plan_and_change_doc = await self.repo.get(filename=CODE_PLAN_AND_CHANGE_FILENAME) + context = "\n".join( + [ + "## User New Requirements\n" + str(requirement_doc) + "\n", + "## Code Plan And Change\n" + str(code_plan_and_change_doc) + "\n", + "## System Design\n" + str(self.i_context.design_doc) + "\n", + "## Task\n" + task_content + "\n", + "## Code Files\n" + code_context + "\n", + ] + ) + context_prompt = PROMPT_TEMPLATE.format( context=context, code=iterative_code, - filename=self.context.code_doc.filename, + filename=self.i_context.code_doc.filename, ) cr_prompt = EXAMPLE_AND_INSTRUCTION.format( format_example=format_example, ) + len1 = len(iterative_code) if iterative_code else 0 + len2 = len(self.i_context.code_doc.content) if self.i_context.code_doc.content else 0 logger.info( - f"Code review and rewrite {self.context.code_doc.filename}: {i + 1}/{k} | {len(iterative_code)=}, " - f"{len(self.context.code_doc.content)=}" + f"Code review and rewrite {self.i_context.code_doc.filename}: {i + 1}/{k} | len(iterative_code)={len1}, " + f"len(self.i_context.code_doc.content)={len2}" ) result, rewrited_code = await self.write_code_review_and_rewrite( - context_prompt, cr_prompt, self.context.code_doc.filename + context_prompt, cr_prompt, self.i_context.code_doc.filename ) if "LBTM" in result: iterative_code = rewrited_code elif "LGTM" in result: - self.context.code_doc.content = iterative_code - return self.context + self.i_context.code_doc.content = iterative_code + return self.i_context # code_rsp = await self._aask_v1(prompt, "code_rsp", OUTPUT_MAPPING) # self._save(context, filename, code) # 如果rewrited_code是None（原code perfect），那么直接返回code - self.context.code_doc.content = iterative_code - return self.context + self.i_context.code_doc.content = iterative_code + return self.i_context diff --git a/metagpt/actions/write_code_steps.py b/metagpt/actions/write_code_steps.py deleted file mode 100644 index 7ba22fde4..000000000 --- a/metagpt/actions/write_code_steps.py +++ /dev/null @@ -1,116 +0,0 @@ -import json - -from metagpt.actions import Action -from metagpt.schema import Plan -from metagpt.utils.common import CodeParser - -# CODE_STEPS_PROMPT_TEMPLATE = """ -# # Context -# {context} -# -# ----- -# Tasks are all code development tasks. -# You are a professional engineer, the main goal is to plan out concise solution steps for Current Task before coding. -# A planning process can reduce the difficulty and improve the quality of coding. -# You may be given some code plans for the tasks ahead, but you don't have to follow the existing plan when planning the current task. -# The output plan should following the subsequent principles: -# 1.The plan is a rough checklist of steps outlining the entire program's structure.Try to keep the number of steps fewer than 5. -# 2.The steps should be written concisely and at a high level, avoiding overly detailed implementation specifics. -# 3.The execution of the plan happens sequentially, but the plan can incorporate conditional (if) and looping(loop) keywords for more complex structures. -# -# Output the code steps in a JSON format, as shown in this example: -# ```json -# { -# "Step 1": "", -# "Step 2": "", -# "Step 3": "", -# ... -# } -# ``` -# """ - -CODE_STEPS_PROMPT_TEMPLATE = """ -# Context -{context} - ------ -Tasks are all code development tasks. -You are a professional engineer, the main goal is to plan out concise solution steps for Current Task before coding. -A planning process can reduce the difficulty and improve the quality of coding. -You may be given some code plans for the tasks ahead, but you don't have to follow the existing plan when planning the current task. -The output plan should following the subsequent principles: -1.The plan is a rough checklist of steps outlining the entire program's structure.Try to keep the number of steps fewer than 5. -2.The steps should be written concisely and at a high level, avoiding overly detailed implementation specifics. -3.The execution of the plan happens sequentially, but the plan can incorporate conditional (if) and looping(loop) keywords for more complex structures. -4.Design and provide code steps by following the code logic. Analyze the provided code step by step and reuse the imported library. - -Output the code steps in a JSON format, as shown in this example: -```json -{ - "Step 1": "", - "Step 2": "", - "Step 3": "", - ... -} -``` -""" - -# STRUCTURAL_CONTEXT = """ -# ## User Requirement -# {user_requirement} -# ## Current Plan -# {tasks} -# ## Current Task -# {current_task} -# """ - -STRUCTURAL_CONTEXT = """ -## User Requirement -{user_requirement} -## Plan -{tasks} -## Codes -{codes} -## Current Task -{current_task} -""" - - -class WriteCodeSteps(Action): - async def run(self, plan: Plan) -> str: - """Run of a task guide writing action, used in ml engineer - - Args: - plan (plan): task plan - useful_memories (list): useful_memories - Returns: - str: The dataset_descriptions string. - """ - - context = self.get_context(plan) - code_steps_prompt = CODE_STEPS_PROMPT_TEMPLATE.replace("{context}", context) - code_steps = await self._aask(code_steps_prompt) - code_steps = CodeParser.parse_code(block=None, text=code_steps) - return code_steps - - def get_context(self, plan: Plan): - user_requirement = plan.goal - # select_task_keys = ['task_id', 'instruction', 'is_finished', 'code'] - # select_task_keys = ['task_id','instruction'] - - def process_task(task): - task_dict = task.dict() - # ptask = {k: task_dict[k] for k in task_dict if k in select_task_keys } - ptask = f"task_id_{task_dict['task_id']}:{task_dict['instruction']}" - return ptask - - tasks = json.dumps([process_task(task) for task in plan.tasks], indent=4, ensure_ascii=False) - - code_lists = [task.code for task in plan.tasks if task.is_finished == True] - codes = "\n\n".join(code_lists) - current_task = json.dumps(process_task(plan.current_task)) if plan.current_task else {} - context = STRUCTURAL_CONTEXT.format( - user_requirement=user_requirement, tasks=tasks, codes=codes, current_task=current_task - ) - # print(context) - return context diff --git a/metagpt/actions/write_docstring.py b/metagpt/actions/write_docstring.py index 8b8335517..79204e6a4 100644 --- a/metagpt/actions/write_docstring.py +++ b/metagpt/actions/write_docstring.py @@ -161,7 +161,7 @@ class WriteDocstring(Action): """ desc: str = "Write docstring for code." - context: Optional[str] = None + i_context: Optional[str] = None async def run( self, diff --git a/metagpt/actions/write_plan.py b/metagpt/actions/write_plan.py index 60dcef43b..335a09841 100644 --- a/metagpt/actions/write_plan.py +++ b/metagpt/actions/write_plan.py @@ -10,7 +10,10 @@ from typing import Dict, List, Tuple from metagpt.actions import Action from metagpt.logs import logger -from metagpt.prompts.ml_engineer import ASSIGN_TASK_TYPE_CONFIG, ASSIGN_TASK_TYPE_PROMPT +from metagpt.prompts.write_analysis_code import ( + ASSIGN_TASK_TYPE_CONFIG, + ASSIGN_TASK_TYPE_PROMPT, +) from metagpt.schema import Message, Plan, Task from metagpt.tools import TOOL_REGISTRY from metagpt.utils.common import CodeParser, create_func_config diff --git a/metagpt/actions/write_prd.py b/metagpt/actions/write_prd.py index 073d8c076..823786893 100644 --- a/metagpt/actions/write_prd.py +++ b/metagpt/actions/write_prd.py @@ -14,26 +14,22 @@ from __future__ import annotations import json -import uuid from pathlib import Path -from typing import Optional from metagpt.actions import Action, ActionOutput from metagpt.actions.action_node import ActionNode from metagpt.actions.fix_bug import FixBug from metagpt.actions.write_prd_an import ( + COMPETITIVE_QUADRANT_CHART, PROJECT_NAME, + REFINED_PRD_NODE, WP_IS_RELATIVE_NODE, WP_ISSUE_TYPE_NODE, WRITE_PRD_NODE, ) -from metagpt.config import CONFIG from metagpt.const import ( BUGFIX_FILENAME, COMPETITIVE_ANALYSIS_FILE_REPO, - DOCS_FILE_REPO, - PRD_PDF_FILE_REPO, - PRDS_FILE_REPO, REQUIREMENT_FILENAME, ) from metagpt.logs import logger @@ -63,135 +59,114 @@ NEW_REQ_TEMPLATE = """ class WritePRD(Action): - name: str = "WritePRD" - content: Optional[str] = None + """WritePRD deal with the following situations: + 1. Bugfix: If the requirement is a bugfix, the bugfix document will be generated. + 2. New requirement: If the requirement is a new requirement, the PRD document will be generated. + 3. Requirement update: If the requirement is an update, the PRD document will be updated. + """ - async def run(self, with_messages, schema=CONFIG.prompt_schema, *args, **kwargs) -> ActionOutput | Message: - # Determine which requirement documents need to be rewritten: Use LLM to assess whether new requirements are - # related to the PRD. If they are related, rewrite the PRD. - docs_file_repo = CONFIG.git_repo.new_file_repository(relative_path=DOCS_FILE_REPO) - requirement_doc = await docs_file_repo.get(filename=REQUIREMENT_FILENAME) - if requirement_doc and await self._is_bugfix(requirement_doc.content): - await docs_file_repo.save(filename=BUGFIX_FILENAME, content=requirement_doc.content) - await docs_file_repo.save(filename=REQUIREMENT_FILENAME, content="") - bug_fix = BugFixContext(filename=BUGFIX_FILENAME) - return Message( - content=bug_fix.model_dump_json(), - instruct_content=bug_fix, - role="", - cause_by=FixBug, - sent_from=self, - send_to="Alex", # the name of Engineer - ) + async def run(self, with_messages, *args, **kwargs) -> ActionOutput | Message: + """Run the action.""" + req: Document = await self.repo.requirement + docs: list[Document] = await self.repo.docs.prd.get_all() + if not req: + raise FileNotFoundError("No requirement document found.") + + if await self._is_bugfix(req.content): + logger.info(f"Bugfix detected: {req.content}") + return await self._handle_bugfix(req) + # remove bugfix file from last round in case of conflict + await self.repo.docs.delete(filename=BUGFIX_FILENAME) + + # if requirement is related to other documents, update them, otherwise create a new one + if related_docs := await self.get_related_docs(req, docs): + logger.info(f"Requirement update detected: {req.content}") + return await self._handle_requirement_update(req, related_docs) else: - await docs_file_repo.delete(filename=BUGFIX_FILENAME) + logger.info(f"New requirement detected: {req.content}") + return await self._handle_new_requirement(req) - prds_file_repo = CONFIG.git_repo.new_file_repository(PRDS_FILE_REPO) - prd_docs = await prds_file_repo.get_all() - change_files = Documents() - for prd_doc in prd_docs: - prd_doc = await self._update_prd( - requirement_doc=requirement_doc, prd_doc=prd_doc, prds_file_repo=prds_file_repo, *args, **kwargs - ) - if not prd_doc: - continue - change_files.docs[prd_doc.filename] = prd_doc - logger.info(f"rewrite prd: {prd_doc.filename}") - # If there is no existing PRD, generate one using 'docs/requirement.txt'. - if not change_files.docs: - prd_doc = await self._update_prd( - requirement_doc=requirement_doc, prd_doc=None, prds_file_repo=prds_file_repo, *args, **kwargs - ) - if prd_doc: - change_files.docs[prd_doc.filename] = prd_doc - logger.debug(f"new prd: {prd_doc.filename}") - # Once all files under 'docs/prds/' have been compared with the newly added requirements, trigger the - # 'publish' message to transition the workflow to the next stage. This design allows room for global - # optimization in subsequent steps. - return ActionOutput(content=change_files.model_dump_json(), instruct_content=change_files) + async def _handle_bugfix(self, req: Document) -> Message: + # ... bugfix logic ... + await self.repo.docs.save(filename=BUGFIX_FILENAME, content=req.content) + await self.repo.docs.save(filename=REQUIREMENT_FILENAME, content="") + bug_fix = BugFixContext(filename=BUGFIX_FILENAME) + return Message( + content=bug_fix.model_dump_json(), + instruct_content=bug_fix, + role="", + cause_by=FixBug, + sent_from=self, + send_to="Alex", # the name of Engineer + ) - async def _run_new_requirement(self, requirements, schema=CONFIG.prompt_schema) -> ActionOutput: - # sas = SearchAndSummarize() - # # rsp = await sas.run(context=requirements, system_text=SEARCH_AND_SUMMARIZE_SYSTEM_EN_US) - # rsp = "" - # info = f"### Search Results\n{sas.result}\n\n### Search Summary\n{rsp}" - # if sas.result: - # logger.info(sas.result) - # logger.info(rsp) - project_name = CONFIG.project_name or "" - context = CONTEXT_TEMPLATE.format(requirements=requirements, project_name=project_name) + async def _handle_new_requirement(self, req: Document) -> ActionOutput: + """handle new requirement""" + project_name = self.project_name + context = CONTEXT_TEMPLATE.format(requirements=req, project_name=project_name) exclude = [PROJECT_NAME.key] if project_name else [] node = await WRITE_PRD_NODE.fill(context=context, llm=self.llm, exclude=exclude) # schema=schema await self._rename_workspace(node) - return node + new_prd_doc = await self.repo.docs.prd.save( + filename=FileRepository.new_filename() + ".json", content=node.instruct_content.model_dump_json() + ) + await self._save_competitive_analysis(new_prd_doc) + await self.repo.resources.prd.save_pdf(doc=new_prd_doc) + return Documents.from_iterable(documents=[new_prd_doc]).to_action_output() - async def _is_relative(self, new_requirement_doc, old_prd_doc) -> bool: - context = NEW_REQ_TEMPLATE.format(old_prd=old_prd_doc.content, requirements=new_requirement_doc.content) + async def _handle_requirement_update(self, req: Document, related_docs: list[Document]) -> ActionOutput: + # ... requirement update logic ... + for doc in related_docs: + await self._update_prd(req, doc) + return Documents.from_iterable(documents=related_docs).to_action_output() + + async def _is_bugfix(self, context: str) -> bool: + if not self.repo.code_files_exists(): + return False + node = await WP_ISSUE_TYPE_NODE.fill(context, self.llm) + return node.get("issue_type") == "BUG" + + async def get_related_docs(self, req: Document, docs: list[Document]) -> list[Document]: + """get the related documents""" + # refine: use gather to speed up + return [i for i in docs if await self._is_related(req, i)] + + async def _is_related(self, req: Document, old_prd: Document) -> bool: + context = NEW_REQ_TEMPLATE.format(old_prd=old_prd.content, requirements=req.content) node = await WP_IS_RELATIVE_NODE.fill(context, self.llm) return node.get("is_relative") == "YES" - async def _merge(self, new_requirement_doc, prd_doc, schema=CONFIG.prompt_schema) -> Document: - if not CONFIG.project_name: - CONFIG.project_name = Path(CONFIG.project_path).name - prompt = NEW_REQ_TEMPLATE.format(requirements=new_requirement_doc.content, old_prd=prd_doc.content) - node = await WRITE_PRD_NODE.fill(context=prompt, llm=self.llm, schema=schema) - prd_doc.content = node.instruct_content.model_dump_json() + async def _merge(self, req: Document, related_doc: Document) -> Document: + if not self.project_name: + self.project_name = Path(self.project_path).name + prompt = NEW_REQ_TEMPLATE.format(requirements=req.content, old_prd=related_doc.content) + node = await REFINED_PRD_NODE.fill(context=prompt, llm=self.llm, schema=self.prompt_schema) + related_doc.content = node.instruct_content.model_dump_json() await self._rename_workspace(node) - return prd_doc + return related_doc - async def _update_prd(self, requirement_doc, prd_doc, prds_file_repo, *args, **kwargs) -> Document | None: - if not prd_doc: - prd = await self._run_new_requirement( - requirements=[requirement_doc.content if requirement_doc else ""], *args, **kwargs - ) - new_prd_doc = Document( - root_path=PRDS_FILE_REPO, - filename=FileRepository.new_filename() + ".json", - content=prd.instruct_content.model_dump_json(), - ) - elif await self._is_relative(requirement_doc, prd_doc): - new_prd_doc = await self._merge(requirement_doc, prd_doc) - else: - return None - await prds_file_repo.save(filename=new_prd_doc.filename, content=new_prd_doc.content) + async def _update_prd(self, req: Document, prd_doc: Document) -> Document: + new_prd_doc: Document = await self._merge(req, prd_doc) + await self.repo.docs.prd.save_doc(doc=new_prd_doc) await self._save_competitive_analysis(new_prd_doc) - await self._save_pdf(new_prd_doc) + await self.repo.resources.prd.save_pdf(doc=new_prd_doc) return new_prd_doc - @staticmethod - async def _save_competitive_analysis(prd_doc): + async def _save_competitive_analysis(self, prd_doc: Document): m = json.loads(prd_doc.content) - quadrant_chart = m.get("Competitive Quadrant Chart") + quadrant_chart = m.get(COMPETITIVE_QUADRANT_CHART.key) if not quadrant_chart: return - pathname = ( - CONFIG.git_repo.workdir / Path(COMPETITIVE_ANALYSIS_FILE_REPO) / Path(prd_doc.filename).with_suffix("") - ) - if not pathname.parent.exists(): - pathname.parent.mkdir(parents=True, exist_ok=True) - await mermaid_to_file(quadrant_chart, pathname) + pathname = self.repo.workdir / COMPETITIVE_ANALYSIS_FILE_REPO / Path(prd_doc.filename).stem + pathname.parent.mkdir(parents=True, exist_ok=True) + await mermaid_to_file(self.config.mermaid_engine, quadrant_chart, pathname) - @staticmethod - async def _save_pdf(prd_doc): - await FileRepository.save_as(doc=prd_doc, with_suffix=".md", relative_path=PRD_PDF_FILE_REPO) - - @staticmethod - async def _rename_workspace(prd): - if not CONFIG.project_name: + async def _rename_workspace(self, prd): + if not self.project_name: if isinstance(prd, (ActionOutput, ActionNode)): ws_name = prd.instruct_content.model_dump()["Project Name"] else: ws_name = CodeParser.parse_str(block="Project Name", text=prd) if ws_name: - CONFIG.project_name = ws_name - if not CONFIG.project_name: # The LLM failed to provide a project name, and the user didn't provide one either. - CONFIG.project_name = "app" + uuid.uuid4().hex[:16] - CONFIG.git_repo.rename_root(CONFIG.project_name) - - async def _is_bugfix(self, context) -> bool: - src_workspace_path = CONFIG.git_repo.workdir / CONFIG.git_repo.workdir.name - code_files = CONFIG.git_repo.get_files(relative_path=src_workspace_path) - if not code_files: - return False - node = await WP_ISSUE_TYPE_NODE.fill(context, self.llm) - return node.get("issue_type") == "BUG" + self.project_name = ws_name + self.repo.git_repo.rename_root(self.project_name) diff --git a/metagpt/actions/write_prd_an.py b/metagpt/actions/write_prd_an.py index 948d7d62f..9898be55b 100644 --- a/metagpt/actions/write_prd_an.py +++ b/metagpt/actions/write_prd_an.py @@ -8,7 +8,6 @@ from typing import List from metagpt.actions.action_node import ActionNode -from metagpt.logs import logger LANGUAGE = ActionNode( key="Language", @@ -31,10 +30,18 @@ ORIGINAL_REQUIREMENTS = ActionNode( example="Create a 2048 game", ) +REFINED_REQUIREMENTS = ActionNode( + key="Refined Requirements", + expected_type=str, + instruction="Place the New user's original requirements here.", + example="Create a 2048 game with a new feature that ...", +) + PROJECT_NAME = ActionNode( key="Project Name", expected_type=str, - instruction="According to the content of \"Original Requirements,\" name the project using snake case style , like 'game_2048' or 'simple_crm.", + instruction='According to the content of "Original Requirements," name the project using snake case style , ' + "like 'game_2048' or 'simple_crm.", example="game_2048", ) @@ -45,6 +52,18 @@ PRODUCT_GOALS = ActionNode( example=["Create an engaging user experience", "Improve accessibility, be responsive", "More beautiful UI"], ) +REFINED_PRODUCT_GOALS = ActionNode( + key="Refined Product Goals", + expected_type=List[str], + instruction="Update and expand the original product goals to reflect the evolving needs due to incremental " + "development.Ensure that the refined goals align with the current project direction and contribute to its success.", + example=[ + "Enhance user engagement through new features", + "Optimize performance for scalability", + "Integrate innovative UI enhancements", + ], +) + USER_STORIES = ActionNode( key="User Stories", expected_type=List[str], @@ -58,6 +77,20 @@ USER_STORIES = ActionNode( ], ) +REFINED_USER_STORIES = ActionNode( + key="Refined User Stories", + expected_type=List[str], + instruction="Update and expand the original scenario-based user stories to reflect the evolving needs due to " + "incremental development. Ensure that the refined user stories capture incremental features and improvements. ", + example=[ + "As a player, I want to choose difficulty levels to challenge my skills", + "As a player, I want a visually appealing score display after each game for a better gaming experience", + "As a player, I want a convenient restart button displayed when I lose to quickly start a new game", + "As a player, I want an enhanced and aesthetically pleasing UI to elevate the overall gaming experience", + "As a player, I want the ability to play the game seamlessly on my mobile phone for on-the-go entertainment", + ], +) + COMPETITIVE_ANALYSIS = ActionNode( key="Competitive Analysis", expected_type=List[str], @@ -97,6 +130,15 @@ REQUIREMENT_ANALYSIS = ActionNode( example="", ) +REFINED_REQUIREMENT_ANALYSIS = ActionNode( + key="Refined Requirement Analysis", + expected_type=List[str], + instruction="Review and refine the existing requirement analysis to align with the evolving needs of the project " + "due to incremental development. Ensure the analysis comprehensively covers the new features and enhancements " + "required for the refined project scope.", + example=["Require add/update/modify ..."], +) + REQUIREMENT_POOL = ActionNode( key="Requirement Pool", expected_type=List[List[str]], @@ -104,6 +146,14 @@ REQUIREMENT_POOL = ActionNode( example=[["P0", "The main code ..."], ["P0", "The game algorithm ..."]], ) +REFINED_REQUIREMENT_POOL = ActionNode( + key="Refined Requirement Pool", + expected_type=List[List[str]], + instruction="List down the top 5 to 7 requirements with their priority (P0, P1, P2). " + "Cover both legacy content and incremental content. Retain content unrelated to incremental development", + example=[["P0", "The main code ..."], ["P0", "The game algorithm ..."]], +) + UI_DESIGN_DRAFT = ActionNode( key="UI Design draft", expected_type=str, @@ -152,15 +202,22 @@ NODES = [ ANYTHING_UNCLEAR, ] +REFINED_NODES = [ + LANGUAGE, + PROGRAMMING_LANGUAGE, + REFINED_REQUIREMENTS, + PROJECT_NAME, + REFINED_PRODUCT_GOALS, + REFINED_USER_STORIES, + COMPETITIVE_ANALYSIS, + COMPETITIVE_QUADRANT_CHART, + REFINED_REQUIREMENT_ANALYSIS, + REFINED_REQUIREMENT_POOL, + UI_DESIGN_DRAFT, + ANYTHING_UNCLEAR, +] + WRITE_PRD_NODE = ActionNode.from_children("WritePRD", NODES) +REFINED_PRD_NODE = ActionNode.from_children("RefinedPRD", REFINED_NODES) WP_ISSUE_TYPE_NODE = ActionNode.from_children("WP_ISSUE_TYPE", [ISSUE_TYPE, REASON]) WP_IS_RELATIVE_NODE = ActionNode.from_children("WP_IS_RELATIVE", [IS_RELATIVE, REASON]) - - -def main(): - prompt = WRITE_PRD_NODE.compile(context="") - logger.info(prompt) - - -if __name__ == "__main__": - main() diff --git a/metagpt/actions/write_prd_review.py b/metagpt/actions/write_prd_review.py index 2babe38db..68fb5d9e8 100644 --- a/metagpt/actions/write_prd_review.py +++ b/metagpt/actions/write_prd_review.py @@ -13,7 +13,7 @@ from metagpt.actions.action import Action class WritePRDReview(Action): name: str = "" - context: Optional[str] = None + i_context: Optional[str] = None prd: Optional[str] = None desc: str = "Based on the PRD, conduct a PRD Review, providing clear and detailed feedback" diff --git a/metagpt/actions/write_teaching_plan.py b/metagpt/actions/write_teaching_plan.py index b824e055e..c5f70ae05 100644 --- a/metagpt/actions/write_teaching_plan.py +++ b/metagpt/actions/write_teaching_plan.py @@ -8,14 +8,14 @@ from typing import Optional from metagpt.actions import Action -from metagpt.config import CONFIG +from metagpt.context import Context from metagpt.logs import logger class WriteTeachingPlanPart(Action): """Write Teaching Plan Part""" - context: Optional[str] = None + i_context: Optional[str] = None topic: str = "" language: str = "Chinese" rsp: Optional[str] = None @@ -24,7 +24,7 @@ class WriteTeachingPlanPart(Action): statement_patterns = TeachingPlanBlock.TOPIC_STATEMENTS.get(self.topic, []) statements = [] for p in statement_patterns: - s = self.format_value(p) + s = self.format_value(p, context=self.context) statements.append(s) formatter = ( TeachingPlanBlock.PROMPT_TITLE_TEMPLATE @@ -35,7 +35,7 @@ class WriteTeachingPlanPart(Action): formation=TeachingPlanBlock.FORMATION, role=self.prefix, statements="\n".join(statements), - lesson=self.context, + lesson=self.i_context, topic=self.topic, language=self.language, ) @@ -68,20 +68,23 @@ class WriteTeachingPlanPart(Action): return self.topic @staticmethod - def format_value(value): + def format_value(value, context: Context): """Fill parameters inside `value` with `options`.""" if not isinstance(value, str): return value if "{" not in value: return value - merged_opts = CONFIG.options or {} + options = context.config.model_dump() + for k, v in context.kwargs: + options[k] = v # None value is allowed to override and disable the value from config. + opts = {k: v for k, v in options.items() if v is not None} try: - return value.format(**merged_opts) + return value.format(**opts) except KeyError as e: logger.warning(f"Parameter is missing:{e}") - for k, v in merged_opts.items(): + for k, v in opts.items(): value = value.replace("{" + f"{k}" + "}", str(v)) return value diff --git a/metagpt/actions/write_test.py b/metagpt/actions/write_test.py index 96486311f..978fa20a6 100644 --- a/metagpt/actions/write_test.py +++ b/metagpt/actions/write_test.py @@ -39,7 +39,7 @@ you should correctly import the necessary classes based on these file locations! class WriteTest(Action): name: str = "WriteTest" - context: Optional[TestingContext] = None + i_context: Optional[TestingContext] = None async def write_code(self, prompt): code_rsp = await self._aask(prompt) @@ -55,16 +55,16 @@ class WriteTest(Action): return code async def run(self, *args, **kwargs) -> TestingContext: - if not self.context.test_doc: - self.context.test_doc = Document( - filename="test_" + self.context.code_doc.filename, root_path=TEST_CODES_FILE_REPO + if not self.i_context.test_doc: + self.i_context.test_doc = Document( + filename="test_" + self.i_context.code_doc.filename, root_path=TEST_CODES_FILE_REPO ) fake_root = "/data" prompt = PROMPT_TEMPLATE.format( - code_to_test=self.context.code_doc.content, - test_file_name=self.context.test_doc.filename, - source_file_path=fake_root + "/" + self.context.code_doc.root_relative_path, + code_to_test=self.i_context.code_doc.content, + test_file_name=self.i_context.test_doc.filename, + source_file_path=fake_root + "/" + self.i_context.code_doc.root_relative_path, workspace=fake_root, ) - self.context.test_doc.content = await self.write_code(prompt) - return self.context + self.i_context.test_doc.content = await self.write_code(prompt) + return self.i_context diff --git a/metagpt/config.py b/metagpt/config.py deleted file mode 100644 index 8cfbbe973..000000000 --- a/metagpt/config.py +++ /dev/null @@ -1,290 +0,0 @@ -#!/usr/bin/env python -# -*- coding: utf-8 -*- -""" -Provide configuration, singleton -@Modified By: mashenquan, 2023/11/27. - 1. According to Section 2.2.3.11 of RFC 135, add git repository support. - 2. Add the parameter `src_workspace` for the old version project path. -""" -import datetime -import json -import os -import warnings -from copy import deepcopy -from enum import Enum -from pathlib import Path -from typing import Any -from uuid import uuid4 - -import yaml - -from metagpt.const import DEFAULT_WORKSPACE_ROOT, METAGPT_ROOT, OPTIONS -from metagpt.logs import logger -from metagpt.tools import SearchEngineType, WebBrowserEngineType -from metagpt.utils.common import require_python_version -from metagpt.utils.cost_manager import CostManager -from metagpt.utils.singleton import Singleton - - -class NotConfiguredException(Exception): - """Exception raised for errors in the configuration. - - Attributes: - message -- explanation of the error - """ - - def __init__(self, message="The required configuration is not set"): - self.message = message - super().__init__(self.message) - - -class LLMProviderEnum(Enum): - OPENAI = "openai" - ANTHROPIC = "anthropic" - SPARK = "spark" - ZHIPUAI = "zhipuai" - FIREWORKS = "fireworks" - OPEN_LLM = "open_llm" - GEMINI = "gemini" - METAGPT = "metagpt" - AZURE_OPENAI = "azure_openai" - OLLAMA = "ollama" - - def __missing__(self, key): - return self.OPENAI - - -class Config(metaclass=Singleton): - """ - Regular usage method: - config = Config("config.yaml") - secret_key = config.get_key("MY_SECRET_KEY") - print("Secret key:", secret_key) - """ - - _instance = None - home_yaml_file = Path.home() / ".metagpt/config.yaml" - key_yaml_file = METAGPT_ROOT / "config/key.yaml" - default_yaml_file = METAGPT_ROOT / "config/config.yaml" - - def __init__(self, yaml_file=default_yaml_file, cost_data=""): - global_options = OPTIONS.get() - # cli paras - self.project_path = "" - self.project_name = "" - self.inc = False - self.reqa_file = "" - self.max_auto_summarize_code = 0 - self.git_reinit = False - - self._init_with_config_files_and_env(yaml_file) - # The agent needs to be billed per user, so billing information cannot be destroyed when the session ends. - self.cost_manager = CostManager(**json.loads(cost_data)) if cost_data else CostManager() - self._update() - global_options.update(OPTIONS.get()) - logger.debug("Config loading done.") - - def get_default_llm_provider_enum(self) -> LLMProviderEnum: - """Get first valid LLM provider enum""" - mappings = { - LLMProviderEnum.OPENAI: bool( - self._is_valid_llm_key(self.OPENAI_API_KEY) and not self.OPENAI_API_TYPE and self.OPENAI_API_MODEL - ), - LLMProviderEnum.ANTHROPIC: self._is_valid_llm_key(self.ANTHROPIC_API_KEY), - LLMProviderEnum.ZHIPUAI: self._is_valid_llm_key(self.ZHIPUAI_API_KEY), - LLMProviderEnum.FIREWORKS: self._is_valid_llm_key(self.FIREWORKS_API_KEY), - LLMProviderEnum.OPEN_LLM: self._is_valid_llm_key(self.OPEN_LLM_API_BASE), - LLMProviderEnum.GEMINI: self._is_valid_llm_key(self.GEMINI_API_KEY), - LLMProviderEnum.METAGPT: bool( - self._is_valid_llm_key(self.OPENAI_API_KEY) and self.OPENAI_API_TYPE == "metagpt" - ), - LLMProviderEnum.AZURE_OPENAI: bool( - self._is_valid_llm_key(self.OPENAI_API_KEY) - and self.OPENAI_API_TYPE == "azure" - and self.DEPLOYMENT_NAME - and self.OPENAI_API_VERSION - ), - LLMProviderEnum.OLLAMA: self._is_valid_llm_key(self.OLLAMA_API_BASE), - } - provider = None - for k, v in mappings.items(): - if v: - provider = k - break - if provider is None: - if self.DEFAULT_PROVIDER: - provider = LLMProviderEnum(self.DEFAULT_PROVIDER) - else: - raise NotConfiguredException("You should config a LLM configuration first") - - if provider is LLMProviderEnum.GEMINI and not require_python_version(req_version=(3, 10)): - warnings.warn("Use Gemini requires Python >= 3.10") - model_name = self.get_model_name(provider=provider) - if model_name: - logger.info(f"{provider} Model: {model_name}") - if provider: - logger.info(f"API: {provider}") - return provider - - def get_model_name(self, provider=None) -> str: - provider = provider or self.get_default_llm_provider_enum() - model_mappings = { - LLMProviderEnum.OPENAI: self.OPENAI_API_MODEL, - LLMProviderEnum.AZURE_OPENAI: self.DEPLOYMENT_NAME, - } - return model_mappings.get(provider, "") - - @staticmethod - def _is_valid_llm_key(k: str) -> bool: - return bool(k and k != "YOUR_API_KEY") - - def _update(self): - self.global_proxy = self._get("GLOBAL_PROXY") - - self.openai_api_key = self._get("OPENAI_API_KEY") - self.anthropic_api_key = self._get("ANTHROPIC_API_KEY") - self.zhipuai_api_key = self._get("ZHIPUAI_API_KEY") - self.open_llm_api_base = self._get("OPEN_LLM_API_BASE") - self.open_llm_api_model = self._get("OPEN_LLM_API_MODEL") - self.fireworks_api_key = self._get("FIREWORKS_API_KEY") - self.gemini_api_key = self._get("GEMINI_API_KEY") - self.ollama_api_base = self._get("OLLAMA_API_BASE") - self.ollama_api_model = self._get("OLLAMA_API_MODEL") - - if not self._get("DISABLE_LLM_PROVIDER_CHECK"): - _ = self.get_default_llm_provider_enum() - - self.openai_base_url = self._get("OPENAI_BASE_URL") - self.openai_proxy = self._get("OPENAI_PROXY") or self.global_proxy - self.openai_api_type = self._get("OPENAI_API_TYPE") - self.openai_api_version = self._get("OPENAI_API_VERSION") - self.openai_api_rpm = self._get("RPM", 3) - self.openai_api_model = self._get("OPENAI_API_MODEL", "gpt-4-1106-preview") - self.max_tokens_rsp = self._get("MAX_TOKENS", 2048) - self.deployment_name = self._get("DEPLOYMENT_NAME", "gpt-4") - - self.spark_appid = self._get("SPARK_APPID") - self.spark_api_secret = self._get("SPARK_API_SECRET") - self.spark_api_key = self._get("SPARK_API_KEY") - self.domain = self._get("DOMAIN") - self.spark_url = self._get("SPARK_URL") - - self.fireworks_api_base = self._get("FIREWORKS_API_BASE") - self.fireworks_api_model = self._get("FIREWORKS_API_MODEL") - - self.claude_api_key = self._get("ANTHROPIC_API_KEY") - self.serpapi_api_key = self._get("SERPAPI_API_KEY") - self.serper_api_key = self._get("SERPER_API_KEY") - self.google_api_key = self._get("GOOGLE_API_KEY") - self.google_cse_id = self._get("GOOGLE_CSE_ID") - self.search_engine = SearchEngineType(self._get("SEARCH_ENGINE", SearchEngineType.SERPAPI_GOOGLE)) - self.web_browser_engine = WebBrowserEngineType(self._get("WEB_BROWSER_ENGINE", WebBrowserEngineType.PLAYWRIGHT)) - self.playwright_browser_type = self._get("PLAYWRIGHT_BROWSER_TYPE", "chromium") - self.selenium_browser_type = self._get("SELENIUM_BROWSER_TYPE", "chrome") - - self.long_term_memory = self._get("LONG_TERM_MEMORY", False) - if self.long_term_memory: - logger.warning("LONG_TERM_MEMORY is True") - self.cost_manager.max_budget = self._get("MAX_BUDGET", 10.0) - self.code_review_k_times = 2 - - self.puppeteer_config = self._get("PUPPETEER_CONFIG", "") - self.mmdc = self._get("MMDC", "mmdc") - self.calc_usage = self._get("CALC_USAGE", True) - self.model_for_researcher_summary = self._get("MODEL_FOR_RESEARCHER_SUMMARY") - self.model_for_researcher_report = self._get("MODEL_FOR_RESEARCHER_REPORT") - self.mermaid_engine = self._get("MERMAID_ENGINE", "nodejs") - self.pyppeteer_executable_path = self._get("PYPPETEER_EXECUTABLE_PATH", "") - - workspace_uid = ( - self._get("WORKSPACE_UID") or f"{datetime.datetime.now().strftime('%Y%m%d%H%M%S')}-{uuid4().hex[-8:]}" - ) - self.repair_llm_output = self._get("REPAIR_LLM_OUTPUT", False) - self.prompt_schema = self._get("PROMPT_FORMAT", "json") - self.workspace_path = Path(self._get("WORKSPACE_PATH", DEFAULT_WORKSPACE_ROOT)) - val = self._get("WORKSPACE_PATH_WITH_UID") - if val and val.lower() == "true": # for agent - self.workspace_path = self.workspace_path / workspace_uid - self._ensure_workspace_exists() - self.max_auto_summarize_code = self.max_auto_summarize_code or self._get("MAX_AUTO_SUMMARIZE_CODE", 1) - self.timeout = int(self._get("TIMEOUT", 3)) - - self.kaggle_username = self._get("KAGGLE_USERNAME", "") - self.kaggle_key = self._get("KAGGLE_KEY", "") - - def update_via_cli(self, project_path, project_name, inc, reqa_file, max_auto_summarize_code): - """update config via cli""" - - # Use in the PrepareDocuments action according to Section 2.2.3.5.1 of RFC 135. - if project_path: - inc = True - project_name = project_name or Path(project_path).name - self.project_path = project_path - self.project_name = project_name - self.inc = inc - self.reqa_file = reqa_file - self.max_auto_summarize_code = max_auto_summarize_code - - def _ensure_workspace_exists(self): - self.workspace_path.mkdir(parents=True, exist_ok=True) - logger.debug(f"WORKSPACE_PATH set to {self.workspace_path}") - - def _init_with_config_files_and_env(self, yaml_file): - """Load from config/key.yaml, config/config.yaml, and env in decreasing order of priority""" - configs = dict(os.environ) - - for _yaml_file in [yaml_file, self.key_yaml_file, self.home_yaml_file]: - if not _yaml_file.exists(): - continue - - # Load local YAML file - with open(_yaml_file, "r", encoding="utf-8") as file: - yaml_data = yaml.safe_load(file) - if not yaml_data: - continue - configs.update(yaml_data) - OPTIONS.set(configs) - - @staticmethod - def _get(*args, **kwargs): - i = OPTIONS.get() - return i.get(*args, **kwargs) - - def get(self, key, *args, **kwargs): - """Retrieve values from config/key.yaml, config/config.yaml, and environment variables. - Throw an error if not found.""" - value = self._get(key, *args, **kwargs) - if value is None: - raise ValueError(f"Key '{key}' not found in environment variables or in the YAML file") - return value - - def __setattr__(self, name: str, value: Any) -> None: - OPTIONS.get()[name] = value - - def __getattr__(self, name: str) -> Any: - i = OPTIONS.get() - return i.get(name) - - def set_context(self, options: dict): - """Update current config""" - if not options: - return - opts = deepcopy(OPTIONS.get()) - opts.update(options) - OPTIONS.set(opts) - self._update() - - @property - def options(self): - """Return all key-values""" - return OPTIONS.get() - - def new_environ(self): - """Return a new os.environ object""" - env = os.environ.copy() - i = self.options - env.update({k: v for k, v in i.items() if isinstance(v, str)}) - return env - - -CONFIG = Config() diff --git a/metagpt/config2.py b/metagpt/config2.py new file mode 100644 index 000000000..dc53ee661 --- /dev/null +++ b/metagpt/config2.py @@ -0,0 +1,145 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/1/4 01:25 +@Author : alexanderwu +@File : config2.py +""" +import os +from pathlib import Path +from typing import Dict, Iterable, List, Literal, Optional + +from pydantic import BaseModel, model_validator + +from metagpt.configs.browser_config import BrowserConfig +from metagpt.configs.llm_config import LLMConfig, LLMType +from metagpt.configs.mermaid_config import MermaidConfig +from metagpt.configs.redis_config import RedisConfig +from metagpt.configs.s3_config import S3Config +from metagpt.configs.search_config import SearchConfig +from metagpt.configs.workspace_config import WorkspaceConfig +from metagpt.const import CONFIG_ROOT, METAGPT_ROOT +from metagpt.utils.yaml_model import YamlModel + + +class CLIParams(BaseModel): + """CLI parameters""" + + project_path: str = "" + project_name: str = "" + inc: bool = False + reqa_file: str = "" + max_auto_summarize_code: int = 0 + git_reinit: bool = False + + @model_validator(mode="after") + def check_project_path(self): + """Check project_path and project_name""" + if self.project_path: + self.inc = True + self.project_name = self.project_name or Path(self.project_path).name + return self + + +class Config(CLIParams, YamlModel): + """Configurations for MetaGPT""" + + # Key Parameters + llm: LLMConfig + + # Global Proxy. Will be used if llm.proxy is not set + proxy: str = "" + + # Tool Parameters + search: Optional[SearchConfig] = None + browser: BrowserConfig = BrowserConfig() + mermaid: MermaidConfig = MermaidConfig() + + # Storage Parameters + s3: Optional[S3Config] = None + redis: Optional[RedisConfig] = None + + # Misc Parameters + repair_llm_output: bool = False + prompt_schema: Literal["json", "markdown", "raw"] = "json" + workspace: WorkspaceConfig = WorkspaceConfig() + enable_longterm_memory: bool = False + code_review_k_times: int = 2 + + # Will be removed in the future + llm_for_researcher_summary: str = "gpt3" + llm_for_researcher_report: str = "gpt3" + METAGPT_TEXT_TO_IMAGE_MODEL_URL: str = "" + language: str = "English" + redis_key: str = "placeholder" + mmdc: str = "mmdc" + puppeteer_config: str = "" + pyppeteer_executable_path: str = "" + IFLYTEK_APP_ID: str = "" + IFLYTEK_API_SECRET: str = "" + IFLYTEK_API_KEY: str = "" + AZURE_TTS_SUBSCRIPTION_KEY: str = "" + AZURE_TTS_REGION: str = "" + mermaid_engine: str = "nodejs" + OPENAI_VISION_MODEL: str = "gpt-4-vision-preview" + VISION_MAX_TOKENS: int = 4096 + + @classmethod + def from_home(cls, path): + """Load config from ~/.metagpt/config.yaml""" + pathname = CONFIG_ROOT / path + if not pathname.exists(): + return None + return Config.from_yaml_file(pathname) + + @classmethod + def default(cls): + """Load default config + - Priority: env < default_config_paths + - Inside default_config_paths, the latter one overwrites the former one + """ + default_config_paths: List[Path] = [ + METAGPT_ROOT / "config/config2.yaml", + Path.home() / ".metagpt/config2.yaml", + ] + + dicts = [dict(os.environ)] + dicts += [Config.read_yaml(path) for path in default_config_paths] + final = merge_dict(dicts) + return Config(**final) + + def update_via_cli(self, project_path, project_name, inc, reqa_file, max_auto_summarize_code): + """update config via cli""" + + # Use in the PrepareDocuments action according to Section 2.2.3.5.1 of RFC 135. + if project_path: + inc = True + project_name = project_name or Path(project_path).name + self.project_path = project_path + self.project_name = project_name + self.inc = inc + self.reqa_file = reqa_file + self.max_auto_summarize_code = max_auto_summarize_code + + def get_openai_llm(self) -> Optional[LLMConfig]: + """Get OpenAI LLMConfig by name. If no OpenAI, raise Exception""" + if self.llm.api_type == LLMType.OPENAI: + return self.llm + return None + + def get_azure_llm(self) -> Optional[LLMConfig]: + """Get Azure LLMConfig by name. If no Azure, raise Exception""" + if self.llm.api_type == LLMType.AZURE: + return self.llm + return None + + +def merge_dict(dicts: Iterable[Dict]) -> Dict: + """Merge multiple dicts into one, with the latter dict overwriting the former""" + result = {} + for dictionary in dicts: + result.update(dictionary) + return result + + +config = Config.default() diff --git a/tests/metagpt/test_action.py b/metagpt/configs/__init__.py similarity index 59% rename from tests/metagpt/test_action.py rename to metagpt/configs/__init__.py index af5106ab4..e42e6788f 100644 --- a/tests/metagpt/test_action.py +++ b/metagpt/configs/__init__.py @@ -1,7 +1,7 @@ #!/usr/bin/env python # -*- coding: utf-8 -*- """ -@Time : 2023/5/11 14:44 +@Time : 2024/1/4 16:33 @Author : alexanderwu -@File : test_action.py +@File : __init__.py """ diff --git a/metagpt/configs/browser_config.py b/metagpt/configs/browser_config.py new file mode 100644 index 000000000..00f918735 --- /dev/null +++ b/metagpt/configs/browser_config.py @@ -0,0 +1,20 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/1/4 19:06 +@Author : alexanderwu +@File : browser_config.py +""" +from typing import Literal + +from metagpt.tools import WebBrowserEngineType +from metagpt.utils.yaml_model import YamlModel + + +class BrowserConfig(YamlModel): + """Config for Browser""" + + engine: WebBrowserEngineType = WebBrowserEngineType.PLAYWRIGHT + browser: Literal["chrome", "firefox", "edge", "ie"] = "chrome" + driver: Literal["chromium", "firefox", "webkit"] = "chromium" + path: str = "" diff --git a/metagpt/configs/llm_config.py b/metagpt/configs/llm_config.py new file mode 100644 index 000000000..626d4242f --- /dev/null +++ b/metagpt/configs/llm_config.py @@ -0,0 +1,78 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/1/4 16:33 +@Author : alexanderwu +@File : llm_config.py +""" +from enum import Enum +from typing import Optional + +from pydantic import field_validator + +from metagpt.utils.yaml_model import YamlModel + + +class LLMType(Enum): + OPENAI = "openai" + ANTHROPIC = "anthropic" + SPARK = "spark" + ZHIPUAI = "zhipuai" + FIREWORKS = "fireworks" + OPEN_LLM = "open_llm" + GEMINI = "gemini" + METAGPT = "metagpt" + AZURE = "azure" + OLLAMA = "ollama" + + def __missing__(self, key): + return self.OPENAI + + +class LLMConfig(YamlModel): + """Config for LLM + + OpenAI: https://github.com/openai/openai-python/blob/main/src/openai/resources/chat/completions.py#L681 + Optional Fields in pydantic: https://docs.pydantic.dev/latest/migration/#required-optional-and-nullable-fields + """ + + api_key: str + api_type: LLMType = LLMType.OPENAI + base_url: str = "https://api.openai.com/v1" + api_version: Optional[str] = None + + model: Optional[str] = None # also stands for DEPLOYMENT_NAME + + # For Spark(Xunfei), maybe remove later + app_id: Optional[str] = None + api_secret: Optional[str] = None + domain: Optional[str] = None + + # For Chat Completion + max_token: int = 4096 + temperature: float = 0.0 + top_p: float = 1.0 + top_k: int = 0 + repetition_penalty: float = 1.0 + stop: Optional[str] = None + presence_penalty: float = 0.0 + frequency_penalty: float = 0.0 + best_of: Optional[int] = None + n: Optional[int] = None + stream: bool = False + logprobs: Optional[bool] = None # https://cookbook.openai.com/examples/using_logprobs + top_logprobs: Optional[int] = None + timeout: int = 60 + + # For Network + proxy: Optional[str] = None + + # Cost Control + calc_usage: bool = True + + @field_validator("api_key") + @classmethod + def check_llm_key(cls, v): + if v in ["", None, "YOUR_API_KEY"]: + raise ValueError("Please set your API key in config.yaml") + return v diff --git a/metagpt/configs/mermaid_config.py b/metagpt/configs/mermaid_config.py new file mode 100644 index 000000000..de4a3865c --- /dev/null +++ b/metagpt/configs/mermaid_config.py @@ -0,0 +1,18 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/1/4 19:07 +@Author : alexanderwu +@File : mermaid_config.py +""" +from typing import Literal + +from metagpt.utils.yaml_model import YamlModel + + +class MermaidConfig(YamlModel): + """Config for Mermaid""" + + engine: Literal["nodejs", "ink", "playwright", "pyppeteer"] = "nodejs" + path: str = "" + puppeteer_config: str = "" # Only for nodejs engine diff --git a/metagpt/configs/redis_config.py b/metagpt/configs/redis_config.py new file mode 100644 index 000000000..c4cfb6764 --- /dev/null +++ b/metagpt/configs/redis_config.py @@ -0,0 +1,26 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/1/4 19:06 +@Author : alexanderwu +@File : redis_config.py +""" +from metagpt.utils.yaml_model import YamlModelWithoutDefault + + +class RedisConfig(YamlModelWithoutDefault): + host: str + port: int + username: str = "" + password: str + db: str + + def to_url(self): + return f"redis://{self.host}:{self.port}" + + def to_kwargs(self): + return { + "username": self.username, + "password": self.password, + "db": self.db, + } diff --git a/metagpt/configs/s3_config.py b/metagpt/configs/s3_config.py new file mode 100644 index 000000000..72b81fae4 --- /dev/null +++ b/metagpt/configs/s3_config.py @@ -0,0 +1,15 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/1/4 19:07 +@Author : alexanderwu +@File : s3_config.py +""" +from metagpt.utils.yaml_model import YamlModelWithoutDefault + + +class S3Config(YamlModelWithoutDefault): + access_key: str + secret_key: str + endpoint: str + bucket: str diff --git a/metagpt/configs/search_config.py b/metagpt/configs/search_config.py new file mode 100644 index 000000000..a8ae918db --- /dev/null +++ b/metagpt/configs/search_config.py @@ -0,0 +1,17 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/1/4 19:06 +@Author : alexanderwu +@File : search_config.py +""" +from metagpt.tools import SearchEngineType +from metagpt.utils.yaml_model import YamlModel + + +class SearchConfig(YamlModel): + """Config for Search""" + + api_key: str + api_type: SearchEngineType = SearchEngineType.SERPAPI_GOOGLE + cse_id: str = "" # for google diff --git a/metagpt/configs/workspace_config.py b/metagpt/configs/workspace_config.py new file mode 100644 index 000000000..df7aeaef9 --- /dev/null +++ b/metagpt/configs/workspace_config.py @@ -0,0 +1,38 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/1/4 19:09 +@Author : alexanderwu +@File : workspace_config.py +""" +from datetime import datetime +from pathlib import Path +from uuid import uuid4 + +from pydantic import field_validator, model_validator + +from metagpt.const import DEFAULT_WORKSPACE_ROOT +from metagpt.utils.yaml_model import YamlModel + + +class WorkspaceConfig(YamlModel): + path: Path = DEFAULT_WORKSPACE_ROOT + use_uid: bool = False + uid: str = "" + + @field_validator("path") + @classmethod + def check_workspace_path(cls, v): + if isinstance(v, str): + v = Path(v) + return v + + @model_validator(mode="after") + def check_uid_and_update_path(self): + if self.use_uid and not self.uid: + self.uid = f"{datetime.now().strftime('%Y%m%d%H%M%S')}-{uuid4().hex[-8:]}" + self.path = self.path / self.uid + + # Create workspace path if not exists + self.path.mkdir(parents=True, exist_ok=True) + return self diff --git a/metagpt/const.py b/metagpt/const.py index 7a19e81d0..2cffaa804 100644 --- a/metagpt/const.py +++ b/metagpt/const.py @@ -9,7 +9,6 @@ @Modified By: mashenquan, 2023-11-27. Defines file repository paths according to Section 2.2.3.4 of RFC 135. @Modified By: mashenquan, 2023/12/5. Add directories for code summarization.. """ -import contextvars import os from pathlib import Path @@ -17,8 +16,6 @@ from loguru import logger import metagpt -OPTIONS = contextvars.ContextVar("OPTIONS", default={}) - def get_metagpt_package_root(): """Get the root directory of the installed package.""" @@ -47,7 +44,7 @@ def get_metagpt_root(): # METAGPT PROJECT ROOT AND VARS - +CONFIG_ROOT = Path.home() / ".metagpt" METAGPT_ROOT = get_metagpt_root() # Dependent on METAGPT_PROJECT_ROOT DEFAULT_WORKSPACE_ROOT = METAGPT_ROOT / "workspace" @@ -73,12 +70,10 @@ SKILL_DIRECTORY = SOURCE_ROOT / "skills" TOOL_SCHEMA_PATH = METAGPT_ROOT / "metagpt/tools/schemas" TOOL_LIBS_PATH = METAGPT_ROOT / "metagpt/tools/libs" - # REAL CONSTS MEM_TTL = 24 * 30 * 3600 - MESSAGE_ROUTE_FROM = "sent_from" MESSAGE_ROUTE_TO = "send_to" MESSAGE_ROUTE_CAUSE_BY = "cause_by" @@ -89,25 +84,28 @@ MESSAGE_ROUTE_TO_NONE = "" REQUIREMENT_FILENAME = "requirement.txt" BUGFIX_FILENAME = "bugfix.txt" PACKAGE_REQUIREMENTS_FILENAME = "requirements.txt" +CODE_PLAN_AND_CHANGE_FILENAME = "code_plan_and_change.json" DOCS_FILE_REPO = "docs" -PRDS_FILE_REPO = "docs/prds" +PRDS_FILE_REPO = "docs/prd" SYSTEM_DESIGN_FILE_REPO = "docs/system_design" -TASK_FILE_REPO = "docs/tasks" +TASK_FILE_REPO = "docs/task" +CODE_PLAN_AND_CHANGE_FILE_REPO = "docs/code_plan_and_change" COMPETITIVE_ANALYSIS_FILE_REPO = "resources/competitive_analysis" DATA_API_DESIGN_FILE_REPO = "resources/data_api_design" SEQ_FLOW_FILE_REPO = "resources/seq_flow" SYSTEM_DESIGN_PDF_FILE_REPO = "resources/system_design" PRD_PDF_FILE_REPO = "resources/prd" -TASK_PDF_FILE_REPO = "resources/api_spec_and_tasks" +TASK_PDF_FILE_REPO = "resources/api_spec_and_task" +CODE_PLAN_AND_CHANGE_PDF_FILE_REPO = "resources/code_plan_and_change" TEST_CODES_FILE_REPO = "tests" TEST_OUTPUTS_FILE_REPO = "test_outputs" -CODE_SUMMARIES_FILE_REPO = "docs/code_summaries" -CODE_SUMMARIES_PDF_FILE_REPO = "resources/code_summaries" +CODE_SUMMARIES_FILE_REPO = "docs/code_summary" +CODE_SUMMARIES_PDF_FILE_REPO = "resources/code_summary" RESOURCES_FILE_REPO = "resources" -SD_OUTPUT_FILE_REPO = "resources/SD_Output" +SD_OUTPUT_FILE_REPO = "resources/sd_output" GRAPH_REPO_FILE_REPO = "docs/graph_repo" -CLASS_VIEW_FILE_REPO = "docs/class_views" +CLASS_VIEW_FILE_REPO = "docs/class_view" YAPI_URL = "http://yapi.deepwisdomai.com/" diff --git a/metagpt/context.py b/metagpt/context.py new file mode 100644 index 000000000..3dfd52d58 --- /dev/null +++ b/metagpt/context.py @@ -0,0 +1,97 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/1/4 16:32 +@Author : alexanderwu +@File : context.py +""" +import os +from pathlib import Path +from typing import Any, Optional + +from pydantic import BaseModel, ConfigDict + +from metagpt.config2 import Config +from metagpt.configs.llm_config import LLMConfig +from metagpt.provider.base_llm import BaseLLM +from metagpt.provider.llm_provider_registry import create_llm_instance +from metagpt.utils.cost_manager import CostManager +from metagpt.utils.git_repository import GitRepository +from metagpt.utils.project_repo import ProjectRepo + + +class AttrDict(BaseModel): + """A dict-like object that allows access to keys as attributes, compatible with Pydantic.""" + + model_config = ConfigDict(extra="allow") + + def __init__(self, **kwargs): + super().__init__(**kwargs) + self.__dict__.update(kwargs) + + def __getattr__(self, key): + return self.__dict__.get(key, None) + + def __setattr__(self, key, value): + self.__dict__[key] = value + + def __delattr__(self, key): + if key in self.__dict__: + del self.__dict__[key] + else: + raise AttributeError(f"No such attribute: {key}") + + def set(self, key, val: Any): + self.__dict__[key] = val + + def get(self, key, default: Any = None): + return self.__dict__.get(key, default) + + def remove(self, key): + if key in self.__dict__: + self.__delattr__(key) + + +class Context(BaseModel): + """Env context for MetaGPT""" + + model_config = ConfigDict(arbitrary_types_allowed=True) + + kwargs: AttrDict = AttrDict() + config: Config = Config.default() + + repo: Optional[ProjectRepo] = None + git_repo: Optional[GitRepository] = None + src_workspace: Optional[Path] = None + cost_manager: CostManager = CostManager() + + _llm: Optional[BaseLLM] = None + + def new_environ(self): + """Return a new os.environ object""" + env = os.environ.copy() + # i = self.options + # env.update({k: v for k, v in i.items() if isinstance(v, str)}) + return env + + # def use_llm(self, name: Optional[str] = None, provider: LLMType = LLMType.OPENAI) -> BaseLLM: + # """Use a LLM instance""" + # self._llm_config = self.config.get_llm_config(name, provider) + # self._llm = None + # return self._llm + + def llm(self) -> BaseLLM: + """Return a LLM instance, fixme: support cache""" + # if self._llm is None: + self._llm = create_llm_instance(self.config.llm) + if self._llm.cost_manager is None: + self._llm.cost_manager = self.cost_manager + return self._llm + + def llm_with_cost_manager_from_llm_config(self, llm_config: LLMConfig) -> BaseLLM: + """Return a LLM instance, fixme: support cache""" + # if self._llm is None: + llm = create_llm_instance(llm_config) + if llm.cost_manager is None: + llm.cost_manager = self.cost_manager + return llm diff --git a/metagpt/context_mixin.py b/metagpt/context_mixin.py new file mode 100644 index 000000000..bdf2d0734 --- /dev/null +++ b/metagpt/context_mixin.py @@ -0,0 +1,102 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/1/11 17:25 +@Author : alexanderwu +@File : context_mixin.py +""" +from typing import Optional + +from pydantic import BaseModel, ConfigDict, Field + +from metagpt.config2 import Config +from metagpt.context import Context +from metagpt.provider.base_llm import BaseLLM + + +class ContextMixin(BaseModel): + """Mixin class for context and config""" + + model_config = ConfigDict(arbitrary_types_allowed=True) + + # Pydantic has bug on _private_attr when using inheritance, so we use private_* instead + # - https://github.com/pydantic/pydantic/issues/7142 + # - https://github.com/pydantic/pydantic/issues/7083 + # - https://github.com/pydantic/pydantic/issues/7091 + + # Env/Role/Action will use this context as private context, or use self.context as public context + private_context: Optional[Context] = Field(default=None, exclude=True) + # Env/Role/Action will use this config as private config, or use self.context.config as public config + private_config: Optional[Config] = Field(default=None, exclude=True) + + # Env/Role/Action will use this llm as private llm, or use self.context._llm instance + private_llm: Optional[BaseLLM] = Field(default=None, exclude=True) + + def __init__( + self, + context: Optional[Context] = None, + config: Optional[Config] = None, + llm: Optional[BaseLLM] = None, + **kwargs, + ): + """Initialize with config""" + super().__init__(**kwargs) + self.set_context(context) + self.set_config(config) + self.set_llm(llm) + + def set(self, k, v, override=False): + """Set attribute""" + if override or not self.__dict__.get(k): + self.__dict__[k] = v + + def set_context(self, context: Context, override=True): + """Set context""" + self.set("private_context", context, override) + + def set_config(self, config: Config, override=False): + """Set config""" + self.set("private_config", config, override) + if config is not None: + _ = self.llm # init llm + + def set_llm(self, llm: BaseLLM, override=False): + """Set llm""" + self.set("private_llm", llm, override) + + @property + def config(self) -> Config: + """Role config: role config > context config""" + if self.private_config: + return self.private_config + return self.context.config + + @config.setter + def config(self, config: Config) -> None: + """Set config""" + self.set_config(config) + + @property + def context(self) -> Context: + """Role context: role context > context""" + if self.private_context: + return self.private_context + return Context() + + @context.setter + def context(self, context: Context) -> None: + """Set context""" + self.set_context(context) + + @property + def llm(self) -> BaseLLM: + """Role llm: if not existed, init from role.config""" + # print(f"class:{self.__class__.__name__}({self.name}), llm: {self._llm}, llm_config: {self._llm_config}") + if not self.private_llm: + self.private_llm = self.context.llm_with_cost_manager_from_llm_config(self.config.llm) + return self.private_llm + + @llm.setter + def llm(self, llm: BaseLLM) -> None: + """Set llm""" + self.private_llm = llm diff --git a/metagpt/document_store/base_store.py b/metagpt/document_store/base_store.py index b719d1083..ddc1d626b 100644 --- a/metagpt/document_store/base_store.py +++ b/metagpt/document_store/base_store.py @@ -8,8 +8,6 @@ from abc import ABC, abstractmethod from pathlib import Path -from metagpt.config import Config - class BaseStore(ABC): """FIXME: consider add_index, set_index and think about granularity.""" @@ -31,7 +29,6 @@ class LocalStore(BaseStore, ABC): def __init__(self, raw_data_path: Path, cache_dir: Path = None): if not raw_data_path: raise FileNotFoundError - self.config = Config() self.raw_data_path = raw_data_path self.fname = self.raw_data_path.stem if not cache_dir: diff --git a/metagpt/document_store/faiss_store.py b/metagpt/document_store/faiss_store.py index 1271f1c23..2359917d5 100644 --- a/metagpt/document_store/faiss_store.py +++ b/metagpt/document_store/faiss_store.py @@ -9,14 +9,13 @@ import asyncio from pathlib import Path from typing import Optional -from langchain.embeddings import OpenAIEmbeddings from langchain.vectorstores import FAISS from langchain_core.embeddings import Embeddings -from metagpt.config import CONFIG from metagpt.document import IndexableDocument from metagpt.document_store.base_store import LocalStore from metagpt.logs import logger +from metagpt.utils.embedding import get_embedding class FaissStore(LocalStore): @@ -25,9 +24,7 @@ class FaissStore(LocalStore): ): self.meta_col = meta_col self.content_col = content_col - self.embedding = embedding or OpenAIEmbeddings( - openai_api_key=CONFIG.openai_api_key, openai_api_base=CONFIG.openai_base_url - ) + self.embedding = embedding or get_embedding() super().__init__(raw_data, cache_dir) def _load(self) -> Optional["FaissStore"]: diff --git a/metagpt/environment.py b/metagpt/environment.py index ddb9ad9dd..5a2dd339b 100644 --- a/metagpt/environment.py +++ b/metagpt/environment.py @@ -12,16 +12,15 @@ functionality is to be consolidated into the `Environment` class. """ import asyncio -from pathlib import Path from typing import Iterable, Set from pydantic import BaseModel, ConfigDict, Field, SerializeAsAny, model_validator -from metagpt.config import CONFIG +from metagpt.context import Context from metagpt.logs import logger from metagpt.roles.role import Role from metagpt.schema import Message -from metagpt.utils.common import is_subscribed, read_json_file, write_json_file +from metagpt.utils.common import is_send_to class Environment(BaseModel): @@ -33,58 +32,22 @@ class Environment(BaseModel): desc: str = Field(default="") # 环境描述 roles: dict[str, SerializeAsAny[Role]] = Field(default_factory=dict, validate_default=True) - members: dict[Role, Set] = Field(default_factory=dict, exclude=True) + member_addrs: dict[Role, Set] = Field(default_factory=dict, exclude=True) history: str = "" # For debug + context: Context = Field(default_factory=Context, exclude=True) @model_validator(mode="after") def init_roles(self): self.add_roles(self.roles.values()) return self - def serialize(self, stg_path: Path): - roles_path = stg_path.joinpath("roles.json") - roles_info = [] - for role_key, role in self.roles.items(): - roles_info.append( - { - "role_class": role.__class__.__name__, - "module_name": role.__module__, - "role_name": role.name, - "role_sub_tags": list(self.members.get(role)), - } - ) - role.serialize(stg_path=stg_path.joinpath(f"roles/{role.__class__.__name__}_{role.name}")) - write_json_file(roles_path, roles_info) - - history_path = stg_path.joinpath("history.json") - write_json_file(history_path, {"content": self.history}) - - @classmethod - def deserialize(cls, stg_path: Path) -> "Environment": - """stg_path: ./storage/team/environment/""" - roles_path = stg_path.joinpath("roles.json") - roles_info = read_json_file(roles_path) - roles = [] - for role_info in roles_info: - # role stored in ./environment/roles/{role_class}_{role_name} - role_path = stg_path.joinpath(f"roles/{role_info.get('role_class')}_{role_info.get('role_name')}") - role = Role.deserialize(role_path) - roles.append(role) - - history = read_json_file(stg_path.joinpath("history.json")) - history = history.get("content") - - environment = Environment(**{"history": history}) - environment.add_roles(roles) - - return environment - def add_role(self, role: Role): """增加一个在当前环境的角色 Add a role in the current environment """ self.roles[role.profile] = role role.set_env(self) + role.context = self.context def add_roles(self, roles: Iterable[Role]): """增加一批在当前环境的角色 @@ -95,6 +58,7 @@ class Environment(BaseModel): for role in roles: # setup system message with roles role.set_env(self) + role.context = self.context def publish_message(self, message: Message, peekable: bool = True) -> bool: """ @@ -108,8 +72,8 @@ class Environment(BaseModel): logger.debug(f"publish_message: {message.dump()}") found = False # According to the routing feature plan in Chapter 2.2.3.2 of RFC 113 - for role, subscription in self.members.items(): - if is_subscribed(message, subscription): + for role, addrs in self.member_addrs.items(): + if is_send_to(message, addrs): role.put_message(message) found = True if not found: @@ -154,15 +118,14 @@ class Environment(BaseModel): return False return True - def get_subscription(self, obj): - """Get the labels for messages to be consumed by the object.""" - return self.members.get(obj, {}) + def get_addresses(self, obj): + """Get the addresses of the object.""" + return self.member_addrs.get(obj, {}) - def set_subscription(self, obj, tags): - """Set the labels for message to be consumed by the object""" - self.members[obj] = tags + def set_addresses(self, obj, addresses): + """Set the addresses of the object""" + self.member_addrs[obj] = addresses - @staticmethod - def archive(auto_archive=True): - if auto_archive and CONFIG.git_repo: - CONFIG.git_repo.archive() + def archive(self, auto_archive=True): + if auto_archive and self.context.git_repo: + self.context.git_repo.archive() diff --git a/metagpt/learn/skill_loader.py b/metagpt/learn/skill_loader.py index 7383af66d..bcf28bb87 100644 --- a/metagpt/learn/skill_loader.py +++ b/metagpt/learn/skill_loader.py @@ -13,7 +13,7 @@ import aiofiles import yaml from pydantic import BaseModel, Field -from metagpt.config import CONFIG +from metagpt.context import Context class Example(BaseModel): @@ -73,14 +73,15 @@ class SkillsDeclaration(BaseModel): skill_data = yaml.safe_load(data) return SkillsDeclaration(**skill_data) - def get_skill_list(self, entity_name: str = "Assistant") -> Dict: + def get_skill_list(self, entity_name: str = "Assistant", context: Context = None) -> Dict: """Return the skill name based on the skill description.""" entity = self.entities.get(entity_name) if not entity: return {} # List of skills that the agent chooses to activate. - agent_skills = CONFIG.agent_skills + ctx = context or Context() + agent_skills = ctx.kwargs.agent_skills if not agent_skills: return {} diff --git a/metagpt/learn/text_to_embedding.py b/metagpt/learn/text_to_embedding.py index 26dab0419..f859ab638 100644 --- a/metagpt/learn/text_to_embedding.py +++ b/metagpt/learn/text_to_embedding.py @@ -6,19 +6,19 @@ @File : text_to_embedding.py @Desc : Text-to-Embedding skill, which provides text-to-embedding functionality. """ - -from metagpt.config import CONFIG +import metagpt.config2 +from metagpt.config2 import Config from metagpt.tools.openai_text_to_embedding import oas3_openai_text_to_embedding -async def text_to_embedding(text, model="text-embedding-ada-002", openai_api_key="", **kwargs): +async def text_to_embedding(text, model="text-embedding-ada-002", config: Config = metagpt.config2.config): """Text to embedding :param text: The text used for embedding. :param model: One of ['text-embedding-ada-002'], ID of the model to use. For more details, checkout: `https://api.openai.com/v1/models`. - :param openai_api_key: OpenAI API key, For more details, checkout: `https://platform.openai.com/account/api-keys` + :param config: OpenAI config with API key, For more details, checkout: `https://platform.openai.com/account/api-keys` :return: A json object of :class:`ResultEmbedding` class if successful, otherwise `{}`. """ - if CONFIG.OPENAI_API_KEY or openai_api_key: - return await oas3_openai_text_to_embedding(text, model=model, openai_api_key=openai_api_key) - raise EnvironmentError + openai_api_key = config.get_openai_llm().api_key + proxy = config.get_openai_llm().proxy + return await oas3_openai_text_to_embedding(text, model=model, openai_api_key=openai_api_key, proxy=proxy) diff --git a/metagpt/learn/text_to_image.py b/metagpt/learn/text_to_image.py index c3c62fb67..e2fac7647 100644 --- a/metagpt/learn/text_to_image.py +++ b/metagpt/learn/text_to_image.py @@ -8,33 +8,37 @@ """ import base64 -from metagpt.config import CONFIG +import metagpt.config2 +from metagpt.config2 import Config from metagpt.const import BASE64_FORMAT +from metagpt.llm import LLM from metagpt.tools.metagpt_text_to_image import oas3_metagpt_text_to_image from metagpt.tools.openai_text_to_image import oas3_openai_text_to_image from metagpt.utils.s3 import S3 -async def text_to_image(text, size_type: str = "512x512", openai_api_key="", model_url="", **kwargs): +async def text_to_image(text, size_type: str = "512x512", config: Config = metagpt.config2.config): """Text to image :param text: The text used for image conversion. - :param openai_api_key: OpenAI API key, For more details, checkout: `https://platform.openai.com/account/api-keys` :param size_type: If using OPENAI, the available size options are ['256x256', '512x512', '1024x1024'], while for MetaGPT, the options are ['512x512', '512x768']. - :param model_url: MetaGPT model url + :param config: Config :return: The image data is returned in Base64 encoding. """ image_declaration = "data:image/png;base64," - if CONFIG.METAGPT_TEXT_TO_IMAGE_MODEL_URL or model_url: + + model_url = config.METAGPT_TEXT_TO_IMAGE_MODEL_URL + if model_url: binary_data = await oas3_metagpt_text_to_image(text, size_type, model_url) - elif CONFIG.OPENAI_API_KEY or openai_api_key: - binary_data = await oas3_openai_text_to_image(text, size_type) + elif config.get_openai_llm(): + llm = LLM(llm_config=config.get_openai_llm()) + binary_data = await oas3_openai_text_to_image(text, size_type, llm=llm) else: raise ValueError("Missing necessary parameters.") base64_data = base64.b64encode(binary_data).decode("utf-8") - s3 = S3() - url = await s3.cache(data=base64_data, file_ext=".png", format=BASE64_FORMAT) if s3.is_valid else "" + s3 = S3(config.s3) + url = await s3.cache(data=base64_data, file_ext=".png", format=BASE64_FORMAT) if url: return f"![{text}]({url})" return image_declaration + base64_data if base64_data else "" diff --git a/metagpt/learn/text_to_speech.py b/metagpt/learn/text_to_speech.py index ecd00c724..37e56eaff 100644 --- a/metagpt/learn/text_to_speech.py +++ b/metagpt/learn/text_to_speech.py @@ -6,8 +6,8 @@ @File : text_to_speech.py @Desc : Text-to-Speech skill, which provides text-to-speech functionality """ - -from metagpt.config import CONFIG +import metagpt.config2 +from metagpt.config2 import Config from metagpt.const import BASE64_FORMAT from metagpt.tools.azure_tts import oas3_azsure_tts from metagpt.tools.iflytek_tts import oas3_iflytek_tts @@ -20,12 +20,7 @@ async def text_to_speech( voice="zh-CN-XiaomoNeural", style="affectionate", role="Girl", - subscription_key="", - region="", - iflytek_app_id="", - iflytek_api_key="", - iflytek_api_secret="", - **kwargs, + config: Config = metagpt.config2.config, ): """Text to speech For more details, check out:`https://learn.microsoft.com/en-us/azure/ai-services/speech-service/language-support?tabs=tts` @@ -44,23 +39,27 @@ async def text_to_speech( """ - if (CONFIG.AZURE_TTS_SUBSCRIPTION_KEY and CONFIG.AZURE_TTS_REGION) or (subscription_key and region): + subscription_key = config.AZURE_TTS_SUBSCRIPTION_KEY + region = config.AZURE_TTS_REGION + if subscription_key and region: audio_declaration = "data:audio/wav;base64," base64_data = await oas3_azsure_tts(text, lang, voice, style, role, subscription_key, region) - s3 = S3() - url = await s3.cache(data=base64_data, file_ext=".wav", format=BASE64_FORMAT) if s3.is_valid else "" + s3 = S3(config.s3) + url = await s3.cache(data=base64_data, file_ext=".wav", format=BASE64_FORMAT) if url: return f"[{text}]({url})" return audio_declaration + base64_data if base64_data else base64_data - if (CONFIG.IFLYTEK_APP_ID and CONFIG.IFLYTEK_API_KEY and CONFIG.IFLYTEK_API_SECRET) or ( - iflytek_app_id and iflytek_api_key and iflytek_api_secret - ): + + iflytek_app_id = config.IFLYTEK_APP_ID + iflytek_api_key = config.IFLYTEK_API_KEY + iflytek_api_secret = config.IFLYTEK_API_SECRET + if iflytek_app_id and iflytek_api_key and iflytek_api_secret: audio_declaration = "data:audio/mp3;base64," base64_data = await oas3_iflytek_tts( text=text, app_id=iflytek_app_id, api_key=iflytek_api_key, api_secret=iflytek_api_secret ) - s3 = S3() - url = await s3.cache(data=base64_data, file_ext=".mp3", format=BASE64_FORMAT) if s3.is_valid else "" + s3 = S3(config.s3) + url = await s3.cache(data=base64_data, file_ext=".mp3", format=BASE64_FORMAT) if url: return f"[{text}]({url})" return audio_declaration + base64_data if base64_data else base64_data diff --git a/metagpt/llm.py b/metagpt/llm.py index 76dd5a0f8..a3fc5613a 100644 --- a/metagpt/llm.py +++ b/metagpt/llm.py @@ -5,20 +5,16 @@ @Author : alexanderwu @File : llm.py """ - from typing import Optional -from metagpt.config import CONFIG, LLMProviderEnum +from metagpt.configs.llm_config import LLMConfig +from metagpt.context import Context from metagpt.provider.base_llm import BaseLLM -from metagpt.provider.human_provider import HumanProvider -from metagpt.provider.llm_provider_registry import LLM_REGISTRY - -_ = HumanProvider() # Avoid pre-commit error -def LLM(provider: Optional[LLMProviderEnum] = None) -> BaseLLM: - """get the default llm provider""" - if provider is None: - provider = CONFIG.get_default_llm_provider_enum() - - return LLM_REGISTRY.get_provider(provider) +def LLM(llm_config: Optional[LLMConfig] = None, context: Context = None) -> BaseLLM: + """get the default llm provider if name is None""" + ctx = context or Context() + if llm_config is not None: + ctx.llm_with_cost_manager_from_llm_config(llm_config) + return ctx.llm() diff --git a/metagpt/memory/brain_memory.py b/metagpt/memory/brain_memory.py index ff29eaddb..044b0b359 100644 --- a/metagpt/memory/brain_memory.py +++ b/metagpt/memory/brain_memory.py @@ -14,8 +14,8 @@ from typing import Dict, List, Optional from pydantic import BaseModel, Field -from metagpt.config import CONFIG -from metagpt.const import DEFAULT_LANGUAGE, DEFAULT_MAX_TOKENS, DEFAULT_TOKEN_SIZE +from metagpt.config2 import config +from metagpt.const import DEFAULT_MAX_TOKENS, DEFAULT_TOKEN_SIZE from metagpt.logs import logger from metagpt.provider import MetaGPTLLM from metagpt.provider.base_llm import BaseLLM @@ -29,9 +29,9 @@ class BrainMemory(BaseModel): historical_summary: str = "" last_history_id: str = "" is_dirty: bool = False - last_talk: str = None + last_talk: Optional[str] = None cacheable: bool = True - llm: Optional[BaseLLM] = None + llm: Optional[BaseLLM] = Field(default=None, exclude=True) class Config: arbitrary_types_allowed = True @@ -56,8 +56,8 @@ class BrainMemory(BaseModel): @staticmethod async def loads(redis_key: str) -> "BrainMemory": - redis = Redis() - if not redis.is_valid or not redis_key: + redis = Redis(config.redis) + if not redis_key: return BrainMemory() v = await redis.get(key=redis_key) logger.debug(f"REDIS GET {redis_key} {v}") @@ -70,8 +70,8 @@ class BrainMemory(BaseModel): async def dumps(self, redis_key: str, timeout_sec: int = 30 * 60): if not self.is_dirty: return - redis = Redis() - if not redis.is_valid or not redis_key: + redis = Redis(config.redis) + if not redis_key: return False v = self.model_dump_json() if self.cacheable: @@ -83,7 +83,7 @@ class BrainMemory(BaseModel): def to_redis_key(prefix: str, user_id: str, chat_id: str): return f"{prefix}:{user_id}:{chat_id}" - async def set_history_summary(self, history_summary, redis_key, redis_conf): + async def set_history_summary(self, history_summary, redis_key): if self.historical_summary == history_summary: if self.is_dirty: await self.dumps(redis_key=redis_key) @@ -140,7 +140,7 @@ class BrainMemory(BaseModel): return text summary = await self._summarize(text=text, max_words=max_words, keep_language=keep_language, limit=limit) if summary: - await self.set_history_summary(history_summary=summary, redis_key=CONFIG.REDIS_KEY, redis_conf=CONFIG.REDIS) + await self.set_history_summary(history_summary=summary, redis_key=config.redis_key) return summary raise ValueError(f"text too long:{text_length}") @@ -164,7 +164,7 @@ class BrainMemory(BaseModel): msgs.reverse() self.history = msgs self.is_dirty = True - await self.dumps(redis_key=CONFIG.REDIS_KEY) + await self.dumps(redis_key=config.redis.key) self.is_dirty = False return BrainMemory.to_metagpt_history_format(self.history) @@ -181,7 +181,7 @@ class BrainMemory(BaseModel): summary = await self.summarize(llm=llm, max_words=500) - language = CONFIG.language or DEFAULT_LANGUAGE + language = config.language command = f"Translate the above summary into a {language} title of less than {max_words} words." summaries = [summary, command] msg = "\n".join(summaries) diff --git a/metagpt/memory/longterm_memory.py b/metagpt/memory/longterm_memory.py index b54653970..5a139a93b 100644 --- a/metagpt/memory/longterm_memory.py +++ b/metagpt/memory/longterm_memory.py @@ -2,7 +2,6 @@ # -*- coding: utf-8 -*- """ @Desc : the implement of Long-term memory -@Modified By: mashenquan, 2023/8/20. Remove global configuration `CONFIG`, enable configuration support for business isolation. """ from typing import Optional diff --git a/metagpt/memory/memory.py b/metagpt/memory/memory.py index 593409648..580361d33 100644 --- a/metagpt/memory/memory.py +++ b/metagpt/memory/memory.py @@ -7,19 +7,13 @@ @Modified By: mashenquan, 2023-11-1. According to RFC 116: Updated the type of index key. """ from collections import defaultdict -from pathlib import Path from typing import DefaultDict, Iterable, Set from pydantic import BaseModel, Field, SerializeAsAny from metagpt.const import IGNORED_MESSAGE_ID from metagpt.schema import Message -from metagpt.utils.common import ( - any_to_str, - any_to_str_set, - read_json_file, - write_json_file, -) +from metagpt.utils.common import any_to_str, any_to_str_set class Memory(BaseModel): @@ -29,22 +23,6 @@ class Memory(BaseModel): index: DefaultDict[str, list[SerializeAsAny[Message]]] = Field(default_factory=lambda: defaultdict(list)) ignore_id: bool = False - def serialize(self, stg_path: Path): - """stg_path = ./storage/team/environment/ or ./storage/team/environment/roles/{role_class}_{role_name}/""" - memory_path = stg_path.joinpath("memory.json") - storage = self.model_dump() - write_json_file(memory_path, storage) - - @classmethod - def deserialize(cls, stg_path: Path) -> "Memory": - """stg_path = ./storage/team/environment/ or ./storage/team/environment/roles/{role_class}_{role_name}/""" - memory_path = stg_path.joinpath("memory.json") - - memory_dict = read_json_file(memory_path) - memory = Memory(**memory_dict) - - return memory - def add(self, message: Message): """Add a new message to storage, while updating the index""" if self.ignore_id: diff --git a/metagpt/memory/memory_storage.py b/metagpt/memory/memory_storage.py index 1850e0ea0..c029d027b 100644 --- a/metagpt/memory/memory_storage.py +++ b/metagpt/memory/memory_storage.py @@ -2,7 +2,6 @@ # -*- coding: utf-8 -*- """ @Desc : the implement of memory storage -@Modified By: mashenquan, 2023/8/20. Remove global configuration `CONFIG`, enable configuration support for business isolation. """ from pathlib import Path diff --git a/metagpt/plan/planner.py b/metagpt/plan/planner.py index 87492e455..0d8870fd3 100644 --- a/metagpt/plan/planner.py +++ b/metagpt/plan/planner.py @@ -32,8 +32,8 @@ class Planner(BaseModel): auto_run: bool = False use_tools: bool = False - def __init__(self, goal: str, **kwargs): - plan = Plan(goal=goal) + def __init__(self, goal: str = "", plan: Plan = None, **kwargs): + plan = plan or Plan(goal=goal) super().__init__(plan=plan, **kwargs) @property @@ -44,6 +44,48 @@ class Planner(BaseModel): def current_task_id(self): return self.plan.current_task_id + async def update_plan(self, goal: str = "", max_tasks: int = 3, max_retries: int = 3): + if goal: + self.plan = Plan(goal=goal) + + plan_confirmed = False + while not plan_confirmed: + context = self.get_useful_memories() + rsp = await WritePlan().run(context, max_tasks=max_tasks, use_tools=self.use_tools) + self.working_memory.add(Message(content=rsp, role="assistant", cause_by=WritePlan)) + + # precheck plan before asking reviews + is_plan_valid, error = precheck_update_plan_from_rsp(rsp, self.plan) + if not is_plan_valid and max_retries > 0: + error_msg = f"The generated plan is not valid with error: {error}, try regenerating, remember to generate either the whole plan or the single changed task only" + logger.warning(error_msg) + self.working_memory.add(Message(content=error_msg, role="assistant", cause_by=WritePlan)) + max_retries -= 1 + continue + + _, plan_confirmed = await self.ask_review(trigger=ReviewConst.TASK_REVIEW_TRIGGER) + + update_plan_from_rsp(rsp=rsp, current_plan=self.plan) + + self.working_memory.clear() + + async def process_task_result(self, task_result: TaskResult): + # ask for acceptance, users can other refuse and change tasks in the plan + review, task_result_confirmed = await self.ask_review(task_result) + + if task_result_confirmed: + # tick off this task and record progress + await self.confirm_task(self.current_task, task_result, review) + + elif "redo" in review: + # Ask the Role to redo this task with help of review feedback, + # useful when the code run is successful but the procedure or result is not what we want + pass # simply pass, not confirming the result + + else: + # update plan according to user's feedback and to take on changed tasks + await self.update_plan() + async def ask_review( self, task_result: TaskResult = None, auto_run: bool = None, trigger: str = ReviewConst.TASK_REVIEW_TRIGGER ): @@ -74,28 +116,6 @@ class Planner(BaseModel): self.working_memory.add(Message(content=review, role="user", cause_by=AskReview)) await self.update_plan(review) - async def update_plan(self, max_tasks: int = 3, max_retries: int = 3): - plan_confirmed = False - while not plan_confirmed: - context = self.get_useful_memories() - rsp = await WritePlan().run(context, max_tasks=max_tasks, use_tools=self.use_tools) - self.working_memory.add(Message(content=rsp, role="assistant", cause_by=WritePlan)) - - # precheck plan before asking reviews - is_plan_valid, error = precheck_update_plan_from_rsp(rsp, self.plan) - if not is_plan_valid and max_retries > 0: - error_msg = f"The generated plan is not valid with error: {error}, try regenerating, remember to generate either the whole plan or the single changed task only" - logger.warning(error_msg) - self.working_memory.add(Message(content=error_msg, role="assistant", cause_by=WritePlan)) - max_retries -= 1 - continue - - _, plan_confirmed = await self.ask_review(trigger=ReviewConst.TASK_REVIEW_TRIGGER) - - update_plan_from_rsp(rsp=rsp, current_plan=self.plan) - - self.working_memory.clear() - def get_useful_memories(self, task_exclude_field=None) -> list[Message]: """find useful memories only to reduce context length and improve performance""" # TODO dataset description , code steps diff --git a/metagpt/prompts/ml_engineer.py b/metagpt/prompts/ml_action.py similarity index 64% rename from metagpt/prompts/ml_engineer.py rename to metagpt/prompts/ml_action.py index ac95e14bd..582b01146 100644 --- a/metagpt/prompts/ml_engineer.py +++ b/metagpt/prompts/ml_action.py @@ -2,7 +2,7 @@ # -*- coding: utf-8 -*- # @Time : 2023/11/24 15:43 # @Author : lidanyang -# @File : ml_engineer +# @File : ml_action # @Desc : UPDATE_DATA_COLUMNS = """ # Background @@ -49,85 +49,6 @@ Output the information in a JSON format, as shown in this example: - Don't contain specific values or examples found in the data column. """ -ASSIGN_TASK_TYPE_PROMPT = """ -Please assign a task type to each task in the list below from the given categories: -{task_list} - -## All Task Type: -{task_type_desc} -""" - -ASSIGN_TASK_TYPE_CONFIG = { - "name": "assign_task_type", - "description": "Assign task type to each task by order.", - "parameters": { - "type": "object", - "properties": { - "task_type": { - "type": "array", - "description": "List of task type. The length should as long as task list", - "items": { - "type": "string", - }, - }, - }, - "required": ["task_type"], - }, -} - -TOOL_RECOMMENDATION_PROMPT = """ -## User Requirement: -{current_task} - -## Task -Recommend up to five tools from 'Available Tools' that can help solve the 'User Requirement'. -This is a detailed code steps for current task. You can refer to it when recommending tools. -{code_steps} - -## Available Tools: -{available_tools} - -## Tool Selection and Instructions: -- Select tools most relevant to completing the 'User Requirement'. -- If you believe that no tools are suitable, indicate with an empty list. -- Only list the names of the tools, not the full schema of each tool. -- Ensure selected tools are listed in 'Available Tools'. -""" - -SELECT_FUNCTION_TOOLS = { - "name": "select_function_tools", - "description": "For current task, select suitable tools for it.", - "parameters": { - "type": "object", - "properties": { - "recommend_tools": { - "type": "array", - "description": "List of tool names. Empty list if no tool is suitable.", - "items": { - "type": "string", - }, - }, - }, - "required": ["recommend_tools"], - }, -} - -CODE_GENERATOR_WITH_TOOLS = { - "name": "add_subtask_code", - "description": "Add new code cell of current task to the end of an active Jupyter notebook.", - "parameters": { - "type": "object", - "properties": { - "code": { - "type": "string", - "description": "The code to be added to a new cell in jupyter.", - }, - }, - "required": ["code"], - }, -} - - PRINT_DATA_COLUMNS = { "name": "print_column_info", "description": "Print the latest column information after 'Done Tasks' code if first read or data changed.", @@ -189,24 +110,6 @@ model.fit(train, y_train) - The output code should contain all steps implemented in 'Code Steps'. """ -TOOL_USAGE_PROMPT = """ -# Instruction -Write complete code for 'Current Task'. And avoid duplicating code from finished tasks, such as repeated import of packages, reading data, etc. -Specifically, {tool_type_usage_prompt} - -# Capabilities -- You can utilize pre-defined tools in any code lines from 'Available Tools' in the form of Python Class. -- You can freely combine the use of any other public packages, like sklearn, numpy, pandas, etc.. - -# Available Tools (can be empty): -Each Class tool is described in JSON format. When you call a tool, import the tool first. -{tool_schemas} - -# Constraints: -- Ensure the output new code is executable in the same Jupyter notebook with previous tasks code have been executed. -- Always prioritize using pre-defined tools for the same functionality. -""" - ML_TOOL_USAGE_PROMPT = """ # Background As a data scientist, you need to help user to achieve their goal [{user_requirement}] step-by-step in an continuous Jupyter notebook. diff --git a/metagpt/prompts/tool_types.py b/metagpt/prompts/tool_types.py index 42d9c1ece..381fb25ad 100644 --- a/metagpt/prompts/tool_types.py +++ b/metagpt/prompts/tool_types.py @@ -25,7 +25,7 @@ The current task is about feature engineering. when performing it, please adhere # Prompt for using tools of "model_train" type MODEL_TRAIN_PROMPT = """ The current task is about training a model, please ensure high performance: -- Keep in mind that your user prioritizes results and is highly focused on model performance. So, when needed, feel free to use models of any complexity to improve effectiveness, such as lightGBM, XGBoost, CatBoost, etc. +- Keep in mind that your user prioritizes results and is highly focused on model performance. So, when needed, feel free to use models of any complexity to improve effectiveness, such as XGBoost, CatBoost, etc. - If non-numeric columns exist, perform label encode together with all steps. - Use the data from previous task result directly, do not mock or reload data yourself. - Set suitable hyperparameters for the model, make metrics as high as possible. diff --git a/metagpt/prompts/write_analysis_code.py b/metagpt/prompts/write_analysis_code.py new file mode 100644 index 000000000..4c8a5081e --- /dev/null +++ b/metagpt/prompts/write_analysis_code.py @@ -0,0 +1,95 @@ +ASSIGN_TASK_TYPE_PROMPT = """ +Please assign a task type to each task in the list below from the given categories: +{task_list} + +## All Task Type: +{task_type_desc} +""" + +ASSIGN_TASK_TYPE_CONFIG = { + "name": "assign_task_type", + "description": "Assign task type to each task by order.", + "parameters": { + "type": "object", + "properties": { + "task_type": { + "type": "array", + "description": "List of task type. The length should as long as task list", + "items": { + "type": "string", + }, + }, + }, + "required": ["task_type"], + }, +} + +TOOL_RECOMMENDATION_PROMPT = """ +## User Requirement: +{current_task} + +## Task +Recommend up to five tools from 'Available Tools' that can help solve the 'User Requirement'. +This is a detailed code steps for current task. You can refer to it when recommending tools. +{code_steps} + +## Available Tools: +{available_tools} + +## Tool Selection and Instructions: +- Select tools most relevant to completing the 'User Requirement'. +- If you believe that no tools are suitable, indicate with an empty list. +- Only list the names of the tools, not the full schema of each tool. +- Ensure selected tools are listed in 'Available Tools'. +""" + +SELECT_FUNCTION_TOOLS = { + "name": "select_function_tools", + "description": "For current task, select suitable tools for it.", + "parameters": { + "type": "object", + "properties": { + "recommend_tools": { + "type": "array", + "description": "List of tool names. Empty list if no tool is suitable.", + "items": { + "type": "string", + }, + }, + }, + "required": ["recommend_tools"], + }, +} + +CODE_GENERATOR_WITH_TOOLS = { + "name": "add_subtask_code", + "description": "Add new code cell of current task to the end of an active Jupyter notebook.", + "parameters": { + "type": "object", + "properties": { + "code": { + "type": "string", + "description": "The code to be added to a new cell in jupyter.", + }, + }, + "required": ["code"], + }, +} + +TOOL_USAGE_PROMPT = """ +# Instruction +Write complete code for 'Current Task'. And avoid duplicating code from finished tasks, such as repeated import of packages, reading data, etc. +Specifically, {tool_type_usage_prompt} + +# Capabilities +- You can utilize pre-defined tools in any code lines from 'Available Tools' in the form of Python Class. +- You can freely combine the use of any other public packages, like sklearn, numpy, pandas, etc.. + +# Available Tools (can be empty): +Each Class tool is described in JSON format. When you call a tool, import the tool first. +{tool_schemas} + +# Constraints: +- Ensure the output new code is executable in the same Jupyter notebook with previous tasks code have been executed. +- Always prioritize using pre-defined tools for the same functionality. +""" diff --git a/metagpt/provider/__init__.py b/metagpt/provider/__init__.py index 28157a4e2..675734811 100644 --- a/metagpt/provider/__init__.py +++ b/metagpt/provider/__init__.py @@ -14,6 +14,8 @@ from metagpt.provider.openai_api import OpenAILLM from metagpt.provider.zhipuai_api import ZhiPuAILLM from metagpt.provider.azure_openai_api import AzureOpenAILLM from metagpt.provider.metagpt_api import MetaGPTLLM +from metagpt.provider.human_provider import HumanProvider +from metagpt.provider.spark_api import SparkLLM __all__ = [ "FireworksLLM", @@ -24,4 +26,6 @@ __all__ = [ "AzureOpenAILLM", "MetaGPTLLM", "OllamaLLM", + "HumanProvider", + "SparkLLM", ] diff --git a/metagpt/provider/anthropic_api.py b/metagpt/provider/anthropic_api.py index b9d7d9e38..f31c2d04d 100644 --- a/metagpt/provider/anthropic_api.py +++ b/metagpt/provider/anthropic_api.py @@ -9,12 +9,15 @@ import anthropic from anthropic import Anthropic, AsyncAnthropic -from metagpt.config import CONFIG +from metagpt.configs.llm_config import LLMConfig class Claude2: + def __init__(self, config: LLMConfig): + self.config = config + def ask(self, prompt: str) -> str: - client = Anthropic(api_key=CONFIG.anthropic_api_key) + client = Anthropic(api_key=self.config.api_key) res = client.completions.create( model="claude-2", @@ -24,7 +27,7 @@ class Claude2: return res.completion async def aask(self, prompt: str) -> str: - aclient = AsyncAnthropic(api_key=CONFIG.anthropic_api_key) + aclient = AsyncAnthropic(api_key=self.config.api_key) res = await aclient.completions.create( model="claude-2", diff --git a/metagpt/provider/azure_openai_api.py b/metagpt/provider/azure_openai_api.py index d15d1c82e..6dc32d380 100644 --- a/metagpt/provider/azure_openai_api.py +++ b/metagpt/provider/azure_openai_api.py @@ -3,8 +3,6 @@ @Time : 2023/5/5 23:08 @Author : alexanderwu @File : openai.py -@Modified By: mashenquan, 2023/8/20. Remove global configuration `CONFIG`, enable configuration support for business isolation; - Change cost control from global to company level. @Modified By: mashenquan, 2023/11/21. Fix bug: ReadTimeout. @Modified By: mashenquan, 2023/12/1. Fix bug: Unclosed connection caused by openai 0.x. """ @@ -13,12 +11,12 @@ from openai import AsyncAzureOpenAI from openai._base_client import AsyncHttpxClientWrapper -from metagpt.config import LLMProviderEnum +from metagpt.configs.llm_config import LLMType from metagpt.provider.llm_provider_registry import register_provider from metagpt.provider.openai_api import OpenAILLM -@register_provider(LLMProviderEnum.AZURE_OPENAI) +@register_provider(LLMType.AZURE) class AzureOpenAILLM(OpenAILLM): """ Check https://platform.openai.com/examples for examples @@ -28,13 +26,13 @@ class AzureOpenAILLM(OpenAILLM): kwargs = self._make_client_kwargs() # https://learn.microsoft.com/zh-cn/azure/ai-services/openai/how-to/migration?tabs=python-new%2Cdalle-fix self.aclient = AsyncAzureOpenAI(**kwargs) - self.model = self.config.DEPLOYMENT_NAME # Used in _calc_usage & _cons_kwargs + self.model = self.config.model # Used in _calc_usage & _cons_kwargs def _make_client_kwargs(self) -> dict: kwargs = dict( - api_key=self.config.OPENAI_API_KEY, - api_version=self.config.OPENAI_API_VERSION, - azure_endpoint=self.config.OPENAI_BASE_URL, + api_key=self.config.api_key, + api_version=self.config.api_version, + azure_endpoint=self.config.base_url, ) # to use proxy, openai v1 needs http_client diff --git a/metagpt/provider/base_llm.py b/metagpt/provider/base_llm.py index dbef15fa1..65e70af00 100644 --- a/metagpt/provider/base_llm.py +++ b/metagpt/provider/base_llm.py @@ -8,15 +8,32 @@ """ import json from abc import ABC, abstractmethod -from typing import Optional +from typing import Optional, Union + +from openai import AsyncOpenAI + +from metagpt.configs.llm_config import LLMConfig +from metagpt.logs import logger +from metagpt.schema import Message +from metagpt.utils.cost_manager import CostManager class BaseLLM(ABC): """LLM API abstract class, requiring all inheritors to provide a series of standard capabilities""" + config: LLMConfig use_system_prompt: bool = True system_prompt = "You are a helpful assistant." + # OpenAI / Azure / Others + aclient: Optional[Union[AsyncOpenAI]] = None + cost_manager: Optional[CostManager] = None + model: Optional[str] = None + + @abstractmethod + def __init__(self, config: LLMConfig): + pass + def _user_msg(self, msg: str) -> dict[str, str]: return {"role": "user", "content": msg} @@ -43,10 +60,13 @@ class BaseLLM(ABC): if system_msgs: message = self._system_msgs(system_msgs) else: - message = [self._default_system_msg()] if self.use_system_prompt else [] + message = [self._default_system_msg()] + if not self.use_system_prompt: + message = [] if format_msgs: message.extend(format_msgs) message.append(self._user_msg(msg)) + logger.debug(message) rsp = await self.acompletion_text(message, stream=stream, timeout=timeout) return rsp @@ -63,10 +83,9 @@ class BaseLLM(ABC): context.append(self._assistant_msg(rsp_text)) return self._extract_assistant_rsp(context) - async def aask_code(self, msgs: list[str], timeout=3) -> str: + async def aask_code(self, messages: Union[str, Message, list[dict]], timeout=3) -> dict: """FIXME: No code segment filtering has been done here, and all results are actually displayed""" - rsp_text = await self.aask_batch(msgs, timeout=timeout) - return rsp_text + raise NotImplementedError @abstractmethod async def acompletion(self, messages: list[dict], timeout=3): @@ -87,6 +106,10 @@ class BaseLLM(ABC): """Required to provide the first text of choice""" return rsp.get("choices")[0]["message"]["content"] + def get_choice_delta_text(self, rsp: dict) -> str: + """Required to provide the first text of stream choice""" + return rsp.get("choices")[0]["delta"]["content"] + def get_choice_function(self, rsp: dict) -> dict: """Required to provide the first function of choice :param dict rsp: OpenAI chat.comletion respond JSON, Note "message" must include "tool_calls", diff --git a/metagpt/provider/fireworks_api.py b/metagpt/provider/fireworks_api.py index f0af68818..d56453a85 100644 --- a/metagpt/provider/fireworks_api.py +++ b/metagpt/provider/fireworks_api.py @@ -15,7 +15,7 @@ from tenacity import ( wait_random_exponential, ) -from metagpt.config import CONFIG, Config, LLMProviderEnum +from metagpt.configs.llm_config import LLMConfig, LLMType from metagpt.logs import logger from metagpt.provider.llm_provider_registry import register_provider from metagpt.provider.openai_api import OpenAILLM, log_and_reraise @@ -64,44 +64,35 @@ class FireworksCostManager(CostManager): token_costs = self.model_grade_token_costs(model) cost = (prompt_tokens * token_costs["prompt"] + completion_tokens * token_costs["completion"]) / 1000000 self.total_cost += cost - max_budget = CONFIG.max_budget if CONFIG.max_budget else CONFIG.cost_manager.max_budget logger.info( - f"Total running cost: ${self.total_cost:.4f} | Max budget: ${max_budget:.3f} | " + f"Total running cost: ${self.total_cost:.4f}" f"Current cost: ${cost:.4f}, prompt_tokens: {prompt_tokens}, completion_tokens: {completion_tokens}" ) - CONFIG.total_cost = self.total_cost -@register_provider(LLMProviderEnum.FIREWORKS) +@register_provider(LLMType.FIREWORKS) class FireworksLLM(OpenAILLM): - def __init__(self): - self.config: Config = CONFIG - self.__init_fireworks() + def __init__(self, config: LLMConfig): + super().__init__(config=config) self.auto_max_tokens = False - self._cost_manager = FireworksCostManager() - - def __init_fireworks(self): - self.is_azure = False - self.rpm = int(self.config.get("RPM", 10)) - self._init_client() - self.model = self.config.fireworks_api_model # `self.model` should after `_make_client` to rewrite it + self.cost_manager = FireworksCostManager() def _make_client_kwargs(self) -> dict: - kwargs = dict(api_key=self.config.fireworks_api_key, base_url=self.config.fireworks_api_base) + kwargs = dict(api_key=self.config.api_key, base_url=self.config.base_url) return kwargs def _update_costs(self, usage: CompletionUsage): if self.config.calc_usage and usage: try: - # use FireworksCostManager not CONFIG.cost_manager - self._cost_manager.update_cost(usage.prompt_tokens, usage.completion_tokens, self.model) + # use FireworksCostManager not context.cost_manager + self.cost_manager.update_cost(usage.prompt_tokens, usage.completion_tokens, self.model) except Exception as e: logger.error(f"updating costs failed!, exp: {e}") def get_costs(self) -> Costs: - return self._cost_manager.get_costs() + return self.cost_manager.get_costs() - async def _achat_completion_stream(self, messages: list[dict]) -> str: + async def _achat_completion_stream(self, messages: list[dict], timeout=3) -> str: response: AsyncStream[ChatCompletionChunk] = await self.aclient.chat.completions.create( **self._cons_kwargs(messages), stream=True ) diff --git a/metagpt/provider/general_api_requestor.py b/metagpt/provider/general_api_requestor.py index cf31fd629..500cd1426 100644 --- a/metagpt/provider/general_api_requestor.py +++ b/metagpt/provider/general_api_requestor.py @@ -79,10 +79,8 @@ class GeneralAPIRequestor(APIRequestor): async def _interpret_async_response( self, result: aiohttp.ClientResponse, stream: bool ) -> Tuple[Union[bytes, AsyncGenerator[bytes, None]], bool]: - if stream and ( - "text/event-stream" in result.headers.get("Content-Type", "") - or "application/x-ndjson" in result.headers.get("Content-Type", "") - ): + content_type = result.headers.get("Content-Type", "") + if stream and ("text/event-stream" in content_type or "application/x-ndjson" in content_type): # the `Content-Type` of ollama stream resp is "application/x-ndjson" return ( self._interpret_response_line(line, result.status, result.headers, stream=True) diff --git a/metagpt/provider/google_gemini_api.py b/metagpt/provider/google_gemini_api.py index c36c677ef..6df814b55 100644 --- a/metagpt/provider/google_gemini_api.py +++ b/metagpt/provider/google_gemini_api.py @@ -19,7 +19,7 @@ from tenacity import ( wait_random_exponential, ) -from metagpt.config import CONFIG, LLMProviderEnum +from metagpt.configs.llm_config import LLMConfig, LLMType from metagpt.logs import log_llm_stream, logger from metagpt.provider.base_llm import BaseLLM from metagpt.provider.llm_provider_registry import register_provider @@ -41,21 +41,22 @@ class GeminiGenerativeModel(GenerativeModel): return await self._async_client.count_tokens(model=self.model_name, contents=contents) -@register_provider(LLMProviderEnum.GEMINI) +@register_provider(LLMType.GEMINI) class GeminiLLM(BaseLLM): """ Refs to `https://ai.google.dev/tutorials/python_quickstart` """ - def __init__(self): + def __init__(self, config: LLMConfig): self.use_system_prompt = False # google gemini has no system prompt when use api - self.__init_gemini(CONFIG) + self.__init_gemini(config) + self.config = config self.model = "gemini-pro" # so far only one model self.llm = GeminiGenerativeModel(model_name=self.model) - def __init_gemini(self, config: CONFIG): - genai.configure(api_key=config.gemini_api_key) + def __init_gemini(self, config: LLMConfig): + genai.configure(api_key=config.api_key) def _user_msg(self, msg: str) -> dict[str, str]: # Not to change BaseLLM default functions but update with Gemini's conversation format. @@ -71,11 +72,11 @@ class GeminiLLM(BaseLLM): def _update_costs(self, usage: dict): """update each request's token cost""" - if CONFIG.calc_usage: + if self.config.calc_usage: try: prompt_tokens = int(usage.get("prompt_tokens", 0)) completion_tokens = int(usage.get("completion_tokens", 0)) - CONFIG.cost_manager.update_cost(prompt_tokens, completion_tokens, self.model) + self.cost_manager.update_cost(prompt_tokens, completion_tokens, self.model) except Exception as e: logger.error(f"google gemini updats costs failed! exp: {e}") @@ -108,7 +109,7 @@ class GeminiLLM(BaseLLM): self._update_costs(usage) return resp - async def acompletion(self, messages: list[dict]) -> dict: + async def acompletion(self, messages: list[dict], timeout=3) -> dict: return await self._achat_completion(messages) async def _achat_completion_stream(self, messages: list[dict]) -> str: diff --git a/metagpt/provider/human_provider.py b/metagpt/provider/human_provider.py index 59d236a3a..fe000b3a6 100644 --- a/metagpt/provider/human_provider.py +++ b/metagpt/provider/human_provider.py @@ -5,6 +5,7 @@ Author: garylin2099 """ from typing import Optional +from metagpt.configs.llm_config import LLMConfig from metagpt.logs import logger from metagpt.provider.base_llm import BaseLLM @@ -14,6 +15,9 @@ class HumanProvider(BaseLLM): This enables replacing LLM anywhere in the framework with a human, thus introducing human interaction """ + def __init__(self, config: LLMConfig): + pass + def ask(self, msg: str, timeout=3) -> str: logger.info("It's your turn, please type in your response. You may also refer to the context below") rsp = input(msg) diff --git a/metagpt/provider/llm_provider_registry.py b/metagpt/provider/llm_provider_registry.py index 2b3ef93a3..df89d36aa 100644 --- a/metagpt/provider/llm_provider_registry.py +++ b/metagpt/provider/llm_provider_registry.py @@ -5,7 +5,8 @@ @Author : alexanderwu @File : llm_provider_registry.py """ -from metagpt.config import LLMProviderEnum +from metagpt.configs.llm_config import LLMConfig, LLMType +from metagpt.provider.base_llm import BaseLLM class LLMProviderRegistry: @@ -15,13 +16,9 @@ class LLMProviderRegistry: def register(self, key, provider_cls): self.providers[key] = provider_cls - def get_provider(self, enum: LLMProviderEnum): + def get_provider(self, enum: LLMType): """get provider instance according to the enum""" - return self.providers[enum]() - - -# Registry instance -LLM_REGISTRY = LLMProviderRegistry() + return self.providers[enum] def register_provider(key): @@ -32,3 +29,12 @@ def register_provider(key): return cls return decorator + + +def create_llm_instance(config: LLMConfig) -> BaseLLM: + """get the default llm provider""" + return LLM_REGISTRY.get_provider(config.api_type)(config) + + +# Registry instance +LLM_REGISTRY = LLMProviderRegistry() diff --git a/metagpt/provider/metagpt_api.py b/metagpt/provider/metagpt_api.py index 69aa7f305..4956746dc 100644 --- a/metagpt/provider/metagpt_api.py +++ b/metagpt/provider/metagpt_api.py @@ -5,12 +5,11 @@ @File : metagpt_api.py @Desc : MetaGPT LLM provider. """ -from metagpt.config import LLMProviderEnum +from metagpt.configs.llm_config import LLMType from metagpt.provider import OpenAILLM from metagpt.provider.llm_provider_registry import register_provider -@register_provider(LLMProviderEnum.METAGPT) +@register_provider(LLMType.METAGPT) class MetaGPTLLM(OpenAILLM): - def __init__(self): - super().__init__() + pass diff --git a/metagpt/provider/ollama_api.py b/metagpt/provider/ollama_api.py index 25086737f..c9103b018 100644 --- a/metagpt/provider/ollama_api.py +++ b/metagpt/provider/ollama_api.py @@ -13,48 +13,34 @@ from tenacity import ( wait_random_exponential, ) -from metagpt.config import CONFIG, LLMProviderEnum +from metagpt.configs.llm_config import LLMConfig, LLMType from metagpt.const import LLM_API_TIMEOUT from metagpt.logs import log_llm_stream, logger from metagpt.provider.base_llm import BaseLLM from metagpt.provider.general_api_requestor import GeneralAPIRequestor from metagpt.provider.llm_provider_registry import register_provider from metagpt.provider.openai_api import log_and_reraise -from metagpt.utils.cost_manager import CostManager +from metagpt.utils.cost_manager import TokenCostManager -class OllamaCostManager(CostManager): - def update_cost(self, prompt_tokens, completion_tokens, model): - """ - Update the total cost, prompt tokens, and completion tokens. - """ - self.total_prompt_tokens += prompt_tokens - self.total_completion_tokens += completion_tokens - max_budget = CONFIG.max_budget if CONFIG.max_budget else CONFIG.cost_manager.max_budget - logger.info( - f"Max budget: ${max_budget:.3f} | " - f"prompt_tokens: {prompt_tokens}, completion_tokens: {completion_tokens}" - ) - CONFIG.total_cost = self.total_cost - - -@register_provider(LLMProviderEnum.OLLAMA) +@register_provider(LLMType.OLLAMA) class OllamaLLM(BaseLLM): """ Refs to `https://github.com/jmorganca/ollama/blob/main/docs/api.md#generate-a-chat-completion` """ - def __init__(self): - self.__init_ollama(CONFIG) - self.client = GeneralAPIRequestor(base_url=CONFIG.ollama_api_base) + def __init__(self, config: LLMConfig): + self.__init_ollama(config) + self.client = GeneralAPIRequestor(base_url=config.base_url) + self.config = config self.suffix_url = "/chat" self.http_method = "post" self.use_system_prompt = False - self._cost_manager = OllamaCostManager() + self._cost_manager = TokenCostManager() - def __init_ollama(self, config: CONFIG): - assert config.ollama_api_base - self.model = config.ollama_api_model + def __init_ollama(self, config: LLMConfig): + assert config.base_url, "ollama base url is required!" + self.model = config.model def _const_kwargs(self, messages: list[dict], stream: bool = False) -> dict: kwargs = {"model": self.model, "messages": messages, "options": {"temperature": 0.3}, "stream": stream} @@ -62,7 +48,7 @@ class OllamaLLM(BaseLLM): def _update_costs(self, usage: dict): """update each request's token cost""" - if CONFIG.calc_usage: + if self.config.calc_usage: try: prompt_tokens = int(usage.get("prompt_tokens", 0)) completion_tokens = int(usage.get("completion_tokens", 0)) diff --git a/metagpt/provider/open_llm_api.py b/metagpt/provider/open_llm_api.py index b0c484f5a..a29b263a4 100644 --- a/metagpt/provider/open_llm_api.py +++ b/metagpt/provider/open_llm_api.py @@ -4,56 +4,27 @@ from openai.types import CompletionUsage -from metagpt.config import CONFIG, Config, LLMProviderEnum +from metagpt.configs.llm_config import LLMConfig, LLMType from metagpt.logs import logger from metagpt.provider.llm_provider_registry import register_provider from metagpt.provider.openai_api import OpenAILLM -from metagpt.utils.cost_manager import CostManager, Costs +from metagpt.utils.cost_manager import Costs, TokenCostManager from metagpt.utils.token_counter import count_message_tokens, count_string_tokens -class OpenLLMCostManager(CostManager): - """open llm model is self-host, it's free and without cost""" - - def update_cost(self, prompt_tokens, completion_tokens, model): - """ - Update the total cost, prompt tokens, and completion tokens. - - Args: - prompt_tokens (int): The number of tokens used in the prompt. - completion_tokens (int): The number of tokens used in the completion. - model (str): The model used for the API call. - """ - self.total_prompt_tokens += prompt_tokens - self.total_completion_tokens += completion_tokens - max_budget = CONFIG.max_budget if CONFIG.max_budget else CONFIG.cost_manager.max_budget - logger.info( - f"Max budget: ${max_budget:.3f} | reference " - f"prompt_tokens: {prompt_tokens}, completion_tokens: {completion_tokens}" - ) - - -@register_provider(LLMProviderEnum.OPEN_LLM) +@register_provider(LLMType.OPEN_LLM) class OpenLLM(OpenAILLM): - def __init__(self): - self.config: Config = CONFIG - self.__init_openllm() - self.auto_max_tokens = False - self._cost_manager = OpenLLMCostManager() - - def __init_openllm(self): - self.is_azure = False - self.rpm = int(self.config.get("RPM", 10)) - self._init_client() - self.model = self.config.open_llm_api_model # `self.model` should after `_make_client` to rewrite it + def __init__(self, config: LLMConfig): + super().__init__(config) + self._cost_manager = TokenCostManager() def _make_client_kwargs(self) -> dict: - kwargs = dict(api_key="sk-xxx", base_url=self.config.open_llm_api_base) + kwargs = dict(api_key="sk-xxx", base_url=self.config.base_url) return kwargs def _calc_usage(self, messages: list[dict], rsp: str) -> CompletionUsage: usage = CompletionUsage(prompt_tokens=0, completion_tokens=0, total_tokens=0) - if not CONFIG.calc_usage: + if not self.config.calc_usage: return usage try: diff --git a/metagpt/provider/openai_api.py b/metagpt/provider/openai_api.py index 386c36c22..206701efd 100644 --- a/metagpt/provider/openai_api.py +++ b/metagpt/provider/openai_api.py @@ -3,15 +3,13 @@ @Time : 2023/5/5 23:08 @Author : alexanderwu @File : openai.py -@Modified By: mashenquan, 2023/8/20. Remove global configuration `CONFIG`, enable configuration support for isolation; - Change cost control from global to company level. @Modified By: mashenquan, 2023/11/21. Fix bug: ReadTimeout. @Modified By: mashenquan, 2023/12/1. Fix bug: Unclosed connection caused by openai 0.x. """ import json import re -from typing import AsyncIterator, Union +from typing import AsyncIterator, Optional, Union from openai import APIConnectionError, AsyncOpenAI, AsyncStream from openai._base_client import AsyncHttpxClientWrapper @@ -25,14 +23,14 @@ from tenacity import ( wait_random_exponential, ) -from metagpt.config import CONFIG, Config, LLMProviderEnum +from metagpt.configs.llm_config import LLMConfig, LLMType from metagpt.logs import log_llm_stream, logger from metagpt.provider.base_llm import BaseLLM from metagpt.provider.constant import GENERAL_FUNCTION_SCHEMA from metagpt.provider.llm_provider_registry import register_provider from metagpt.schema import Message from metagpt.utils.common import CodeParser -from metagpt.utils.cost_manager import Costs +from metagpt.utils.cost_manager import CostManager, Costs from metagpt.utils.exceptions import handle_exception from metagpt.utils.token_counter import ( count_message_tokens, @@ -52,18 +50,19 @@ See FAQ 5.8 raise retry_state.outcome.exception() -@register_provider(LLMProviderEnum.OPENAI) +@register_provider(LLMType.OPENAI) class OpenAILLM(BaseLLM): """Check https://platform.openai.com/examples for examples""" - def __init__(self): - self.config: Config = CONFIG - self._init_openai() + def __init__(self, config: LLMConfig): + self.config = config + self._init_model() self._init_client() self.auto_max_tokens = False + self.cost_manager: Optional[CostManager] = None - def _init_openai(self): - self.model = self.config.OPENAI_API_MODEL # Used in _calc_usage & _cons_kwargs + def _init_model(self): + self.model = self.config.model # Used in _calc_usage & _cons_kwargs def _init_client(self): """https://github.com/openai/openai-python#async-usage""" @@ -71,7 +70,7 @@ class OpenAILLM(BaseLLM): self.aclient = AsyncOpenAI(**kwargs) def _make_client_kwargs(self) -> dict: - kwargs = {"api_key": self.config.openai_api_key, "base_url": self.config.openai_base_url} + kwargs = {"api_key": self.config.api_key, "base_url": self.config.base_url} # to use proxy, openai v1 needs http_client if proxy_params := self._get_proxy_params(): @@ -81,10 +80,10 @@ class OpenAILLM(BaseLLM): def _get_proxy_params(self) -> dict: params = {} - if self.config.openai_proxy: - params = {"proxies": self.config.openai_proxy} - if self.config.openai_base_url: - params["base_url"] = self.config.openai_base_url + if self.config.proxy: + params = {"proxies": self.config.proxy} + if self.config.base_url: + params["base_url"] = self.config.base_url return params @@ -105,7 +104,7 @@ class OpenAILLM(BaseLLM): "stop": None, "temperature": 0.3, "model": self.model, - "timeout": max(CONFIG.timeout, timeout), + "timeout": max(self.config.timeout, timeout), } if extra_kwargs: kwargs.update(extra_kwargs) @@ -266,7 +265,7 @@ class OpenAILLM(BaseLLM): def _calc_usage(self, messages: list[dict], rsp: str) -> CompletionUsage: usage = CompletionUsage(prompt_tokens=0, completion_tokens=0, total_tokens=0) - if not CONFIG.calc_usage: + if not self.config.calc_usage: return usage try: @@ -279,18 +278,28 @@ class OpenAILLM(BaseLLM): @handle_exception def _update_costs(self, usage: CompletionUsage): - if CONFIG.calc_usage and usage: - CONFIG.cost_manager.update_cost(usage.prompt_tokens, usage.completion_tokens, self.model) + if self.config.calc_usage and usage and self.cost_manager: + self.cost_manager.update_cost(usage.prompt_tokens, usage.completion_tokens, self.model) def get_costs(self) -> Costs: - return CONFIG.cost_manager.get_costs() + if not self.cost_manager: + return Costs(0, 0, 0, 0) + return self.cost_manager.get_costs() def _get_max_tokens(self, messages: list[dict]): if not self.auto_max_tokens: - return CONFIG.max_tokens_rsp - return get_max_completion_tokens(messages, self.model, CONFIG.max_tokens_rsp) + return self.config.max_token + return get_max_completion_tokens(messages, self.model, self.config.max_tokens) @handle_exception async def amoderation(self, content: Union[str, list[str]]): """Moderate content.""" return await self.aclient.moderations.create(input=content) + + async def atext_to_speech(self, **kwargs): + """text to speech""" + return await self.aclient.audio.speech.create(**kwargs) + + async def aspeech_to_text(self, **kwargs): + """speech to text""" + return await self.aclient.audio.transcriptions.create(**kwargs) diff --git a/metagpt/provider/spark_api.py b/metagpt/provider/spark_api.py index ce889529a..5e89c26d5 100644 --- a/metagpt/provider/spark_api.py +++ b/metagpt/provider/spark_api.py @@ -16,29 +16,30 @@ from wsgiref.handlers import format_date_time import websocket # 使用websocket_client -from metagpt.config import CONFIG, LLMProviderEnum +from metagpt.configs.llm_config import LLMConfig, LLMType from metagpt.logs import logger from metagpt.provider.base_llm import BaseLLM from metagpt.provider.llm_provider_registry import register_provider -@register_provider(LLMProviderEnum.SPARK) +@register_provider(LLMType.SPARK) class SparkLLM(BaseLLM): - def __init__(self): - logger.warning("当前方法无法支持异步运行。当你使用acompletion时，并不能并行访问。") + def __init__(self, config: LLMConfig): + self.config = config + logger.warning("SparkLLM：当前方法无法支持异步运行。当你使用acompletion时，并不能并行访问。") def get_choice_text(self, rsp: dict) -> str: return rsp["payload"]["choices"]["text"][-1]["content"] async def acompletion_text(self, messages: list[dict], stream=False, timeout: int = 3) -> str: # 不支持 - logger.error("该功能禁用。") - w = GetMessageFromWeb(messages) + # logger.warning("当前方法无法支持异步运行。当你使用acompletion时，并不能并行访问。") + w = GetMessageFromWeb(messages, self.config) return w.run() async def acompletion(self, messages: list[dict], timeout=3): # 不支持异步 - w = GetMessageFromWeb(messages) + w = GetMessageFromWeb(messages, self.config) return w.run() @@ -89,14 +90,14 @@ class GetMessageFromWeb: # 此处打印出建立连接时候的url,参考本demo的时候可取消上方打印的注释，比对相同参数时生成的url与自己代码生成的url是否一致 return url - def __init__(self, text): + def __init__(self, text, config: LLMConfig): self.text = text self.ret = "" - self.spark_appid = CONFIG.spark_appid - self.spark_api_secret = CONFIG.spark_api_secret - self.spark_api_key = CONFIG.spark_api_key - self.domain = CONFIG.domain - self.spark_url = CONFIG.spark_url + self.spark_appid = config.app_id + self.spark_api_secret = config.api_secret + self.spark_api_key = config.api_key + self.domain = config.domain + self.spark_url = config.base_url def on_message(self, ws, message): data = json.loads(message) diff --git a/metagpt/provider/zhipuai/async_sse_client.py b/metagpt/provider/zhipuai/async_sse_client.py index d7168202a..054865652 100644 --- a/metagpt/provider/zhipuai/async_sse_client.py +++ b/metagpt/provider/zhipuai/async_sse_client.py @@ -1,75 +1,31 @@ #!/usr/bin/env python # -*- coding: utf-8 -*- # @Desc : async_sse_client to make keep the use of Event to access response -# refs to `https://github.com/zhipuai/zhipuai-sdk-python/blob/main/zhipuai/utils/sse_client.py` +# refs to `zhipuai/core/_sse_client.py` -from zhipuai.utils.sse_client import _FIELD_SEPARATOR, Event, SSEClient +import json +from typing import Any, Iterator -class AsyncSSEClient(SSEClient): - async def _aread(self): - data = b"" +class AsyncSSEClient(object): + def __init__(self, event_source: Iterator[Any]): + self._event_source = event_source + + async def stream(self) -> dict: + if isinstance(self._event_source, bytes): + raise RuntimeError( + f"Request failed, msg: {self._event_source.decode('utf-8')}, please ref to `https://open.bigmodel.cn/dev/api#error-code-v3`" + ) async for chunk in self._event_source: - for line in chunk.splitlines(True): - data += line - if data.endswith((b"\r\r", b"\n\n", b"\r\n\r\n")): - yield data - data = b"" - if data: - yield data + line = chunk.decode("utf-8") + if line.startswith(":") or not line: + return - async def async_events(self): - async for chunk in self._aread(): - event = Event() - # Split before decoding so splitlines() only uses \r and \n - for line in chunk.splitlines(): - # Decode the line. - line = line.decode(self._char_enc) - - # Lines starting with a separator are comments and are to be - # ignored. - if not line.strip() or line.startswith(_FIELD_SEPARATOR): - continue - - data = line.split(_FIELD_SEPARATOR, 1) - field = data[0] - - # Ignore unknown fields. - if field not in event.__dict__: - self._logger.debug("Saw invalid field %s while parsing " "Server Side Event", field) - continue - - if len(data) > 1: - # From the spec: - # "If value starts with a single U+0020 SPACE character, - # remove it from value." - if data[1].startswith(" "): - value = data[1][1:] - else: - value = data[1] - else: - # If no value is present after the separator, - # assume an empty value. - value = "" - - # The data field may come over multiple lines and their values - # are concatenated with each other. - if field == "data": - event.__dict__[field] += value + "\n" - else: - event.__dict__[field] = value - - # Events with no data are not dispatched. - if not event.data: - continue - - # If the data field ends with a newline, remove it. - if event.data.endswith("\n"): - event.data = event.data[0:-1] - - # Empty event names default to 'message' - event.event = event.event or "message" - - # Dispatch the event - self._logger.debug("Dispatching %s...", event) - yield event + field, _p, value = line.partition(":") + if value.startswith(" "): + value = value[1:] + if field == "data": + if value.startswith("[DONE]"): + break + data = json.loads(value) + yield data diff --git a/metagpt/provider/zhipuai/zhipu_model_api.py b/metagpt/provider/zhipuai/zhipu_model_api.py index 16d4102d4..a7d49623a 100644 --- a/metagpt/provider/zhipuai/zhipu_model_api.py +++ b/metagpt/provider/zhipuai/zhipu_model_api.py @@ -4,46 +4,27 @@ import json -import zhipuai -from zhipuai.model_api.api import InvokeType, ModelAPI -from zhipuai.utils.http_client import headers as zhipuai_default_headers +from zhipuai import ZhipuAI +from zhipuai.core._http_client import ZHIPUAI_DEFAULT_TIMEOUT from metagpt.provider.general_api_requestor import GeneralAPIRequestor from metagpt.provider.zhipuai.async_sse_client import AsyncSSEClient -class ZhiPuModelAPI(ModelAPI): - @classmethod - def get_header(cls) -> dict: - token = cls._generate_token() - zhipuai_default_headers.update({"Authorization": token}) - return zhipuai_default_headers - - @classmethod - def get_sse_header(cls) -> dict: - token = cls._generate_token() - headers = {"Authorization": token} - return headers - - @classmethod - def split_zhipu_api_url(cls, invoke_type: InvokeType, kwargs): +class ZhiPuModelAPI(ZhipuAI): + def split_zhipu_api_url(self): # use this method to prevent zhipu api upgrading to different version. # and follow the GeneralAPIRequestor implemented based on openai sdk - zhipu_api_url = cls._build_api_url(kwargs, invoke_type) - """ - example: - zhipu_api_url: https://open.bigmodel.cn/api/paas/v3/model-api/{model}/{invoke_method} - """ + zhipu_api_url = "https://open.bigmodel.cn/api/paas/v4/chat/completions" arr = zhipu_api_url.split("/api/") - # ("https://open.bigmodel.cn/api" , "/paas/v3/model-api/chatglm_turbo/invoke") + # ("https://open.bigmodel.cn/api" , "/paas/v4/chat/completions") return f"{arr[0]}/api", f"/{arr[1]}" - @classmethod - async def arequest(cls, invoke_type: InvokeType, stream: bool, method: str, headers: dict, kwargs): + async def arequest(self, stream: bool, method: str, headers: dict, kwargs): # TODO to make the async request to be more generic for models in http mode. assert method in ["post", "get"] - base_url, url = cls.split_zhipu_api_url(invoke_type, kwargs) + base_url, url = self.split_zhipu_api_url() requester = GeneralAPIRequestor(base_url=base_url) result, _, api_key = await requester.arequest( method=method, @@ -51,25 +32,23 @@ class ZhiPuModelAPI(ModelAPI): headers=headers, stream=stream, params=kwargs, - request_timeout=zhipuai.api_timeout_seconds, + request_timeout=ZHIPUAI_DEFAULT_TIMEOUT.read, ) return result - @classmethod - async def ainvoke(cls, **kwargs) -> dict: + async def acreate(self, **kwargs) -> dict: """async invoke different from raw method `async_invoke` which get the final result by task_id""" - headers = cls.get_header() - resp = await cls.arequest( - invoke_type=InvokeType.SYNC, stream=False, method="post", headers=headers, kwargs=kwargs - ) + headers = self._default_headers + resp = await self.arequest(stream=False, method="post", headers=headers, kwargs=kwargs) resp = resp.decode("utf-8") resp = json.loads(resp) + if "error" in resp: + raise RuntimeError( + f"Request failed, msg: {resp}, please ref to `https://open.bigmodel.cn/dev/api#error-code-v3`" + ) return resp - @classmethod - async def asse_invoke(cls, **kwargs) -> AsyncSSEClient: + async def acreate_stream(self, **kwargs) -> AsyncSSEClient: """async sse_invoke""" - headers = cls.get_sse_header() - return AsyncSSEClient( - await cls.arequest(invoke_type=InvokeType.SSE, stream=True, method="post", headers=headers, kwargs=kwargs) - ) + headers = self._default_headers + return AsyncSSEClient(await self.arequest(stream=True, method="post", headers=headers, kwargs=kwargs)) diff --git a/metagpt/provider/zhipuai_api.py b/metagpt/provider/zhipuai_api.py index e1ccf0de5..9108a1fba 100644 --- a/metagpt/provider/zhipuai_api.py +++ b/metagpt/provider/zhipuai_api.py @@ -2,7 +2,6 @@ # -*- coding: utf-8 -*- # @Desc : zhipuai LLM from https://open.bigmodel.cn/dev/api#sdk -import json from enum import Enum import openai @@ -16,7 +15,7 @@ from tenacity import ( wait_random_exponential, ) -from metagpt.config import CONFIG, LLMProviderEnum +from metagpt.configs.llm_config import LLMConfig, LLMType from metagpt.logs import log_llm_stream, logger from metagpt.provider.base_llm import BaseLLM from metagpt.provider.llm_provider_registry import register_provider @@ -31,57 +30,52 @@ class ZhiPuEvent(Enum): FINISH = "finish" -@register_provider(LLMProviderEnum.ZHIPUAI) +@register_provider(LLMType.ZHIPUAI) class ZhiPuAILLM(BaseLLM): """ Refs to `https://open.bigmodel.cn/dev/api#chatglm_turbo` - From now, there is only one model named `chatglm_turbo` + From now, support glm-3-turbo、glm-4, and also system_prompt. """ - def __init__(self): - self.__init_zhipuai(CONFIG) + def __init__(self, config: LLMConfig): + self.__init_zhipuai(config) self.llm = ZhiPuModelAPI self.model = "chatglm_turbo" # so far only one model, just use it self.use_system_prompt: bool = False # zhipuai has no system prompt when use api + self.config = config - def __init_zhipuai(self, config: CONFIG): - assert config.zhipuai_api_key - zhipuai.api_key = config.zhipuai_api_key + def __init_zhipuai(self, config: LLMConfig): + assert config.api_key + zhipuai.api_key = config.api_key # due to use openai sdk, set the api_key but it will't be used. # openai.api_key = zhipuai.api_key # due to use openai sdk, set the api_key but it will't be used. - if config.openai_proxy: + if config.proxy: # FIXME: openai v1.x sdk has no proxy support - openai.proxy = config.openai_proxy + openai.proxy = config.proxy - def _const_kwargs(self, messages: list[dict]) -> dict: - kwargs = {"model": self.model, "prompt": messages, "temperature": 0.3} + def _const_kwargs(self, messages: list[dict], stream: bool = False) -> dict: + kwargs = {"model": self.model, "messages": messages, "stream": stream, "temperature": 0.3} return kwargs def _update_costs(self, usage: dict): """update each request's token cost""" - if CONFIG.calc_usage: + if self.config.calc_usage: try: prompt_tokens = int(usage.get("prompt_tokens", 0)) completion_tokens = int(usage.get("completion_tokens", 0)) - CONFIG.cost_manager.update_cost(prompt_tokens, completion_tokens, self.model) + self.config.cost_manager.update_cost(prompt_tokens, completion_tokens, self.model) except Exception as e: logger.error(f"zhipuai updats costs failed! exp: {e}") - def get_choice_text(self, resp: dict) -> str: - """get the first text of choice from llm response""" - assist_msg = resp.get("data", {}).get("choices", [{"role": "error"}])[-1] - assert assist_msg["role"] == "assistant" - return assist_msg.get("content") - def completion(self, messages: list[dict], timeout=3) -> dict: - resp = self.llm.invoke(**self._const_kwargs(messages)) - usage = resp.get("data").get("usage") + resp = self.llm.chat.completions.create(**self._const_kwargs(messages)) + usage = resp.usage.model_dump() self._update_costs(usage) - return resp + return resp.model_dump() async def _achat_completion(self, messages: list[dict], timeout=3) -> dict: - resp = await self.llm.ainvoke(**self._const_kwargs(messages)) - usage = resp.get("data").get("usage") + resp = await self.llm.acreate(**self._const_kwargs(messages)) + usage = resp.get("usage", {}) self._update_costs(usage) return resp @@ -89,35 +83,18 @@ class ZhiPuAILLM(BaseLLM): return await self._achat_completion(messages, timeout=timeout) async def _achat_completion_stream(self, messages: list[dict], timeout=3) -> str: - response = await self.llm.asse_invoke(**self._const_kwargs(messages)) + response = await self.llm.acreate_stream(**self._const_kwargs(messages, stream=True)) collected_content = [] usage = {} - async for event in response.async_events(): - if event.event == ZhiPuEvent.ADD.value: - content = event.data + async for chunk in response.stream(): + finish_reason = chunk.get("choices")[0].get("finish_reason") + if finish_reason == "stop": + usage = chunk.get("usage", {}) + else: + content = self.get_choice_delta_text(chunk) collected_content.append(content) log_llm_stream(content) - elif event.event == ZhiPuEvent.ERROR.value or event.event == ZhiPuEvent.INTERRUPTED.value: - content = event.data - logger.error(f"event error: {content}", end="") - elif event.event == ZhiPuEvent.FINISH.value: - """ - event.meta - { - "task_status":"SUCCESS", - "usage":{ - "completion_tokens":351, - "prompt_tokens":595, - "total_tokens":946 - }, - "task_id":"xx", - "request_id":"xxx" - } - """ - meta = json.loads(event.meta) - usage = meta.get("usage") - else: - print(f"zhipuapi else event: {event.data}", end="") + log_llm_stream("\n") self._update_costs(usage) diff --git a/metagpt/roles/architect.py b/metagpt/roles/architect.py index c6ceaccb7..166f8cfd0 100644 --- a/metagpt/roles/architect.py +++ b/metagpt/roles/architect.py @@ -33,7 +33,7 @@ class Architect(Role): def __init__(self, **kwargs) -> None: super().__init__(**kwargs) # Initialize actions specific to the Architect role - self._init_actions([WriteDesign]) + self.set_actions([WriteDesign]) # Set events or actions the Architect should watch or be aware of self._watch({WritePRD}) diff --git a/metagpt/roles/assistant.py b/metagpt/roles/assistant.py index 227578a63..2774bd9b6 100644 --- a/metagpt/roles/assistant.py +++ b/metagpt/roles/assistant.py @@ -22,7 +22,6 @@ from pydantic import Field from metagpt.actions.skill_action import ArgumentsParingAction, SkillAction from metagpt.actions.talk_action import TalkAction -from metagpt.config import CONFIG from metagpt.learn.skill_loader import SkillsDeclaration from metagpt.logs import logger from metagpt.memory.brain_memory import BrainMemory @@ -48,7 +47,8 @@ class Assistant(Role): def __init__(self, **kwargs): super().__init__(**kwargs) - self.constraints = self.constraints.format(language=kwargs.get("language") or CONFIG.language or "Chinese") + language = kwargs.get("language") or self.context.kwargs.language + self.constraints = self.constraints.format(language=language) async def think(self) -> bool: """Everything will be done part by part.""" @@ -56,16 +56,16 @@ class Assistant(Role): if not last_talk: return False if not self.skills: - skill_path = Path(CONFIG.SKILL_PATH) if CONFIG.SKILL_PATH else None + skill_path = Path(self.context.kwargs.SKILL_PATH) if self.context.kwargs.SKILL_PATH else None self.skills = await SkillsDeclaration.load(skill_yaml_file_name=skill_path) prompt = "" - skills = self.skills.get_skill_list() + skills = self.skills.get_skill_list(context=self.context) for desc, name in skills.items(): prompt += f"If the text explicitly want you to {desc}, return `[SKILL]: {name}` brief and clear. For instance: [SKILL]: {name}\n" prompt += 'Otherwise, return `[TALK]: {talk}` brief and clear. For instance: if {talk} is "xxxx" return [TALK]: xxxx\n\n' prompt += f"Now what specific action is explicitly mentioned in the text: {last_talk}\n" - rsp = await self.llm.aask(prompt, []) + rsp = await self.llm.aask(prompt, ["You are an action classifier"]) logger.info(f"THINK: {prompt}\n, THINK RESULT: {rsp}\n") return await self._plan(rsp, last_talk=last_talk) @@ -97,8 +97,8 @@ class Assistant(Role): async def talk_handler(self, text, **kwargs) -> bool: history = self.memory.history_text text = kwargs.get("last_talk") or text - self.rc.todo = TalkAction( - context=text, knowledge=self.memory.get_knowledge(), history_summary=history, llm=self.llm, **kwargs + self.set_todo( + TalkAction(i_context=text, knowledge=self.memory.get_knowledge(), history_summary=history, llm=self.llm) ) return True @@ -108,11 +108,11 @@ class Assistant(Role): if not skill: logger.info(f"skill not found: {text}") return await self.talk_handler(text=last_talk, **kwargs) - action = ArgumentsParingAction(skill=skill, llm=self.llm, ask=last_talk, **kwargs) + action = ArgumentsParingAction(skill=skill, llm=self.llm, ask=last_talk) await action.run(**kwargs) if action.args is None: return await self.talk_handler(text=last_talk, **kwargs) - self.rc.todo = SkillAction(skill=skill, args=action.args, llm=self.llm, name=skill.name, desc=skill.description) + self.set_todo(SkillAction(skill=skill, args=action.args, llm=self.llm, name=skill.name, desc=skill.description)) return True async def refine_memory(self) -> str: diff --git a/metagpt/roles/code_interpreter.py b/metagpt/roles/code_interpreter.py index 3991862d1..b4f9622d3 100644 --- a/metagpt/roles/code_interpreter.py +++ b/metagpt/roles/code_interpreter.py @@ -1,23 +1,16 @@ -from datetime import datetime - from pydantic import Field from metagpt.actions.ask_review import ReviewConst from metagpt.actions.execute_code import ExecutePyCode from metagpt.actions.write_analysis_code import WriteCodeByGenerate, WriteCodeWithTools -from metagpt.actions.write_code_steps import WriteCodeSteps from metagpt.logs import logger from metagpt.roles import Role -from metagpt.roles.tool_maker import ToolMaker from metagpt.schema import Message, Task, TaskResult -from metagpt.utils.save_code import save_code_file class CodeInterpreter(Role): auto_run: bool = True use_tools: bool = False - make_udfs: bool = False # whether to save user-defined functions - use_code_steps: bool = False execute_code: ExecutePyCode = Field(default_factory=ExecutePyCode, exclude=True) tools: list[str] = [] @@ -47,29 +40,12 @@ class CodeInterpreter(Role): def working_memory(self): return self.rc.working_memory - async def _plan_and_act(self): - rsp = await super()._plan_and_act() - - # save code using datetime.now or keywords related to the goal of your project (plan.goal). - project_record = datetime.now().strftime("%Y-%m-%d_%H-%M-%S") - save_code_file(name=project_record, code_context=self.execute_code.nb, file_format="ipynb") - - # make tools out of workable codes for future use - if self.make_udfs: - await self.make_tools() - - return rsp - async def _act_on_task(self, current_task: Task) -> TaskResult: code, result, is_success = await self._write_and_exec_code() task_result = TaskResult(code=code, result=result, is_success=is_success) return task_result async def _write_and_exec_code(self, max_retry: int = 3): - self.planner.current_task.code_steps = ( - await WriteCodeSteps().run(self.planner.plan) if self.use_code_steps else "" - ) - counter = 0 success = False @@ -97,7 +73,11 @@ class CodeInterpreter(Role): if ReviewConst.CHANGE_WORD[0] in review: counter = 0 # redo the task again with help of human suggestions - return code["code"] if code["language"] != "markdown" else "", result, success + py_code = ( + code["code"] if code.get("language") != "markdown" else "" + ) # use python code as final code; for markdown, return the rendered result instead of the code itself + + return py_code, result, success async def _write_code(self): todo = WriteCodeByGenerate() if not self.use_tools else WriteCodeWithTools(selected_tools=self.tools) @@ -108,12 +88,3 @@ class CodeInterpreter(Role): code = await todo.run(context=context, plan=self.planner.plan, temperature=0.0) return code, todo - - async def make_tools(self): - """Make user-defined functions(udfs, aka tools) for pure generation code.""" - logger.info("Plan completed. Now start to make tools ...") - tool_maker = ToolMaker() - for task in self.planner.plan.get_finished_tasks(): - await tool_maker.make_tool( - code=task.code, instruction=task.instruction, task_id=task.task_id, auto_run=self.auto_run - ) diff --git a/metagpt/roles/engineer.py b/metagpt/roles/engineer.py index e05e69cbb..40ade2110 100644 --- a/metagpt/roles/engineer.py +++ b/metagpt/roles/engineer.py @@ -20,23 +20,27 @@ from __future__ import annotations import json +import os from collections import defaultdict from pathlib import Path from typing import Set from metagpt.actions import Action, WriteCode, WriteCodeReview, WriteTasks from metagpt.actions.fix_bug import FixBug +from metagpt.actions.project_management_an import REFINED_TASK_LIST, TASK_LIST from metagpt.actions.summarize_code import SummarizeCode -from metagpt.config import CONFIG +from metagpt.actions.write_code_plan_and_change_an import WriteCodePlanAndChange from metagpt.const import ( - CODE_SUMMARIES_FILE_REPO, - CODE_SUMMARIES_PDF_FILE_REPO, + CODE_PLAN_AND_CHANGE_FILE_REPO, + CODE_PLAN_AND_CHANGE_FILENAME, + REQUIREMENT_FILENAME, SYSTEM_DESIGN_FILE_REPO, TASK_FILE_REPO, ) from metagpt.logs import logger from metagpt.roles import Role from metagpt.schema import ( + CodePlanAndChangeContext, CodeSummarizeContext, CodingContext, Document, @@ -80,12 +84,13 @@ class Engineer(Role): code_todos: list = [] summarize_todos: list = [] next_todo_action: str = "" + n_summarize: int = 0 def __init__(self, **kwargs) -> None: super().__init__(**kwargs) - self._init_actions([WriteCode]) - self._watch([WriteTasks, SummarizeCode, WriteCode, WriteCodeReview, FixBug]) + self.set_actions([WriteCode]) + self._watch([WriteTasks, SummarizeCode, WriteCode, WriteCodeReview, FixBug, WriteCodePlanAndChange]) self.code_todos = [] self.summarize_todos = [] self.next_todo_action = any_to_name(WriteCode) @@ -93,11 +98,10 @@ class Engineer(Role): @staticmethod def _parse_tasks(task_msg: Document) -> list[str]: m = json.loads(task_msg.content) - return m.get("Task list") + return m.get(TASK_LIST.key) or m.get(REFINED_TASK_LIST.key) async def _act_sp_with_cr(self, review=False) -> Set[str]: changed_files = set() - src_file_repo = CONFIG.git_repo.new_file_repository(CONFIG.src_workspace) for todo in self.code_todos: """ # Select essential information from the historical data to reduce the length of the prompt (summarized from human experience): @@ -109,12 +113,16 @@ class Engineer(Role): coding_context = await todo.run() # Code review if review: - action = WriteCodeReview(context=coding_context, llm=self.llm) - self._init_action_system_message(action) + action = WriteCodeReview(i_context=coding_context, context=self.context, llm=self.llm) + self._init_action(action) coding_context = await action.run() - await src_file_repo.save( - coding_context.filename, - dependencies={coding_context.design_doc.root_relative_path, coding_context.task_doc.root_relative_path}, + + dependencies = {coding_context.design_doc.root_relative_path, coding_context.task_doc.root_relative_path} + if self.config.inc: + dependencies.add(os.path.join(CODE_PLAN_AND_CHANGE_FILE_REPO, CODE_PLAN_AND_CHANGE_FILENAME)) + await self.project_repo.srcs.save( + filename=coding_context.filename, + dependencies=dependencies, content=coding_context.code_doc.content, ) msg = Message( @@ -134,6 +142,9 @@ class Engineer(Role): """Determines the mode of action based on whether code review is used.""" if self.rc.todo is None: return None + if isinstance(self.rc.todo, WriteCodePlanAndChange): + self.next_todo_action = any_to_name(WriteCode) + return await self._act_code_plan_and_change() if isinstance(self.rc.todo, WriteCode): self.next_todo_action = any_to_name(SummarizeCode) return await self._act_write_code() @@ -153,34 +164,32 @@ class Engineer(Role): ) async def _act_summarize(self): - code_summaries_file_repo = CONFIG.git_repo.new_file_repository(CODE_SUMMARIES_FILE_REPO) - code_summaries_pdf_file_repo = CONFIG.git_repo.new_file_repository(CODE_SUMMARIES_PDF_FILE_REPO) tasks = [] - src_relative_path = CONFIG.src_workspace.relative_to(CONFIG.git_repo.workdir) for todo in self.summarize_todos: summary = await todo.run() - summary_filename = Path(todo.context.design_filename).with_suffix(".md").name - dependencies = {todo.context.design_filename, todo.context.task_filename} - for filename in todo.context.codes_filenames: - rpath = src_relative_path / filename + summary_filename = Path(todo.i_context.design_filename).with_suffix(".md").name + dependencies = {todo.i_context.design_filename, todo.i_context.task_filename} + for filename in todo.i_context.codes_filenames: + rpath = self.project_repo.src_relative_path / filename dependencies.add(str(rpath)) - await code_summaries_pdf_file_repo.save( + await self.project_repo.resources.code_summary.save( filename=summary_filename, content=summary, dependencies=dependencies ) is_pass, reason = await self._is_pass(summary) if not is_pass: - todo.context.reason = reason - tasks.append(todo.context.dict()) - await code_summaries_file_repo.save( - filename=Path(todo.context.design_filename).name, - content=todo.context.model_dump_json(), + todo.i_context.reason = reason + tasks.append(todo.i_context.model_dump()) + + await self.project_repo.docs.code_summary.save( + filename=Path(todo.i_context.design_filename).name, + content=todo.i_context.model_dump_json(), dependencies=dependencies, ) else: - await code_summaries_file_repo.delete(filename=Path(todo.context.design_filename).name) + await self.project_repo.docs.code_summary.delete(filename=Path(todo.i_context.design_filename).name) - logger.info(f"--max-auto-summarize-code={CONFIG.max_auto_summarize_code}") - if not tasks or CONFIG.max_auto_summarize_code == 0: + logger.info(f"--max-auto-summarize-code={self.config.max_auto_summarize_code}") + if not tasks or self.config.max_auto_summarize_code == 0: return Message( content="", role=self.profile, @@ -190,11 +199,39 @@ class Engineer(Role): ) # The maximum number of times the 'SummarizeCode' action is automatically invoked, with -1 indicating unlimited. # This parameter is used for debugging the workflow. - CONFIG.max_auto_summarize_code -= 1 if CONFIG.max_auto_summarize_code > 0 else 0 + self.n_summarize += 1 if self.config.max_auto_summarize_code > self.n_summarize else 0 return Message( content=json.dumps(tasks), role=self.profile, cause_by=SummarizeCode, send_to=self, sent_from=self ) + async def _act_code_plan_and_change(self): + """Write code plan and change that guides subsequent WriteCode and WriteCodeReview""" + logger.info("Writing code plan and change..") + node = await self.rc.todo.run() + code_plan_and_change = node.instruct_content.model_dump_json() + dependencies = { + REQUIREMENT_FILENAME, + self.rc.todo.i_context.prd_filename, + self.rc.todo.i_context.design_filename, + self.rc.todo.i_context.task_filename, + } + await self.project_repo.docs.code_plan_and_change.save( + filename=self.rc.todo.i_context.filename, content=code_plan_and_change, dependencies=dependencies + ) + await self.project_repo.resources.code_plan_and_change.save( + filename=Path(self.rc.todo.i_context.filename).with_suffix(".md").name, + content=node.content, + dependencies=dependencies, + ) + + return Message( + content=code_plan_and_change, + role=self.profile, + cause_by=WriteCodePlanAndChange, + send_to=self, + sent_from=self, + ) + async def _is_pass(self, summary) -> (str, str): rsp = await self.llm.aask(msg=IS_PASS_PROMPT.format(context=summary), stream=False) logger.info(rsp) @@ -203,13 +240,18 @@ class Engineer(Role): return False, rsp async def _think(self) -> Action | None: - if not CONFIG.src_workspace: - CONFIG.src_workspace = CONFIG.git_repo.workdir / CONFIG.git_repo.workdir.name - write_code_filters = any_to_str_set([WriteTasks, SummarizeCode, FixBug]) + if not self.src_workspace: + self.src_workspace = self.git_repo.workdir / self.git_repo.workdir.name + write_plan_and_change_filters = any_to_str_set([WriteTasks]) + write_code_filters = any_to_str_set([WriteTasks, WriteCodePlanAndChange, SummarizeCode, FixBug]) summarize_code_filters = any_to_str_set([WriteCode, WriteCodeReview]) if not self.rc.news: return None msg = self.rc.news[0] + if self.config.inc and msg.cause_by in write_plan_and_change_filters: + logger.debug(f"TODO WriteCodePlanAndChange:{msg.model_dump_json()}") + await self._new_code_plan_and_change_action() + return self.rc.todo if msg.cause_by in write_code_filters: logger.debug(f"TODO WriteCode:{msg.model_dump_json()}") await self._new_code_actions(bug_fix=msg.cause_by == any_to_str(FixBug)) @@ -220,60 +262,54 @@ class Engineer(Role): return self.rc.todo return None - @staticmethod - async def _new_coding_context( - filename, src_file_repo, task_file_repo, design_file_repo, dependency - ) -> CodingContext: - old_code_doc = await src_file_repo.get(filename) + async def _new_coding_context(self, filename, dependency) -> CodingContext: + old_code_doc = await self.project_repo.srcs.get(filename) if not old_code_doc: - old_code_doc = Document(root_path=str(src_file_repo.root_path), filename=filename, content="") + old_code_doc = Document(root_path=str(self.project_repo.src_relative_path), filename=filename, content="") dependencies = {Path(i) for i in await dependency.get(old_code_doc.root_relative_path)} task_doc = None design_doc = None for i in dependencies: if str(i.parent) == TASK_FILE_REPO: - task_doc = await task_file_repo.get(i.name) + task_doc = await self.project_repo.docs.task.get(i.name) elif str(i.parent) == SYSTEM_DESIGN_FILE_REPO: - design_doc = await design_file_repo.get(i.name) + design_doc = await self.project_repo.docs.system_design.get(i.name) if not task_doc or not design_doc: logger.error(f'Detected source code "{filename}" from an unknown origin.') raise ValueError(f'Detected source code "{filename}" from an unknown origin.') context = CodingContext(filename=filename, design_doc=design_doc, task_doc=task_doc, code_doc=old_code_doc) return context - @staticmethod - async def _new_coding_doc(filename, src_file_repo, task_file_repo, design_file_repo, dependency): - context = await Engineer._new_coding_context( - filename, src_file_repo, task_file_repo, design_file_repo, dependency - ) + async def _new_coding_doc(self, filename, dependency): + context = await self._new_coding_context(filename, dependency) coding_doc = Document( - root_path=str(src_file_repo.root_path), filename=filename, content=context.model_dump_json() + root_path=str(self.project_repo.src_relative_path), filename=filename, content=context.model_dump_json() ) return coding_doc async def _new_code_actions(self, bug_fix=False): # Prepare file repos - src_file_repo = CONFIG.git_repo.new_file_repository(CONFIG.src_workspace) - changed_src_files = src_file_repo.all_files if bug_fix else src_file_repo.changed_files - task_file_repo = CONFIG.git_repo.new_file_repository(TASK_FILE_REPO) - changed_task_files = task_file_repo.changed_files - design_file_repo = CONFIG.git_repo.new_file_repository(SYSTEM_DESIGN_FILE_REPO) - + changed_src_files = self.project_repo.srcs.all_files if bug_fix else self.project_repo.srcs.changed_files + changed_task_files = self.project_repo.docs.task.changed_files changed_files = Documents() # Recode caused by upstream changes. for filename in changed_task_files: - design_doc = await design_file_repo.get(filename) - task_doc = await task_file_repo.get(filename) + design_doc = await self.project_repo.docs.system_design.get(filename) + task_doc = await self.project_repo.docs.task.get(filename) task_list = self._parse_tasks(task_doc) for task_filename in task_list: - old_code_doc = await src_file_repo.get(task_filename) + old_code_doc = await self.project_repo.srcs.get(task_filename) if not old_code_doc: - old_code_doc = Document(root_path=str(src_file_repo.root_path), filename=task_filename, content="") + old_code_doc = Document( + root_path=str(self.project_repo.src_relative_path), filename=task_filename, content="" + ) context = CodingContext( filename=task_filename, design_doc=design_doc, task_doc=task_doc, code_doc=old_code_doc ) coding_doc = Document( - root_path=str(src_file_repo.root_path), filename=task_filename, content=context.model_dump_json() + root_path=str(self.project_repo.src_relative_path), + filename=task_filename, + content=context.model_dump_json(), ) if task_filename in changed_files.docs: logger.warning( @@ -281,41 +317,44 @@ class Engineer(Role): f"{changed_files.docs[task_filename].model_dump_json()}" ) changed_files.docs[task_filename] = coding_doc - self.code_todos = [WriteCode(context=i, llm=self.llm) for i in changed_files.docs.values()] + self.code_todos = [ + WriteCode(i_context=i, context=self.context, llm=self.llm) for i in changed_files.docs.values() + ] # Code directly modified by the user. - dependency = await CONFIG.git_repo.get_dependency() + dependency = await self.git_repo.get_dependency() for filename in changed_src_files: if filename in changed_files.docs: continue - coding_doc = await self._new_coding_doc( - filename=filename, - src_file_repo=src_file_repo, - task_file_repo=task_file_repo, - design_file_repo=design_file_repo, - dependency=dependency, - ) + coding_doc = await self._new_coding_doc(filename=filename, dependency=dependency) changed_files.docs[filename] = coding_doc - self.code_todos.append(WriteCode(context=coding_doc, llm=self.llm)) + self.code_todos.append(WriteCode(i_context=coding_doc, context=self.context, llm=self.llm)) if self.code_todos: - self.rc.todo = self.code_todos[0] + self.set_todo(self.code_todos[0]) async def _new_summarize_actions(self): - src_file_repo = CONFIG.git_repo.new_file_repository(CONFIG.src_workspace) - src_files = src_file_repo.all_files + src_files = self.project_repo.srcs.all_files # Generate a SummarizeCode action for each pair of (system_design_doc, task_doc). summarizations = defaultdict(list) for filename in src_files: - dependencies = await src_file_repo.get_dependency(filename=filename) - ctx = CodeSummarizeContext.loads(filenames=dependencies) + dependencies = await self.project_repo.srcs.get_dependency(filename=filename) + ctx = CodeSummarizeContext.loads(filenames=list(dependencies)) summarizations[ctx].append(filename) for ctx, filenames in summarizations.items(): ctx.codes_filenames = filenames - self.summarize_todos.append(SummarizeCode(context=ctx, llm=self.llm)) + self.summarize_todos.append(SummarizeCode(i_context=ctx, context=self.context, llm=self.llm)) if self.summarize_todos: - self.rc.todo = self.summarize_todos[0] + self.set_todo(self.summarize_todos[0]) + + async def _new_code_plan_and_change_action(self): + """Create a WriteCodePlanAndChange action for subsequent to-do actions.""" + files = self.project_repo.all_files + requirement_doc = await self.project_repo.docs.get(REQUIREMENT_FILENAME) + requirement = requirement_doc.content if requirement_doc else "" + code_plan_and_change_ctx = CodePlanAndChangeContext.loads(files, requirement=requirement) + self.rc.todo = WriteCodePlanAndChange(i_context=code_plan_and_change_ctx, context=self.context, llm=self.llm) @property - def todo(self) -> str: + def action_description(self) -> str: """AgentStore uses this attribute to display to the user what actions the current role should take.""" return self.next_todo_action diff --git a/metagpt/roles/invoice_ocr_assistant.py b/metagpt/roles/invoice_ocr_assistant.py index f5588974b..a39a48b97 100644 --- a/metagpt/roles/invoice_ocr_assistant.py +++ b/metagpt/roles/invoice_ocr_assistant.py @@ -60,7 +60,7 @@ class InvoiceOCRAssistant(Role): def __init__(self, **kwargs): super().__init__(**kwargs) - self._init_actions([InvoiceOCR]) + self.set_actions([InvoiceOCR]) self._set_react_mode(react_mode=RoleReactMode.BY_ORDER.value) async def _act(self) -> Message: @@ -82,12 +82,12 @@ class InvoiceOCRAssistant(Role): resp = await todo.run(file_path) if len(resp) == 1: # Single file support for questioning based on OCR recognition results - self._init_actions([GenerateTable, ReplyQuestion]) + self.set_actions([GenerateTable, ReplyQuestion]) self.orc_data = resp[0] else: - self._init_actions([GenerateTable]) + self.set_actions([GenerateTable]) - self.rc.todo = None + self.set_todo(None) content = INVOICE_OCR_SUCCESS resp = OCRResults(ocr_result=json.dumps(resp)) msg = Message(content=content, instruct_content=resp) diff --git a/metagpt/roles/kaggle_manager.py b/metagpt/roles/kaggle_manager.py deleted file mode 100644 index 3ef573a8c..000000000 --- a/metagpt/roles/kaggle_manager.py +++ /dev/null @@ -1,153 +0,0 @@ -import json -import os -import subprocess - -import fire -import pandas as pd - -from metagpt.actions import Action, UserRequirement -from metagpt.actions.ml_da_action import SummarizeAnalysis -from metagpt.config import CONFIG -from metagpt.logs import logger -from metagpt.roles import Role -from metagpt.schema import Message -from metagpt.utils.common import CodeParser - -os.environ["KAGGLE_USERNAME"] = CONFIG.kaggle_username -os.environ["KAGGLE_KEY"] = CONFIG.kaggle_key - - -def run_command(cmd): - print(cmd) - output = subprocess.run(cmd, shell=True, capture_output=True, text=True) - if output.returncode != 0: - print("Error output:", output.stderr) - exit() - else: - print(output.stdout) - return output.stdout - - -class DownloadData(Action): - async def run(self, competition, data_desc="") -> str: - data_path = CONFIG.workspace_path / competition - - output = run_command(f"kaggle competitions list --search {competition}") - assert output != "No competitions found", "You must provide the correct competition name" - - run_command(f"kaggle competitions download {competition} --path {WORKSPACE_ROOT}") - - if not os.path.exists(data_path): - # if True: - # run_command(f"rm -r {data_path / '*'}") - run_command(f"unzip -o {CONFIG.workspace_path / '*.zip'} -d {data_path}") # FIXME: not safe - - file_list = run_command(f"ls {data_path}") - - rsp = f""" - Location: - Data downloaded at {data_path} folder, including {file_list} - Data Description: - {data_desc} - """ - return rsp - - -class SubmitResult(Action): - PROMPT_TEMPLATE: str = """ - # Summary - __summary__ - # Your task - Extract the file path for test set prediction from the summary above, output a json following the format: - ```json - {"file_path": str = "the file path, for example, /path/to/the/prediction/file/xxx.csv, /path/to/the/prediction/file/xxx.xlsx"} - ``` - """ - - def __init__(self, name: str = "", context=None, llm=None) -> str: - super().__init__(name, context, llm) - - async def _parse_submit_file_path(self, context) -> str: - prompt = self.PROMPT_TEMPLATE.replace("__summary__", context) - rsp = await self._aask(prompt) - rsp = CodeParser.parse_code(block=None, text=rsp) - file_path = json.loads(rsp)["file_path"] - return file_path - - async def run(self, competition, submit_message="") -> str: - submit_file_path = await self._parse_submit_file_path(submit_message) - - data_path = CONFIG.workspace_path / competition - submit_message = submit_message.replace("'", "") - - run_command(f"kaggle competitions submit {competition} -f {submit_file_path} -m '{submit_message}'") - run_command(f"kaggle competitions leaderboard --show --csv {competition} > {data_path / 'leaderboard.csv'}") - run_command(f"kaggle competitions submissions --csv {competition} > {data_path / 'submission.csv'}") - - leaderboard = pd.read_csv(data_path / "leaderboard.csv") - submission = pd.read_csv(data_path / "submission.csv") - print(submission) # submission.to_json(orient="records") - - submission_score = submission.loc[0, "publicScore"] - best_score = max(submission["publicScore"]) # might be min - rank = leaderboard.loc[leaderboard["score"] == best_score].index[0] - rank_pct = round(rank / len(leaderboard), 4) * 100 - - submission_summary = f""" - # All histories: - {submission.head(5).to_string()} - # Current - Current submission score: {submission_score}, best score: {best_score}, best rank: {rank} (top {rank_pct}%) - """ - logger.info(submission_summary) - return submission_summary - - -class KaggleManager(Role): - def __init__(self, name="ABC", profile="KaggleManager", goal="", competition="titanic", data_desc=""): - super().__init__(name=name, profile=profile, goal=goal) - self._init_actions([DownloadData, SubmitResult]) - self._watch([UserRequirement, SummarizeAnalysis]) - self.competition = competition - self.data_desc = data_desc # currently passed in, later can be scrapped down from web by another Role - - async def _think(self): - observed = self.get_memories()[-1].cause_by - if observed == UserRequirement: - self._set_state(0) # DownloadData, get competition of interest from human, download datasets - elif observed == SummarizeAnalysis: - self._set_state(1) # SubmitResult, get prediction from MLEngineer and submit it to Kaggle - - async def _act(self): - todo = self.rc.todo - logger.info(f"{self._setting}: ready to {self.rc.todo}") - - if isinstance(todo, DownloadData): - rsp = await todo.run(self.competition, self.data_desc) - - elif isinstance(todo, SubmitResult): - submit_message = self.get_memories()[ - -1 - ].content # use analysis summary from MLEngineer as submission message - rsp = await todo.run(competition=self.competition, submit_message=submit_message) - - msg = Message(content=rsp, role="user", cause_by=type(todo)) - - return msg - - -if __name__ == "__main__": - competition, data_desc, requirement = ( - "titanic", - "Training set is train.csv.\nTest set is test.csv. We also include gender_submission.csv, a set of predictions that assume all and only female passengers survive, as an example of what a submission file should look like.", - "Run EDA on the train dataset, train a model to predict survival (20% as validation) and save it, predict the test set using saved model, save the test result according to format", - ) - - summary = "I used Python with pandas for data preprocessing, sklearn's RandomForestClassifier for modeling, and achieved 82.12% accuracy on validation. Predictions saved at '/Users/gary/Desktop/data_agents_opt/workspace/titanic/gender_submission.csv'." - - async def main(requirement: str = requirement): - role = KaggleManager(competition=competition, data_desc=data_desc) - # await role.run(Message(content="", cause_by=UserRequirement)) - await role.run(Message(content=summary, cause_by=SummarizeAnalysis)) - - fire.Fire(main) diff --git a/metagpt/roles/ml_engineer.py b/metagpt/roles/ml_engineer.py index d1a22b9d3..e7abee560 100644 --- a/metagpt/roles/ml_engineer.py +++ b/metagpt/roles/ml_engineer.py @@ -1,7 +1,6 @@ from metagpt.actions.debug_code import DebugCode from metagpt.actions.execute_code import ExecutePyCode -from metagpt.actions.ml_da_action import UpdateDataColumns -from metagpt.actions.write_analysis_code import WriteCodeWithToolsML +from metagpt.actions.ml_action import UpdateDataColumns, WriteCodeWithToolsML from metagpt.logs import logger from metagpt.roles.code_interpreter import CodeInterpreter from metagpt.tools.tool_data_type import ToolTypeEnum diff --git a/metagpt/roles/product_manager.py b/metagpt/roles/product_manager.py index 1d82ac3f2..fbe139a99 100644 --- a/metagpt/roles/product_manager.py +++ b/metagpt/roles/product_manager.py @@ -9,7 +9,6 @@ from metagpt.actions import UserRequirement, WritePRD from metagpt.actions.prepare_documents import PrepareDocuments -from metagpt.config import CONFIG from metagpt.roles.role import Role from metagpt.utils.common import any_to_name @@ -34,24 +33,19 @@ class ProductManager(Role): def __init__(self, **kwargs) -> None: super().__init__(**kwargs) - self._init_actions([PrepareDocuments, WritePRD]) + self.set_actions([PrepareDocuments, WritePRD]) self._watch([UserRequirement, PrepareDocuments]) self.todo_action = any_to_name(PrepareDocuments) async def _think(self) -> bool: """Decide what to do""" - if CONFIG.git_repo and not CONFIG.git_reinit: + if self.git_repo and not self.config.git_reinit: self._set_state(1) else: self._set_state(0) - CONFIG.git_reinit = False + self.config.git_reinit = False self.todo_action = any_to_name(WritePRD) return bool(self.rc.todo) async def _observe(self, ignore_memory=False) -> int: return await super()._observe(ignore_memory=True) - - @property - def todo(self) -> str: - """AgentStore uses this attribute to display to the user what actions the current role should take.""" - return self.todo_action diff --git a/metagpt/roles/project_manager.py b/metagpt/roles/project_manager.py index 1fad4afc2..422d2889b 100644 --- a/metagpt/roles/project_manager.py +++ b/metagpt/roles/project_manager.py @@ -33,5 +33,5 @@ class ProjectManager(Role): def __init__(self, **kwargs) -> None: super().__init__(**kwargs) - self._init_actions([WriteTasks]) + self.set_actions([WriteTasks]) self._watch([WriteDesign]) diff --git a/metagpt/roles/qa_engineer.py b/metagpt/roles/qa_engineer.py index b1d06d122..c73c10ef3 100644 --- a/metagpt/roles/qa_engineer.py +++ b/metagpt/roles/qa_engineer.py @@ -15,20 +15,13 @@ of SummarizeCode. """ - from metagpt.actions import DebugError, RunCode, WriteTest from metagpt.actions.summarize_code import SummarizeCode -from metagpt.config import CONFIG -from metagpt.const import ( - MESSAGE_ROUTE_TO_NONE, - TEST_CODES_FILE_REPO, - TEST_OUTPUTS_FILE_REPO, -) +from metagpt.const import MESSAGE_ROUTE_TO_NONE from metagpt.logs import logger from metagpt.roles import Role from metagpt.schema import Document, Message, RunCodeContext, TestingContext from metagpt.utils.common import any_to_str_set, parse_recipient -from metagpt.utils.file_repository import FileRepository class QaEngineer(Role): @@ -36,7 +29,8 @@ class QaEngineer(Role): profile: str = "QaEngineer" goal: str = "Write comprehensive and robust tests to ensure codes will work as expected without bugs" constraints: str = ( - "The test code you write should conform to code standard like PEP8, be modular, " "easy to read and maintain" + "The test code you write should conform to code standard like PEP8, be modular, easy to read and maintain." + "Use same language as user requirement" ) test_round_allowed: int = 5 test_round: int = 0 @@ -46,34 +40,35 @@ class QaEngineer(Role): # FIXME: a bit hack here, only init one action to circumvent _think() logic, # will overwrite _think() in future updates - self._init_actions([WriteTest]) + self.set_actions([WriteTest]) self._watch([SummarizeCode, WriteTest, RunCode, DebugError]) self.test_round = 0 async def _write_test(self, message: Message) -> None: - src_file_repo = CONFIG.git_repo.new_file_repository(CONFIG.src_workspace) + src_file_repo = self.project_repo.with_src_path(self.context.src_workspace).srcs changed_files = set(src_file_repo.changed_files.keys()) # Unit tests only. - if CONFIG.reqa_file and CONFIG.reqa_file not in changed_files: - changed_files.add(CONFIG.reqa_file) - tests_file_repo = CONFIG.git_repo.new_file_repository(TEST_CODES_FILE_REPO) + if self.config.reqa_file and self.config.reqa_file not in changed_files: + changed_files.add(self.config.reqa_file) for filename in changed_files: # write tests if not filename or "test" in filename: continue code_doc = await src_file_repo.get(filename) - test_doc = await tests_file_repo.get("test_" + code_doc.filename) + if not code_doc: + continue + if not code_doc.filename.endswith(".py"): + continue + test_doc = await self.project_repo.tests.get("test_" + code_doc.filename) if not test_doc: test_doc = Document( - root_path=str(tests_file_repo.root_path), filename="test_" + code_doc.filename, content="" + root_path=str(self.project_repo.tests.root_path), filename="test_" + code_doc.filename, content="" ) logger.info(f"Writing {test_doc.filename}..") context = TestingContext(filename=test_doc.filename, test_doc=test_doc, code_doc=code_doc) - context = await WriteTest(context=context, llm=self.llm).run() - await tests_file_repo.save( - filename=context.test_doc.filename, - content=context.test_doc.content, - dependencies={context.code_doc.root_relative_path}, + context = await WriteTest(i_context=context, context=self.context, llm=self.llm).run() + await self.project_repo.tests.save_doc( + doc=context.test_doc, dependencies={context.code_doc.root_relative_path} ) # prepare context for run tests in next round @@ -81,8 +76,8 @@ class QaEngineer(Role): command=["python", context.test_doc.root_relative_path], code_filename=context.code_doc.filename, test_filename=context.test_doc.filename, - working_directory=str(CONFIG.git_repo.workdir), - additional_python_paths=[str(CONFIG.src_workspace)], + working_directory=str(self.project_repo.workdir), + additional_python_paths=[str(self.context.src_workspace)], ) self.publish_message( Message( @@ -94,21 +89,23 @@ class QaEngineer(Role): ) ) - logger.info(f"Done {str(tests_file_repo.workdir)} generating.") + logger.info(f"Done {str(self.project_repo.tests.workdir)} generating.") async def _run_code(self, msg): run_code_context = RunCodeContext.loads(msg.content) - src_doc = await CONFIG.git_repo.new_file_repository(CONFIG.src_workspace).get(run_code_context.code_filename) + src_doc = await self.project_repo.with_src_path(self.context.src_workspace).srcs.get( + run_code_context.code_filename + ) if not src_doc: return - test_doc = await CONFIG.git_repo.new_file_repository(TEST_CODES_FILE_REPO).get(run_code_context.test_filename) + test_doc = await self.project_repo.tests.get(run_code_context.test_filename) if not test_doc: return run_code_context.code = src_doc.content run_code_context.test_code = test_doc.content - result = await RunCode(context=run_code_context, llm=self.llm).run() + result = await RunCode(i_context=run_code_context, context=self.context, llm=self.llm).run() run_code_context.output_filename = run_code_context.test_filename + ".json" - await CONFIG.git_repo.new_file_repository(TEST_OUTPUTS_FILE_REPO).save( + await self.project_repo.test_outputs.save( filename=run_code_context.output_filename, content=result.model_dump_json(), dependencies={src_doc.root_relative_path, test_doc.root_relative_path}, @@ -130,10 +127,8 @@ class QaEngineer(Role): async def _debug_error(self, msg): run_code_context = RunCodeContext.loads(msg.content) - code = await DebugError(context=run_code_context, llm=self.llm).run() - await FileRepository.save_file( - filename=run_code_context.test_filename, content=code, relative_path=TEST_CODES_FILE_REPO - ) + code = await DebugError(i_context=run_code_context, context=self.context, llm=self.llm).run() + await self.project_repo.tests.save(filename=run_code_context.test_filename, content=code) run_code_context.output = None self.publish_message( Message( diff --git a/metagpt/roles/researcher.py b/metagpt/roles/researcher.py index 15f6c9a22..137cfdb4c 100644 --- a/metagpt/roles/researcher.py +++ b/metagpt/roles/researcher.py @@ -34,7 +34,7 @@ class Researcher(Role): def __init__(self, **kwargs): super().__init__(**kwargs) - self._init_actions( + self.set_actions( [CollectLinks(name=self.name), WebBrowseAndSummarize(name=self.name), ConductResearch(name=self.name)] ) self._set_react_mode(react_mode=RoleReactMode.BY_ORDER.value) @@ -49,7 +49,7 @@ class Researcher(Role): if self.rc.state + 1 < len(self.states): self._set_state(self.rc.state + 1) else: - self.rc.todo = None + self.set_todo(None) return False async def _act(self) -> Message: diff --git a/metagpt/roles/role.py b/metagpt/roles/role.py index 21e48a127..9efcf470e 100644 --- a/metagpt/roles/role.py +++ b/metagpt/roles/role.py @@ -23,29 +23,21 @@ from __future__ import annotations from enum import Enum -from pathlib import Path -from typing import Any, Iterable, Optional, Set, Type +from typing import Any, Iterable, Optional, Set, Type, Union from pydantic import BaseModel, ConfigDict, Field, SerializeAsAny, model_validator from metagpt.actions import Action, ActionOutput from metagpt.actions.action_node import ActionNode from metagpt.actions.add_requirement import UserRequirement -from metagpt.const import SERDESER_PATH -from metagpt.llm import LLM, HumanProvider +from metagpt.context_mixin import ContextMixin from metagpt.logs import logger from metagpt.memory import Memory from metagpt.plan.planner import Planner -from metagpt.provider.base_llm import BaseLLM -from metagpt.schema import Message, MessageQueue, SerializationMixin, Task, TaskResult -from metagpt.utils.common import ( - any_to_name, - any_to_str, - import_class, - read_json_file, - role_raise_decorator, - write_json_file, -) +from metagpt.provider import HumanProvider +from metagpt.schema import Message, MessageQueue, SerializationMixin +from metagpt.utils.common import any_to_name, any_to_str, role_raise_decorator +from metagpt.utils.project_repo import ProjectRepo from metagpt.utils.repair_llm_raw_output import extract_state_value_from_output PREFIX_TEMPLATE = """You are a {profile}, named {name}, your goal is {goal}. """ @@ -113,7 +105,7 @@ class RoleContext(BaseModel): max_react_loop: int = 1 def check(self, role_id: str): - # if hasattr(CONFIG, "long_term_memory") and CONFIG.long_term_memory: + # if hasattr(CONFIG, "enable_longterm_memory") and CONFIG.enable_longterm_memory: # self.long_term_memory.recover_memory(role_id, self) # self.memory = self.long_term_memory # use memory to act as long_term_memory for unify operation pass @@ -128,10 +120,10 @@ class RoleContext(BaseModel): return self.memory.get() -class Role(SerializationMixin, is_polymorphic_base=True): +class Role(SerializationMixin, ContextMixin, BaseModel): """Role/Agent""" - model_config = ConfigDict(arbitrary_types_allowed=True, exclude=["llm"]) + model_config = ConfigDict(arbitrary_types_allowed=True, extra="ignore") name: str = "" profile: str = "" @@ -140,13 +132,19 @@ class Role(SerializationMixin, is_polymorphic_base=True): desc: str = "" is_human: bool = False - llm: BaseLLM = Field(default_factory=LLM, exclude=True) # Each role has its own LLM, use different system message role_id: str = "" states: list[str] = [] + + # scenarios to set action system_prompt: + # 1. `__init__` while using Role(actions=[...]) + # 2. add action to role while using `role.set_action(action)` + # 3. set_todo while using `role.set_todo(action)` + # 4. when role.system_prompt is being updated (e.g. by `role.system_prompt = "..."`) + # Additional, if llm is not set, we will use role's llm actions: list[SerializeAsAny[Action]] = Field(default=[], validate_default=True) rc: RoleContext = Field(default_factory=RoleContext) - subscription: set[str] = set() - planner: Planner = None + addresses: set[str] = set() + planner: Planner = Field(default_factory=Planner) # builtin variables recovered: bool = False # to tag if a recovered role @@ -154,27 +152,85 @@ class Role(SerializationMixin, is_polymorphic_base=True): __hash__ = object.__hash__ # support Role as hashable type in `Environment.members` - @model_validator(mode="after") - def check_subscription(self): - if not self.subscription: - self.subscription = {any_to_str(self), self.name} if self.name else {any_to_str(self)} - return self - def __init__(self, **data: Any): - # --- avoid PydanticUndefinedAnnotation name 'Environment' is not defined # - from metagpt.environment import Environment - - Environment - # ------ - Role.model_rebuild() + self.pydantic_rebuild_model() super().__init__(**data) if self.is_human: - self.llm = HumanProvider() + self.llm = HumanProvider(None) + self._check_actions() self.llm.system_prompt = self._get_prefix() self._watch(data.get("watch") or [UserRequirement]) + if self.latest_observed_msg: + self.recovered = True + + @staticmethod + def pydantic_rebuild_model(): + """Rebuild model to avoid `RecursionError: maximum recursion depth exceeded in comparison`""" + from metagpt.environment import Environment + + Environment + Role.model_rebuild() + + @property + def todo(self) -> Action: + """Get action to do""" + return self.rc.todo + + def set_todo(self, value: Optional[Action]): + """Set action to do and update context""" + if value: + value.context = self.context + self.rc.todo = value + + @property + def git_repo(self): + """Git repo""" + return self.context.git_repo + + @git_repo.setter + def git_repo(self, value): + self.context.git_repo = value + + @property + def src_workspace(self): + """Source workspace under git repo""" + return self.context.src_workspace + + @src_workspace.setter + def src_workspace(self, value): + self.context.src_workspace = value + + @property + def project_repo(self) -> ProjectRepo: + project_repo = ProjectRepo(self.context.git_repo) + return project_repo.with_src_path(self.context.src_workspace) if self.context.src_workspace else project_repo + + @property + def prompt_schema(self): + """Prompt schema: json/markdown""" + return self.config.prompt_schema + + @property + def project_name(self): + return self.config.project_name + + @project_name.setter + def project_name(self, value): + self.config.project_name = value + + @property + def project_path(self): + return self.config.project_path + + @model_validator(mode="after") + def check_addresses(self): + if not self.addresses: + self.addresses = {any_to_str(self), self.name} if self.name else {any_to_str(self)} + return self + def _reset(self): self.states = [] self.actions = [] @@ -183,59 +239,32 @@ class Role(SerializationMixin, is_polymorphic_base=True): def _setting(self): return f"{self.name}({self.profile})" - def serialize(self, stg_path: Path = None): - stg_path = ( - SERDESER_PATH.joinpath(f"team/environment/roles/{self.__class__.__name__}_{self.name}") - if stg_path is None - else stg_path - ) + def _check_actions(self): + """Check actions and set llm and prefix for each action.""" + self.set_actions(self.actions) + return self - role_info = self.model_dump(exclude={"rc": {"memory": True, "msg_buffer": True}, "llm": True}) - role_info.update({"role_class": self.__class__.__name__, "module_name": self.__module__}) - role_info_path = stg_path.joinpath("role_info.json") - write_json_file(role_info_path, role_info) - - self.rc.memory.serialize(stg_path) # serialize role's memory alone - - @classmethod - def deserialize(cls, stg_path: Path) -> "Role": - """stg_path = ./storage/team/environment/roles/{role_class}_{role_name}""" - role_info_path = stg_path.joinpath("role_info.json") - role_info = read_json_file(role_info_path) - - role_class_str = role_info.pop("role_class") - module_name = role_info.pop("module_name") - role_class = import_class(class_name=role_class_str, module_name=module_name) - - role = role_class(**role_info) # initiate particular Role - role.set_recovered(True) # set True to make a tag - - role_memory = Memory.deserialize(stg_path) - role.set_memory(role_memory) - - return role - - def _init_action_system_message(self, action: Action): + def _init_action(self, action: Action): + if not action.private_config: + action.set_llm(self.llm, override=True) + else: + action.set_llm(self.llm, override=False) action.set_prefix(self._get_prefix()) - def refresh_system_message(self): - self.llm.system_prompt = self._get_prefix() + def set_action(self, action: Action): + """Add action to the role.""" + self.set_actions([action]) - def set_recovered(self, recovered: bool = False): - self.recovered = recovered + def set_actions(self, actions: list[Union[Action, Type[Action]]]): + """Add actions to the role. - def set_memory(self, memory: Memory): - self.rc.memory = memory - - def init_actions(self, actions): - self._init_actions(actions) - - def _init_actions(self, actions): + Args: + actions: list of Action classes or instances + """ self._reset() - for idx, action in enumerate(actions): + for action in actions: if not isinstance(action, Action): - ## 默认初始化 - i = action(name="", llm=self.llm) + i = action(context=self.context) else: if self.is_human and not isinstance(action.llm, HumanProvider): logger.warning( @@ -244,9 +273,9 @@ class Role(SerializationMixin, is_polymorphic_base=True): f"try passing in Action classes instead of initialized instances" ) i = action - self._init_action_system_message(i) + self._init_action(i) self.actions.append(i) - self.states.append(f"{idx}. {action}") + self.states.append(f"{len(self.actions)}. {action}") def _set_react_mode(self, react_mode: str, max_react_loop: int = 1, auto_run: bool = True, use_tools: bool = False): """Set strategy of the Role reacting to observed Message. Variation lies in how @@ -284,33 +313,29 @@ class Role(SerializationMixin, is_polymorphic_base=True): def is_watch(self, caused_by: str): return caused_by in self.rc.watch - def subscribe(self, tags: Set[str]): + def set_addresses(self, addresses: Set[str]): """Used to receive Messages with certain tags from the environment. Message will be put into personal message buffer to be further processed in _observe. By default, a Role subscribes Messages with a tag of its own name or profile. """ - self.subscription = tags + self.addresses = addresses if self.rc.env: # According to the routing feature plan in Chapter 2.2.3.2 of RFC 113 - self.rc.env.set_subscription(self, self.subscription) + self.rc.env.set_addresses(self, self.addresses) def _set_state(self, state: int): """Update the current state.""" self.rc.state = state logger.debug(f"actions={self.actions}, state={state}") - self.rc.todo = self.actions[self.rc.state] if state >= 0 else None + self.set_todo(self.actions[self.rc.state] if state >= 0 else None) def set_env(self, env: "Environment"): """Set the environment in which the role works. The role can talk to the environment and can also receive messages by observing.""" self.rc.env = env if env: - env.set_subscription(self, self.subscription) - self.refresh_system_message() # add env message to system message - - @property - def action_count(self): - """Return number of action""" - return len(self.actions) + env.set_addresses(self, self.addresses) + self.llm.system_prompt = self._get_prefix() + self.set_actions(self.actions) # reset actions to update llm and prefix def _get_prefix(self): """Get the role prefix""" @@ -323,7 +348,8 @@ class Role(SerializationMixin, is_polymorphic_base=True): prefix += CONSTRAINT_TEMPLATE.format(**{"constraints": self.constraints}) if self.rc.env and self.rc.env.desc: - other_role_names = ", ".join(self.rc.env.role_names()) + all_roles = self.rc.env.role_names() + other_role_names = ", ".join([r for r in all_roles if r != self.name]) env_desc = f"You are in {self.rc.env.desc} with roles({other_role_names})." prefix += env_desc return prefix @@ -338,7 +364,7 @@ class Role(SerializationMixin, is_polymorphic_base=True): if self.recovered and self.rc.state >= 0: self._set_state(self.rc.state) # action to run from recovered state - self.set_recovered(False) # avoid max_react_loop out of work + self.recovered = False # avoid max_react_loop out of work return True prompt = self._get_prefix() @@ -436,7 +462,7 @@ class Role(SerializationMixin, is_polymorphic_base=True): break # act logger.debug(f"{self._setting}: {self.rc.state=}, will do {self.rc.todo}") - rsp = await self._act() # 这个rsp是否需要publish_message？ + rsp = await self._act() actions_taken += 1 return rsp # return output from the last action @@ -452,10 +478,11 @@ class Role(SerializationMixin, is_polymorphic_base=True): async def _plan_and_act(self) -> Message: """first plan, then execute an action sequence, i.e. _think (of a plan) -> _act -> _act -> ... Use llm to come up with the plan dynamically.""" - ### Common Procedure in both single- and multi-agent setting ### - # create initial plan and update until confirmation - await self.planner.update_plan() + # create initial plan and update it until confirmation + goal = self.rc.memory.get()[-1].content # retreive latest user requirement + await self.planner.update_plan(goal=goal) + # take on tasks until all finished while self.planner.current_task: task = self.planner.current_task logger.info(f"ready to take on task {task}") @@ -463,25 +490,10 @@ class Role(SerializationMixin, is_polymorphic_base=True): # take on current task task_result = await self._act_on_task(task) - # ask for acceptance, users can other refuse and change tasks in the plan - review, task_result_confirmed = await self.planner.ask_review(task_result) + # process the result, such as reviewing, confirming, plan updating + await self.planner.process_task_result(task_result) - if task_result_confirmed: - # tick off this task and record progress - await self.planner.confirm_task(task, task_result, review) - - elif "redo" in review: - # Ask the Role to redo this task with help of review feedback, - # useful when the code run is successful but the procedure or result is not what we want - continue - - else: - # update plan according to user's feedback and to take on changed tasks - await self.planner.update_plan() - - completed_plan_memory = self.planner.get_useful_memories() # completed plan as a outcome - - rsp = completed_plan_memory[0] + rsp = self.planner.get_useful_memories()[0] # return the completed plan as a response self.rc.memory.add(rsp) # add to persistent memory @@ -509,6 +521,8 @@ class Role(SerializationMixin, is_polymorphic_base=True): rsp = await self._act_by_order() elif self.rc.react_mode == RoleReactMode.PLAN_AND_ACT: rsp = await self._plan_and_act() + else: + raise ValueError(f"Unsupported react mode: {self.rc.react_mode}") self._set_state(state=-1) # current reaction is complete, reset state to -1 and todo back to None return rsp @@ -530,7 +544,6 @@ class Role(SerializationMixin, is_polymorphic_base=True): if not msg.cause_by: msg.cause_by = UserRequirement self.put_message(msg) - if not await self._observe(): # If there is no new information, suspend and wait logger.debug(f"{self._setting}: no news. waiting.") @@ -539,7 +552,7 @@ class Role(SerializationMixin, is_polymorphic_base=True): rsp = await self.react() # Reset the next action to be taken. - self.rc.todo = None + self.set_todo(None) # Send the response message to the Environment object to have it relay the message to the subscribers. self.publish_message(rsp) return rsp @@ -550,18 +563,34 @@ class Role(SerializationMixin, is_polymorphic_base=True): return not self.rc.news and not self.rc.todo and self.rc.msg_buffer.empty() async def think(self) -> Action: - """The exported `think` function""" + """ + Export SDK API, used by AgentStore RPC. + The exported `think` function + """ + await self._observe() # For compatibility with the old version of the Agent. await self._think() return self.rc.todo async def act(self) -> ActionOutput: - """The exported `act` function""" + """ + Export SDK API, used by AgentStore RPC. + The exported `act` function + """ msg = await self._act() return ActionOutput(content=msg.content, instruct_content=msg.instruct_content) @property - def todo(self) -> str: - """AgentStore uses this attribute to display to the user what actions the current role should take.""" + def action_description(self) -> str: + """ + Export SDK API, used by AgentStore RPC and Agent. + AgentStore uses this attribute to display to the user what actions the current role should take. + `Role` provides the default property, and this property should be overridden by children classes if necessary, + as demonstrated by the `Engineer` class. + """ + if self.rc.todo: + if self.rc.todo.desc: + return self.rc.todo.desc + return any_to_name(self.rc.todo) if self.actions: return any_to_name(self.actions[0]) return "" diff --git a/metagpt/roles/sales.py b/metagpt/roles/sales.py index ca1cfee85..7929ce7fe 100644 --- a/metagpt/roles/sales.py +++ b/metagpt/roles/sales.py @@ -38,5 +38,5 @@ class Sales(Role): action = SearchAndSummarize(name="", engine=SearchEngineType.CUSTOM_ENGINE, search_func=store.asearch) else: action = SearchAndSummarize() - self._init_actions([action]) + self.set_actions([action]) self._watch([UserRequirement]) diff --git a/metagpt/roles/searcher.py b/metagpt/roles/searcher.py index e713f7697..19a73a40e 100644 --- a/metagpt/roles/searcher.py +++ b/metagpt/roles/searcher.py @@ -10,8 +10,9 @@ from pydantic import Field -from metagpt.actions import ActionOutput, SearchAndSummarize +from metagpt.actions import SearchAndSummarize from metagpt.actions.action_node import ActionNode +from metagpt.actions.action_output import ActionOutput from metagpt.logs import logger from metagpt.roles import Role from metagpt.schema import Message @@ -48,12 +49,12 @@ class Searcher(Role): engine (SearchEngineType): The type of search engine to use. """ super().__init__(**kwargs) - self._init_actions([SearchAndSummarize(engine=self.engine)]) + self.set_actions([SearchAndSummarize(engine=self.engine)]) def set_search_func(self, search_func): """Sets a custom search function for the searcher.""" action = SearchAndSummarize(name="", engine=SearchEngineType.CUSTOM_ENGINE, search_func=search_func) - self._init_actions([action]) + self.set_actions([action]) async def _act_sp(self) -> Message: """Performs the search action in a single process.""" diff --git a/metagpt/roles/sk_agent.py b/metagpt/roles/sk_agent.py index 8921774f0..71df55fcc 100644 --- a/metagpt/roles/sk_agent.py +++ b/metagpt/roles/sk_agent.py @@ -17,9 +17,7 @@ from semantic_kernel.planning.basic_planner import BasicPlanner, Plan from metagpt.actions import UserRequirement from metagpt.actions.execute_task import ExecuteTask -from metagpt.llm import LLM from metagpt.logs import logger -from metagpt.provider.base_llm import BaseLLM from metagpt.roles import Role from metagpt.schema import Message from metagpt.utils.make_sk_kernel import make_sk_kernel @@ -44,7 +42,6 @@ class SkAgent(Role): plan: Plan = Field(default=None, exclude=True) planner_cls: Any = None planner: Union[BasicPlanner, SequentialPlanner, ActionPlanner] = None - llm: BaseLLM = Field(default_factory=LLM) kernel: Kernel = Field(default_factory=Kernel) import_semantic_skill_from_directory: Callable = Field(default=None, exclude=True) import_skill: Callable = Field(default=None, exclude=True) @@ -52,7 +49,7 @@ class SkAgent(Role): def __init__(self, **data: Any) -> None: """Initializes the Engineer role with given attributes.""" super().__init__(**data) - self._init_actions([ExecuteTask()]) + self.set_actions([ExecuteTask()]) self._watch([UserRequirement]) self.kernel = make_sk_kernel() diff --git a/metagpt/roles/teacher.py b/metagpt/roles/teacher.py index 5449fe828..d6715dcd1 100644 --- a/metagpt/roles/teacher.py +++ b/metagpt/roles/teacher.py @@ -11,15 +11,12 @@ import re -import aiofiles - from metagpt.actions import UserRequirement from metagpt.actions.write_teaching_plan import TeachingPlanBlock, WriteTeachingPlanPart -from metagpt.config import CONFIG from metagpt.logs import logger from metagpt.roles import Role from metagpt.schema import Message -from metagpt.utils.common import any_to_str +from metagpt.utils.common import any_to_str, awrite class Teacher(Role): @@ -34,11 +31,11 @@ class Teacher(Role): def __init__(self, **kwargs): super().__init__(**kwargs) - self.name = WriteTeachingPlanPart.format_value(self.name) - self.profile = WriteTeachingPlanPart.format_value(self.profile) - self.goal = WriteTeachingPlanPart.format_value(self.goal) - self.constraints = WriteTeachingPlanPart.format_value(self.constraints) - self.desc = WriteTeachingPlanPart.format_value(self.desc) + self.name = WriteTeachingPlanPart.format_value(self.name, self.context) + self.profile = WriteTeachingPlanPart.format_value(self.profile, self.context) + self.goal = WriteTeachingPlanPart.format_value(self.goal, self.context) + self.constraints = WriteTeachingPlanPart.format_value(self.constraints, self.context) + self.desc = WriteTeachingPlanPart.format_value(self.desc, self.context) async def _think(self) -> bool: """Everything will be done part by part.""" @@ -48,9 +45,9 @@ class Teacher(Role): actions = [] print(TeachingPlanBlock.TOPICS) for topic in TeachingPlanBlock.TOPICS: - act = WriteTeachingPlanPart(context=self.rc.news[0].content, topic=topic, llm=self.llm) + act = WriteTeachingPlanPart(i_context=self.rc.news[0].content, topic=topic, llm=self.llm) actions.append(act) - self._init_actions(actions) + self.set_actions(actions) if self.rc.todo is None: self._set_state(0) @@ -60,7 +57,7 @@ class Teacher(Role): self._set_state(self.rc.state + 1) return True - self.rc.todo = None + self.set_todo(None) return False async def _react(self) -> Message: @@ -81,14 +78,10 @@ class Teacher(Role): async def save(self, content): """Save teaching plan""" filename = Teacher.new_file_name(self.course_title) - pathname = CONFIG.workspace_path / "teaching_plan" + pathname = self.config.workspace.path / "teaching_plan" pathname.mkdir(exist_ok=True) pathname = pathname / filename - try: - async with aiofiles.open(str(pathname), mode="w", encoding="utf-8") as writer: - await writer.write(content) - except Exception as e: - logger.error(f"Save failed：{e}") + await awrite(pathname, content) logger.info(f"Save to:{pathname}") @staticmethod diff --git a/metagpt/roles/tool_maker.py b/metagpt/roles/tool_maker.py deleted file mode 100644 index 68d84b1e6..000000000 --- a/metagpt/roles/tool_maker.py +++ /dev/null @@ -1,53 +0,0 @@ -from pydantic import Field - -from metagpt.actions.ask_review import AskReview -from metagpt.actions.execute_code import ExecutePyCode -from metagpt.actions.write_analysis_code import MakeTools -from metagpt.logs import logger -from metagpt.roles import Role -from metagpt.utils.common import remove_comments - - -class ToolMaker(Role): - execute_code: ExecutePyCode = Field(default_factory=ExecutePyCode, exclude=True) - - async def make_tool(self, code: str, instruction: str, task_id: str = "", auto_run=True): - if len(remove_comments(code).split("\n")) < 5: # no need to consider trivial codes with fewer than 5 lines - return - - logger.warning( - f"Making tools for task_id {task_id}: \ - `{instruction}` \n code: \n {code}" - ) - make_tools = MakeTools() - make_tool_retries, make_tool_current_retry = 3, 0 - while True: - # start make tools - tool_code = await make_tools.run(code, instruction) - make_tool_current_retry += 1 - - # check tool_code by execute_code - logger.info(f"Checking task_id {task_id} tool code by executor...") - execute_result, execute_success = await self.execute_code.run(tool_code) - if not execute_success: - logger.error(f"Tool code faild to execute, \n{execute_result}\n.We will try to fix it ...") - # end make tools - if execute_success or make_tool_current_retry >= make_tool_retries: - if make_tool_current_retry >= make_tool_retries: - logger.error( - f"We have tried the maximum number of attempts {make_tool_retries}\ - and still have not created tools for task_id {task_id} successfully,\ - we will skip it." - ) - break - # save successful tool code in udf - if execute_success: - _, confirmed = await self.ask_review(auto_run=auto_run) - if confirmed: - make_tools.save(tool_code) - - async def ask_review(self, auto_run: bool = True): - if not auto_run: - review, confirmed = await AskReview().run() - return review, confirmed - return "", True diff --git a/metagpt/roles/tutorial_assistant.py b/metagpt/roles/tutorial_assistant.py index 10bd82c60..6cf3a6469 100644 --- a/metagpt/roles/tutorial_assistant.py +++ b/metagpt/roles/tutorial_assistant.py @@ -40,7 +40,7 @@ class TutorialAssistant(Role): def __init__(self, **kwargs): super().__init__(**kwargs) - self._init_actions([WriteDirectory(language=self.language)]) + self.set_actions([WriteDirectory(language=self.language)]) self._set_react_mode(react_mode=RoleReactMode.BY_ORDER.value) async def _handle_directory(self, titles: Dict) -> Message: @@ -63,7 +63,7 @@ class TutorialAssistant(Role): directory += f"- {key}\n" for second_dir in first_dir[key]: directory += f" - {second_dir}\n" - self._init_actions(actions) + self.set_actions(actions) async def _act(self) -> Message: """Perform an action as determined by the role. diff --git a/metagpt/schema.py b/metagpt/schema.py index e69f432db..08f97be94 100644 --- a/metagpt/schema.py +++ b/metagpt/schema.py @@ -23,7 +23,7 @@ from abc import ABC from asyncio import Queue, QueueEmpty, wait_for from json import JSONDecodeError from pathlib import Path -from typing import Any, Callable, Dict, List, Optional, Type, TypeVar, Union +from typing import Any, Dict, Iterable, List, Optional, Type, TypeVar, Union from pydantic import ( BaseModel, @@ -32,15 +32,17 @@ from pydantic import ( PrivateAttr, field_serializer, field_validator, + model_serializer, + model_validator, ) -from pydantic_core import core_schema -from metagpt.config import CONFIG from metagpt.const import ( + CODE_PLAN_AND_CHANGE_FILENAME, MESSAGE_ROUTE_CAUSE_BY, MESSAGE_ROUTE_FROM, MESSAGE_ROUTE_TO, MESSAGE_ROUTE_TO_ALL, + PRDS_FILE_REPO, SYSTEM_DESIGN_FILE_REPO, TASK_FILE_REPO, ) @@ -54,7 +56,7 @@ from metagpt.utils.serialize import ( ) -class SerializationMixin(BaseModel): +class SerializationMixin(BaseModel, extra="forbid"): """ PolyMorphic subclasses Serialization / Deserialization Mixin - First of all, we need to know that pydantic is not designed for polymorphism. @@ -69,49 +71,44 @@ class SerializationMixin(BaseModel): __is_polymorphic_base = False __subclasses_map__ = {} - @classmethod - def __get_pydantic_core_schema__( - cls, source: type["SerializationMixin"], handler: Callable[[Any], core_schema.CoreSchema] - ) -> core_schema.CoreSchema: - schema = handler(source) - og_schema_ref = schema["ref"] - schema["ref"] += ":mixin" - - return core_schema.no_info_before_validator_function( - cls.__deserialize_with_real_type__, - schema=schema, - ref=og_schema_ref, - serialization=core_schema.wrap_serializer_function_ser_schema(cls.__serialize_add_class_type__), - ) - - @classmethod - def __serialize_add_class_type__( - cls, - value, - handler: core_schema.SerializerFunctionWrapHandler, - ) -> Any: - ret = handler(value) - if not len(cls.__subclasses__()): - # only subclass add `__module_class_name` - ret["__module_class_name"] = f"{cls.__module__}.{cls.__qualname__}" + @model_serializer(mode="wrap") + def __serialize_with_class_type__(self, default_serializer) -> Any: + # default serializer, then append the `__module_class_name` field and return + ret = default_serializer(self) + ret["__module_class_name"] = f"{self.__class__.__module__}.{self.__class__.__qualname__}" return ret + @model_validator(mode="wrap") @classmethod - def __deserialize_with_real_type__(cls, value: Any): - if not isinstance(value, dict): - return value + def __convert_to_real_type__(cls, value: Any, handler): + if isinstance(value, dict) is False: + return handler(value) - if not cls.__is_polymorphic_base or (len(cls.__subclasses__()) and "__module_class_name" not in value): - # add right condition to init BaseClass like Action() - return value - module_class_name = value.get("__module_class_name", None) - if module_class_name is None: - raise ValueError("Missing field: __module_class_name") + # it is a dict so make sure to remove the __module_class_name + # because we don't allow extra keywords but want to ensure + # e.g Cat.model_validate(cat.model_dump()) works + class_full_name = value.pop("__module_class_name", None) - class_type = cls.__subclasses_map__.get(module_class_name, None) + # if it's not the polymorphic base we construct via default handler + if not cls.__is_polymorphic_base: + if class_full_name is None: + return handler(value) + elif str(cls) == f"": + return handler(value) + else: + # f"Trying to instantiate {class_full_name} but this is not the polymorphic base class") + pass + + # otherwise we lookup the correct polymorphic type and construct that + # instead + if class_full_name is None: + raise ValueError("Missing __module_class_name field") + + class_type = cls.__subclasses_map__.get(class_full_name, None) if class_type is None: - raise TypeError("Trying to instantiate {module_class_name} which not defined yet.") + # TODO could try dynamic import + raise TypeError("Trying to instantiate {class_full_name}, which has not yet been defined!") return class_type(**value) @@ -151,12 +148,6 @@ class Document(BaseModel): """ return os.path.join(self.root_path, self.filename) - @property - def full_path(self): - if not CONFIG.git_repo: - return None - return str(CONFIG.git_repo.workdir / self.root_path / self.filename) - def __str__(self): return self.content @@ -173,6 +164,26 @@ class Documents(BaseModel): docs: Dict[str, Document] = Field(default_factory=dict) + @classmethod + def from_iterable(cls, documents: Iterable[Document]) -> Documents: + """Create a Documents instance from a list of Document instances. + + :param documents: A list of Document instances. + :return: A Documents instance. + """ + + docs = {doc.filename: doc for doc in documents} + return Documents(docs=docs) + + def to_action_output(self) -> "ActionOutput": + """Convert to action output string. + + :return: A string representing action output. + """ + from metagpt.actions.action_output import ActionOutput + + return ActionOutput(content=self.model_dump_json(), instruct_content=self) + class Message(BaseModel): """list[: ]""" @@ -193,12 +204,17 @@ class Message(BaseModel): @field_validator("instruct_content", mode="before") @classmethod def check_instruct_content(cls, ic: Any) -> BaseModel: - if ic and not isinstance(ic, BaseModel) and "class" in ic: - # compatible with custom-defined ActionOutput - mapping = actionoutput_str_to_mapping(ic["mapping"]) - - actionnode_class = import_class("ActionNode", "metagpt.actions.action_node") # avoid circular import - ic_obj = actionnode_class.create_model_class(class_name=ic["class"], mapping=mapping) + if ic and isinstance(ic, dict) and "class" in ic: + if "mapping" in ic: + # compatible with custom-defined ActionOutput + mapping = actionoutput_str_to_mapping(ic["mapping"]) + actionnode_class = import_class("ActionNode", "metagpt.actions.action_node") # avoid circular import + ic_obj = actionnode_class.create_model_class(class_name=ic["class"], mapping=mapping) + elif "module" in ic: + # subclasses of BaseModel + ic_obj = import_class(ic["class"], ic["module"]) + else: + raise KeyError("missing required key to init Message.instruct_content from dict") ic = ic_obj(**ic["value"]) return ic @@ -218,18 +234,21 @@ class Message(BaseModel): return any_to_str_set(send_to if send_to else {MESSAGE_ROUTE_TO_ALL}) @field_serializer("instruct_content", mode="plain") - def ser_instruct_content(self, ic: BaseModel) -> Union[str, None]: + def ser_instruct_content(self, ic: BaseModel) -> Union[dict, None]: ic_dict = None if ic: # compatible with custom-defined ActionOutput schema = ic.model_json_schema() - # `Documents` contain definitions - if "definitions" not in schema: - # TODO refine with nested BaseModel + ic_type = str(type(ic)) + if " CodePlanAndChangeContext: + ctx = CodePlanAndChangeContext(requirement=kwargs.get("requirement", "")) + for filename in filenames: + filename = Path(filename) + if filename.is_relative_to(PRDS_FILE_REPO): + ctx.prd_filename = filename.name + continue + if filename.is_relative_to(SYSTEM_DESIGN_FILE_REPO): + ctx.design_filename = filename.name + continue + if filename.is_relative_to(TASK_FILE_REPO): + ctx.task_filename = filename.name + continue + return ctx + + # mermaid class view class ClassMeta(BaseModel): name: str = "" diff --git a/metagpt/startup.py b/metagpt/startup.py index 767a19a9d..4a077cab7 100644 --- a/metagpt/startup.py +++ b/metagpt/startup.py @@ -1,18 +1,80 @@ #!/usr/bin/env python # -*- coding: utf-8 -*- + import asyncio +import shutil from pathlib import Path import typer -from metagpt.config import CONFIG +from metagpt.config2 import config +from metagpt.const import CONFIG_ROOT, METAGPT_ROOT +from metagpt.context import Context +from metagpt.utils.project_repo import ProjectRepo -app = typer.Typer(add_completion=False) +app = typer.Typer(add_completion=False, pretty_exceptions_show_locals=False) -@app.command() +def generate_repo( + idea, + investment, + n_round, + code_review, + run_tests, + implement, + project_name, + inc, + project_path, + reqa_file, + max_auto_summarize_code, + recover_path, +) -> ProjectRepo: + """Run the startup logic. Can be called from CLI or other Python scripts.""" + from metagpt.roles import ( + Architect, + Engineer, + ProductManager, + ProjectManager, + QaEngineer, + ) + from metagpt.team import Team + + config.update_via_cli(project_path, project_name, inc, reqa_file, max_auto_summarize_code) + ctx = Context(config=config) + + if not recover_path: + company = Team(context=ctx) + company.hire( + [ + ProductManager(), + Architect(), + ProjectManager(), + ] + ) + + if implement or code_review: + company.hire([Engineer(n_borg=5, use_code_review=code_review)]) + + if run_tests: + company.hire([QaEngineer()]) + else: + stg_path = Path(recover_path) + if not stg_path.exists() or not str(stg_path).endswith("team"): + raise FileNotFoundError(f"{recover_path} not exists or not endswith `team`") + + company = Team.deserialize(stg_path=stg_path, context=ctx) + idea = company.idea + + company.invest(investment) + company.run_project(idea) + asyncio.run(company.run(n_round=n_round)) + + return ctx.repo + + +@app.command("", help="Start a new project.") def startup( - idea: str = typer.Argument(..., help="Your innovative idea, such as 'Create a 2048 game.'"), + idea: str = typer.Argument(None, help="Your innovative idea, such as 'Create a 2048 game.'"), investment: float = typer.Option(default=3.0, help="Dollar amount to invest in the AI company."), n_round: int = typer.Option(default=5, help="Number of rounds for the simulation."), code_review: bool = typer.Option(default=True, help="Whether to use code review."), @@ -33,46 +95,49 @@ def startup( "unlimited. This parameter is used for debugging the workflow.", ), recover_path: str = typer.Option(default=None, help="recover the project from existing serialized storage"), + init_config: bool = typer.Option(default=False, help="Initialize the configuration file for MetaGPT."), ): """Run a startup. Be a boss.""" - from metagpt.roles import ( - Architect, - Engineer, - ProductManager, - ProjectManager, - QaEngineer, + if init_config: + copy_config_to() + return + + if idea is None: + typer.echo("Missing argument 'IDEA'. Run 'metagpt --help' for more information.") + raise typer.Exit() + + return generate_repo( + idea, + investment, + n_round, + code_review, + run_tests, + implement, + project_name, + inc, + project_path, + reqa_file, + max_auto_summarize_code, + recover_path, ) - from metagpt.team import Team - CONFIG.update_via_cli(project_path, project_name, inc, reqa_file, max_auto_summarize_code) - if not recover_path: - company = Team() - company.hire( - [ - ProductManager(), - Architect(), - ProjectManager(), - ] - ) +def copy_config_to(config_path=METAGPT_ROOT / "config" / "config2.yaml"): + """Initialize the configuration file for MetaGPT.""" + target_path = CONFIG_ROOT / "config2.yaml" - if implement or code_review: - company.hire([Engineer(n_borg=5, use_code_review=code_review)]) + # 创建目标目录（如果不存在） + target_path.parent.mkdir(parents=True, exist_ok=True) - if run_tests: - company.hire([QaEngineer()]) - else: - # # stg_path = SERDESER_PATH.joinpath("team") - stg_path = Path(recover_path) - if not stg_path.exists() or not str(stg_path).endswith("team"): - raise FileNotFoundError(f"{recover_path} not exists or not endswith `team`") + # 如果目标文件已经存在，则重命名为 .bak + if target_path.exists(): + backup_path = target_path.with_suffix(".bak") + target_path.rename(backup_path) + print(f"Existing configuration file backed up at {backup_path}") - company = Team.deserialize(stg_path=stg_path) - idea = company.idea # use original idea - - company.invest(investment) - company.run_project(idea) - asyncio.run(company.run(n_round=n_round)) + # 复制文件 + shutil.copy(str(config_path), target_path) + print(f"Configuration file initialized at {target_path}") if __name__ == "__main__": diff --git a/metagpt/subscription.py b/metagpt/subscription.py index e2b0916ac..d225a5d87 100644 --- a/metagpt/subscription.py +++ b/metagpt/subscription.py @@ -13,7 +13,7 @@ class SubscriptionRunner(BaseModel): Example: >>> import asyncio - >>> from metagpt.subscription import SubscriptionRunner + >>> from metagpt.address import SubscriptionRunner >>> from metagpt.roles import Searcher >>> from metagpt.schema import Message diff --git a/metagpt/team.py b/metagpt/team.py index b98fc2efb..35f987b57 100644 --- a/metagpt/team.py +++ b/metagpt/team.py @@ -10,13 +10,13 @@ import warnings from pathlib import Path -from typing import Any +from typing import Any, Optional from pydantic import BaseModel, ConfigDict, Field from metagpt.actions import UserRequirement -from metagpt.config import CONFIG from metagpt.const import MESSAGE_ROUTE_TO_ALL, SERDESER_PATH +from metagpt.context import Context from metagpt.environment import Environment from metagpt.logs import logger from metagpt.roles import Role @@ -37,12 +37,17 @@ class Team(BaseModel): model_config = ConfigDict(arbitrary_types_allowed=True) - env: Environment = Field(default_factory=Environment) + env: Optional[Environment] = None investment: float = Field(default=10.0) idea: str = Field(default="") - def __init__(self, **data: Any): + def __init__(self, context: Context = None, **data: Any): super(Team, self).__init__(**data) + ctx = context or Context() + if not self.env: + self.env = Environment(context=ctx) + else: + self.env.context = ctx # The `env` object is allocated by deserialization if "roles" in data: self.hire(data["roles"]) if "env_desc" in data: @@ -50,47 +55,43 @@ class Team(BaseModel): def serialize(self, stg_path: Path = None): stg_path = SERDESER_PATH.joinpath("team") if stg_path is None else stg_path + team_info_path = stg_path.joinpath("team.json") - team_info_path = stg_path.joinpath("team_info.json") - write_json_file(team_info_path, self.model_dump(exclude={"env": True})) - - self.env.serialize(stg_path.joinpath("environment")) # save environment alone + write_json_file(team_info_path, self.model_dump()) @classmethod - def deserialize(cls, stg_path: Path) -> "Team": + def deserialize(cls, stg_path: Path, context: Context = None) -> "Team": """stg_path = ./storage/team""" # recover team_info - team_info_path = stg_path.joinpath("team_info.json") + team_info_path = stg_path.joinpath("team.json") if not team_info_path.exists(): raise FileNotFoundError( - "recover storage meta file `team_info.json` not exist, " - "not to recover and please start a new project." + "recover storage meta file `team.json` not exist, " "not to recover and please start a new project." ) team_info: dict = read_json_file(team_info_path) - - # recover environment - environment = Environment.deserialize(stg_path=stg_path.joinpath("environment")) - team_info.update({"env": environment}) - team = Team(**team_info) + ctx = context or Context() + team = Team(**team_info, context=ctx) return team def hire(self, roles: list[Role]): """Hire roles to cooperate""" self.env.add_roles(roles) + @property + def cost_manager(self): + """Get cost manager""" + return self.env.context.cost_manager + def invest(self, investment: float): """Invest company. raise NoMoneyException when exceed max_budget.""" self.investment = investment - CONFIG.max_budget = investment + self.cost_manager.max_budget = investment logger.info(f"Investment: ${investment}.") - @staticmethod - def _check_balance(): - if CONFIG.cost_manager.total_cost > CONFIG.cost_manager.max_budget: - raise NoMoneyException( - CONFIG.cost_manager.total_cost, f"Insufficient funds: {CONFIG.cost_manager.max_budget}" - ) + def _check_balance(self): + if self.cost_manager.total_cost >= self.cost_manager.max_budget: + raise NoMoneyException(self.cost_manager.total_cost, f"Insufficient funds: {self.cost_manager.max_budget}") def run_project(self, idea, send_to: str = ""): """Run a project from publishing user requirement.""" diff --git a/metagpt/tools/azure_tts.py b/metagpt/tools/azure_tts.py index f4f8aa0a2..2e0e2267c 100644 --- a/metagpt/tools/azure_tts.py +++ b/metagpt/tools/azure_tts.py @@ -13,7 +13,6 @@ from uuid import uuid4 import aiofiles from azure.cognitiveservices.speech import AudioConfig, SpeechConfig, SpeechSynthesizer -from metagpt.config import CONFIG from metagpt.logs import logger @@ -25,8 +24,8 @@ class AzureTTS: :param subscription_key: key is used to access your Azure AI service API, see: `https://portal.azure.com/` > `Resource Management` > `Keys and Endpoint` :param region: This is the location (or region) of your resource. You may need to use this field when making calls to this API. """ - self.subscription_key = subscription_key if subscription_key else CONFIG.AZURE_TTS_SUBSCRIPTION_KEY - self.region = region if region else CONFIG.AZURE_TTS_REGION + self.subscription_key = subscription_key + self.region = region # 参数参考：https://learn.microsoft.com/zh-cn/azure/cognitive-services/speech-service/language-support?tabs=tts#voice-styles-and-roles async def synthesize_speech(self, lang, voice, text, output_file): @@ -83,10 +82,6 @@ async def oas3_azsure_tts(text, lang="", voice="", style="", role="", subscripti role = "Girl" if not style: style = "affectionate" - if not subscription_key: - subscription_key = CONFIG.AZURE_TTS_SUBSCRIPTION_KEY - if not region: - region = CONFIG.AZURE_TTS_REGION xml_value = AzureTTS.role_style_text(role=role, style=style, text=text) tts = AzureTTS(subscription_key=subscription_key, region=region) diff --git a/metagpt/tools/iflytek_tts.py b/metagpt/tools/iflytek_tts.py index ad2395362..6ce48826b 100644 --- a/metagpt/tools/iflytek_tts.py +++ b/metagpt/tools/iflytek_tts.py @@ -23,7 +23,6 @@ import aiofiles import websockets as websockets from pydantic import BaseModel -from metagpt.config import CONFIG from metagpt.logs import logger @@ -56,9 +55,9 @@ class IFlyTekTTS(object): :param api_key: WebAPI argument, see: `https://console.xfyun.cn/services/tts` :param api_secret: WebAPI argument, see: `https://console.xfyun.cn/services/tts` """ - self.app_id = app_id or CONFIG.IFLYTEK_APP_ID - self.api_key = api_key or CONFIG.IFLYTEK_API_KEY - self.api_secret = api_secret or CONFIG.API_SECRET + self.app_id = app_id + self.api_key = api_key + self.api_secret = api_secret async def synthesize_speech(self, text, output_file: str, voice=DEFAULT_IFLYTEK_VOICE): url = self._create_url() @@ -127,14 +126,6 @@ async def oas3_iflytek_tts(text: str, voice: str = "", app_id: str = "", api_key :return: Returns the Base64-encoded .mp3 file data if successful, otherwise an empty string. """ - if not app_id: - app_id = CONFIG.IFLYTEK_APP_ID - if not api_key: - api_key = CONFIG.IFLYTEK_API_KEY - if not api_secret: - api_secret = CONFIG.IFLYTEK_API_SECRET - if not voice: - voice = CONFIG.IFLYTEK_VOICE or DEFAULT_IFLYTEK_VOICE filename = Path(__file__).parent / (uuid.uuid4().hex + ".mp3") try: diff --git a/metagpt/tools/libs/gpt_v_generator.py b/metagpt/tools/libs/gpt_v_generator.py index adc3b1051..e079a8eef 100644 --- a/metagpt/tools/libs/gpt_v_generator.py +++ b/metagpt/tools/libs/gpt_v_generator.py @@ -33,12 +33,12 @@ Now, please generate the corresponding webpage code including HTML, CSS and Java @register_tool(tool_type=ToolTypeEnum.IMAGE2WEBPAGE.value) class GPTvGenerator: def __init__(self): - from metagpt.config import CONFIG + from metagpt.config2 import config - OPENAI_API_BASE = CONFIG.OPENAI_BASE_URL - API_KEY = CONFIG.OPENAI_API_KEY - MODEL = CONFIG.OPENAI_VISION_MODEL - MAX_TOKENS = CONFIG.VISION_MAX_TOKENS + OPENAI_API_BASE = config.llm.base_url + API_KEY = config.llm.api_key + MODEL = config.OPENAI_VISION_MODEL + MAX_TOKENS = config.VISION_MAX_TOKENS self.api_key = API_KEY self.api_base = OPENAI_API_BASE self.model = MODEL diff --git a/metagpt/tools/libs/sd_engine.py b/metagpt/tools/libs/sd_engine.py index 7f182f380..57a025f3c 100644 --- a/metagpt/tools/libs/sd_engine.py +++ b/metagpt/tools/libs/sd_engine.py @@ -56,11 +56,11 @@ default_negative_prompt = "(easynegative:0.8),black, dark,Low resolution" @register_tool(tool_type=ToolTypeEnum.STABLE_DIFFUSION.value) class SDEngine: def __init__(self, sd_url=""): - from metagpt.config import CONFIG + from metagpt.config2 import config # Initialize the SDEngine with configuration - self.sd_url = sd_url if sd_url else CONFIG.get("SD_URL") - self.sd_t2i_url = f"{self.sd_url}{CONFIG.get('SD_T2I_API')}" + self.sd_url = sd_url if sd_url else config.get("SD_URL") + self.sd_t2i_url = f"{self.sd_url}{config.get('SD_T2I_API')}" # Define default payload settings for SD API self.payload = payload logger.info(self.sd_t2i_url) diff --git a/metagpt/tools/libs/web_scraping.py b/metagpt/tools/libs/web_scraping.py index e8e73f123..921fca809 100644 --- a/metagpt/tools/libs/web_scraping.py +++ b/metagpt/tools/libs/web_scraping.py @@ -19,4 +19,4 @@ async def scrape_web_playwright(url, *urls): web = await PlaywrightWrapper("chromium").run(url, *urls) # Return the inner text content of the web page - return {"inner_text": web.inner_text, "html": web.html} + return {"inner_text": web.inner_text.strip(), "html": web.html.strip()} diff --git a/metagpt/tools/metagpt_text_to_image.py b/metagpt/tools/metagpt_text_to_image.py index 9a84e69eb..cf7bf97e7 100644 --- a/metagpt/tools/metagpt_text_to_image.py +++ b/metagpt/tools/metagpt_text_to_image.py @@ -13,7 +13,6 @@ import aiohttp import requests from pydantic import BaseModel -from metagpt.config import CONFIG from metagpt.logs import logger @@ -22,7 +21,7 @@ class MetaGPTText2Image: """ :param model_url: Model reset api url """ - self.model_url = model_url if model_url else CONFIG.METAGPT_TEXT_TO_IMAGE_MODEL + self.model_url = model_url async def text_2_image(self, text, size_type="512x512"): """Text to image @@ -93,6 +92,4 @@ async def oas3_metagpt_text_to_image(text, size_type: str = "512x512", model_url """ if not text: return "" - if not model_url: - model_url = CONFIG.METAGPT_TEXT_TO_IMAGE_MODEL_URL return await MetaGPTText2Image(model_url).text_2_image(text, size_type=size_type) diff --git a/metagpt/tools/moderation.py b/metagpt/tools/moderation.py index cda164ec5..f00b0e1f2 100644 --- a/metagpt/tools/moderation.py +++ b/metagpt/tools/moderation.py @@ -7,12 +7,12 @@ """ from typing import Union -from metagpt.llm import LLM +from metagpt.provider.base_llm import BaseLLM class Moderation: - def __init__(self): - self.llm = LLM() + def __init__(self, llm: BaseLLM): + self.llm = llm def handle_moderation_results(self, results): resp = [] diff --git a/metagpt/tools/openai_text_to_embedding.py b/metagpt/tools/openai_text_to_embedding.py index 52b2cc9eb..e93bfb271 100644 --- a/metagpt/tools/openai_text_to_embedding.py +++ b/metagpt/tools/openai_text_to_embedding.py @@ -13,7 +13,6 @@ import aiohttp import requests from pydantic import BaseModel, Field -from metagpt.config import CONFIG from metagpt.logs import logger @@ -43,11 +42,12 @@ class ResultEmbedding(BaseModel): class OpenAIText2Embedding: - def __init__(self, openai_api_key): + def __init__(self, api_key: str, proxy: str): """ :param openai_api_key: OpenAI API key, For more details, checkout: `https://platform.openai.com/account/api-keys` """ - self.openai_api_key = openai_api_key or CONFIG.OPENAI_API_KEY + self.api_key = api_key + self.proxy = proxy async def text_2_embedding(self, text, model="text-embedding-ada-002"): """Text to embedding @@ -57,8 +57,8 @@ class OpenAIText2Embedding: :return: A json object of :class:`ResultEmbedding` class if successful, otherwise `{}`. """ - proxies = {"proxy": CONFIG.openai_proxy} if CONFIG.openai_proxy else {} - headers = {"Content-Type": "application/json", "Authorization": f"Bearer {self.openai_api_key}"} + proxies = {"proxy": self.proxy} if self.proxy else {} + headers = {"Content-Type": "application/json", "Authorization": f"Bearer {self.api_key}"} data = {"input": text, "model": model} url = "https://api.openai.com/v1/embeddings" try: @@ -72,16 +72,14 @@ class OpenAIText2Embedding: # Export -async def oas3_openai_text_to_embedding(text, model="text-embedding-ada-002", openai_api_key=""): +async def oas3_openai_text_to_embedding(text, openai_api_key: str, model="text-embedding-ada-002", proxy: str = ""): """Text to embedding :param text: The text used for embedding. :param model: One of ['text-embedding-ada-002'], ID of the model to use. For more details, checkout: `https://api.openai.com/v1/models`. - :param openai_api_key: OpenAI API key, For more details, checkout: `https://platform.openai.com/account/api-keys` + :param config: OpenAI config with API key, For more details, checkout: `https://platform.openai.com/account/api-keys` :return: A json object of :class:`ResultEmbedding` class if successful, otherwise `{}`. """ if not text: return "" - if not openai_api_key: - openai_api_key = CONFIG.OPENAI_API_KEY - return await OpenAIText2Embedding(openai_api_key).text_2_embedding(text, model=model) + return await OpenAIText2Embedding(api_key=openai_api_key, proxy=proxy).text_2_embedding(text, model=model) diff --git a/metagpt/tools/openai_text_to_image.py b/metagpt/tools/openai_text_to_image.py index aa00abdcc..bf7c5e799 100644 --- a/metagpt/tools/openai_text_to_image.py +++ b/metagpt/tools/openai_text_to_image.py @@ -10,16 +10,13 @@ import aiohttp import requests -from metagpt.llm import LLM from metagpt.logs import logger +from metagpt.provider.base_llm import BaseLLM class OpenAIText2Image: - def __init__(self): - """ - :param openai_api_key: OpenAI API key, For more details, checkout: `https://platform.openai.com/account/api-keys` - """ - self._llm = LLM() + def __init__(self, llm: BaseLLM): + self.llm = llm async def text_2_image(self, text, size_type="1024x1024"): """Text to image @@ -29,7 +26,7 @@ class OpenAIText2Image: :return: The image data is returned in Base64 encoding. """ try: - result = await self._llm.aclient.images.generate(prompt=text, n=1, size=size_type) + result = await self.llm.aclient.images.generate(prompt=text, n=1, size=size_type) except Exception as e: logger.error(f"An error occurred:{e}") return "" @@ -57,13 +54,14 @@ class OpenAIText2Image: # Export -async def oas3_openai_text_to_image(text, size_type: str = "1024x1024"): +async def oas3_openai_text_to_image(text, size_type: str = "1024x1024", llm: BaseLLM = None): """Text to image :param text: The text used for image conversion. :param size_type: One of ['256x256', '512x512', '1024x1024'] + :param llm: LLM instance :return: The image data is returned in Base64 encoding. """ if not text: return "" - return await OpenAIText2Image().text_2_image(text, size_type=size_type) + return await OpenAIText2Image(llm).text_2_image(text, size_type=size_type) diff --git a/metagpt/tools/search_engine.py b/metagpt/tools/search_engine.py index 64388a11f..0d0db9147 100644 --- a/metagpt/tools/search_engine.py +++ b/metagpt/tools/search_engine.py @@ -10,7 +10,6 @@ from typing import Callable, Coroutine, Literal, Optional, Union, overload from semantic_kernel.skill_definition import sk_function -from metagpt.config import CONFIG from metagpt.tools import SearchEngineType @@ -43,22 +42,22 @@ class SearchEngine: def __init__( self, - engine: Optional[SearchEngineType] = None, + engine: Optional[SearchEngineType] = SearchEngineType.SERPER_GOOGLE, run_func: Callable[[str, int, bool], Coroutine[None, None, Union[str, list[str]]]] = None, + **kwargs, ): - engine = engine or CONFIG.search_engine if engine == SearchEngineType.SERPAPI_GOOGLE: module = "metagpt.tools.search_engine_serpapi" - run_func = importlib.import_module(module).SerpAPIWrapper().run + run_func = importlib.import_module(module).SerpAPIWrapper(**kwargs).run elif engine == SearchEngineType.SERPER_GOOGLE: module = "metagpt.tools.search_engine_serper" - run_func = importlib.import_module(module).SerperWrapper().run + run_func = importlib.import_module(module).SerperWrapper(**kwargs).run elif engine == SearchEngineType.DIRECT_GOOGLE: module = "metagpt.tools.search_engine_googleapi" - run_func = importlib.import_module(module).GoogleAPIWrapper().run + run_func = importlib.import_module(module).GoogleAPIWrapper(**kwargs).run elif engine == SearchEngineType.DUCK_DUCK_GO: module = "metagpt.tools.search_engine_ddg" - run_func = importlib.import_module(module).DDGAPIWrapper().run + run_func = importlib.import_module(module).DDGAPIWrapper(**kwargs).run elif engine == SearchEngineType.CUSTOM_ENGINE: pass # run_func = run_func else: diff --git a/metagpt/tools/search_engine_ddg.py b/metagpt/tools/search_engine_ddg.py index 57bc61b82..3d004a4ee 100644 --- a/metagpt/tools/search_engine_ddg.py +++ b/metagpt/tools/search_engine_ddg.py @@ -7,6 +7,8 @@ import json from concurrent import futures from typing import Literal, overload +from metagpt.config2 import config + try: from duckduckgo_search import DDGS except ImportError: @@ -15,8 +17,6 @@ except ImportError: "You can install it by running the command: `pip install -e.[search-ddg]`" ) -from metagpt.config import CONFIG - class DDGAPIWrapper: """Wrapper around duckduckgo_search API. @@ -31,8 +31,8 @@ class DDGAPIWrapper: executor: futures.Executor | None = None, ): kwargs = {} - if CONFIG.global_proxy: - kwargs["proxies"] = CONFIG.global_proxy + if config.proxy: + kwargs["proxies"] = config.proxy self.loop = loop self.executor = executor self.ddgs = DDGS(**kwargs) diff --git a/metagpt/tools/search_engine_googleapi.py b/metagpt/tools/search_engine_googleapi.py index 8aca3aee2..0a8f796cb 100644 --- a/metagpt/tools/search_engine_googleapi.py +++ b/metagpt/tools/search_engine_googleapi.py @@ -11,7 +11,7 @@ from urllib.parse import urlparse import httplib2 from pydantic import BaseModel, ConfigDict, Field, field_validator -from metagpt.config import CONFIG +from metagpt.config2 import config from metagpt.logs import logger try: @@ -35,7 +35,7 @@ class GoogleAPIWrapper(BaseModel): @field_validator("google_api_key", mode="before") @classmethod def check_google_api_key(cls, val: str): - val = val or CONFIG.google_api_key + val = val or config.search.api_key if not val: raise ValueError( "To use, make sure you provide the google_api_key when constructing an object. Alternatively, " @@ -47,7 +47,7 @@ class GoogleAPIWrapper(BaseModel): @field_validator("google_cse_id", mode="before") @classmethod def check_google_cse_id(cls, val: str): - val = val or CONFIG.google_cse_id + val = val or config.search.cse_id if not val: raise ValueError( "To use, make sure you provide the google_cse_id when constructing an object. Alternatively, " @@ -59,8 +59,8 @@ class GoogleAPIWrapper(BaseModel): @property def google_api_client(self): build_kwargs = {"developerKey": self.google_api_key} - if CONFIG.global_proxy: - parse_result = urlparse(CONFIG.global_proxy) + if config.proxy: + parse_result = urlparse(config.proxy) proxy_type = parse_result.scheme if proxy_type == "https": proxy_type = "http" diff --git a/metagpt/tools/search_engine_serpapi.py b/metagpt/tools/search_engine_serpapi.py index 9d2d20af6..a8d5b49d0 100644 --- a/metagpt/tools/search_engine_serpapi.py +++ b/metagpt/tools/search_engine_serpapi.py @@ -10,7 +10,7 @@ from typing import Any, Dict, Optional, Tuple import aiohttp from pydantic import BaseModel, ConfigDict, Field, field_validator -from metagpt.config import CONFIG +from metagpt.config2 import config class SerpAPIWrapper(BaseModel): @@ -32,7 +32,7 @@ class SerpAPIWrapper(BaseModel): @field_validator("serpapi_api_key", mode="before") @classmethod def check_serpapi_api_key(cls, val: str): - val = val or CONFIG.serpapi_api_key + val = val or config.search.api_key if not val: raise ValueError( "To use, make sure you provide the serpapi_api_key when constructing an object. Alternatively, " diff --git a/metagpt/tools/search_engine_serper.py b/metagpt/tools/search_engine_serper.py index 3dc1d3591..39cb936b8 100644 --- a/metagpt/tools/search_engine_serper.py +++ b/metagpt/tools/search_engine_serper.py @@ -11,7 +11,7 @@ from typing import Any, Dict, Optional, Tuple import aiohttp from pydantic import BaseModel, ConfigDict, Field, field_validator -from metagpt.config import CONFIG +from metagpt.config2 import config class SerperWrapper(BaseModel): @@ -25,7 +25,7 @@ class SerperWrapper(BaseModel): @field_validator("serper_api_key", mode="before") @classmethod def check_serper_api_key(cls, val: str): - val = val or CONFIG.serper_api_key + val = val or config.search.api_key if not val: raise ValueError( "To use, make sure you provide the serper_api_key when constructing an object. Alternatively, " diff --git a/metagpt/tools/tool_registry.py b/metagpt/tools/tool_registry.py index d16defa0a..7e4ee5ead 100644 --- a/metagpt/tools/tool_registry.py +++ b/metagpt/tools/tool_registry.py @@ -24,9 +24,10 @@ class ToolRegistry(BaseModel): tool_types: dict = {} tools_by_types: dict = defaultdict(dict) # two-layer k-v, {tool_type: {tool_name: {...}, ...}, ...} - def register_tool_type(self, tool_type: ToolType): + def register_tool_type(self, tool_type: ToolType, verbose: bool = False): self.tool_types[tool_type.name] = tool_type - logger.info(f"tool type {tool_type.name} registered") + if verbose: + logger.info(f"tool type {tool_type.name} registered") def register_tool( self, @@ -38,6 +39,7 @@ class ToolRegistry(BaseModel): tool_source_object=None, include_functions=[], make_schema_if_not_exists=True, + verbose=False, ): if self.has_tool(tool_name): return @@ -68,7 +70,8 @@ class ToolRegistry(BaseModel): tool = Tool(name=tool_name, path=tool_path, schemas=schemas, code=tool_code) self.tools[tool_name] = tool self.tools_by_types[tool_type][tool_name] = tool - logger.info(f"{tool_name} registered") + if verbose: + logger.info(f"{tool_name} registered") def has_tool(self, key: str) -> Tool: return key in self.tools diff --git a/metagpt/tools/ut_writer.py b/metagpt/tools/ut_writer.py index f2f2bf51c..243871aff 100644 --- a/metagpt/tools/ut_writer.py +++ b/metagpt/tools/ut_writer.py @@ -4,6 +4,7 @@ import json from pathlib import Path +from metagpt.config2 import config from metagpt.provider.openai_api import OpenAILLM as GPTAPI from metagpt.utils.common import awrite @@ -281,6 +282,6 @@ class UTGenerator: """Choose based on different calling methods""" result = "" if self.chatgpt_method == "API": - result = await GPTAPI().aask_code(messages=messages) + result = await GPTAPI(config.get_openai_llm()).aask_code(messages=messages) return result diff --git a/metagpt/tools/web_browser_engine.py b/metagpt/tools/web_browser_engine.py index abd84cc8d..411c1604b 100644 --- a/metagpt/tools/web_browser_engine.py +++ b/metagpt/tools/web_browser_engine.py @@ -1,14 +1,11 @@ #!/usr/bin/env python -""" -@Modified By: mashenquan, 2023/8/20. Remove global configuration `CONFIG`, enable configuration support for business isolation. -""" +# -*- coding: utf-8 -*- from __future__ import annotations import importlib from typing import Any, Callable, Coroutine, overload -from metagpt.config import CONFIG from metagpt.tools import WebBrowserEngineType from metagpt.utils.parse_html import WebPage @@ -16,10 +13,9 @@ from metagpt.utils.parse_html import WebPage class WebBrowserEngine: def __init__( self, - engine: WebBrowserEngineType | None = None, + engine: WebBrowserEngineType = WebBrowserEngineType.PLAYWRIGHT, run_func: Callable[..., Coroutine[Any, Any, WebPage | list[WebPage]]] | None = None, ): - engine = engine or CONFIG.web_browser_engine if engine is None: raise NotImplementedError diff --git a/metagpt/tools/web_browser_engine_playwright.py b/metagpt/tools/web_browser_engine_playwright.py index 15c8a78d7..7c33da923 100644 --- a/metagpt/tools/web_browser_engine_playwright.py +++ b/metagpt/tools/web_browser_engine_playwright.py @@ -1,7 +1,5 @@ #!/usr/bin/env python -""" -@Modified By: mashenquan, 2023/8/20. Remove global configuration `CONFIG`, enable configuration support for business isolation. -""" +# -*- coding: utf-8 -*- from __future__ import annotations @@ -27,20 +25,20 @@ class PlaywrightWrapper: def __init__( self, - browser_type: Literal["chromium", "firefox", "webkit"] | None = None, + browser_type: Literal["chromium", "firefox", "webkit"] | None = "chromium", launch_kwargs: dict | None = None, **kwargs, ) -> None: - from metagpt.config import CONFIG + from metagpt.config2 import ( + config, # avoid circular import error when importing tools" + ) - if browser_type is None: - browser_type = CONFIG.playwright_browser_type self.browser_type = browser_type launch_kwargs = launch_kwargs or {} - if CONFIG.global_proxy and "proxy" not in launch_kwargs: + if config.proxy and "proxy" not in launch_kwargs: args = launch_kwargs.get("args", []) if not any(str.startswith(i, "--proxy-server=") for i in args): - launch_kwargs["proxy"] = {"server": CONFIG.global_proxy} + launch_kwargs["proxy"] = {"server": config.proxy} self.launch_kwargs = launch_kwargs context_kwargs = {} if "ignore_https_errors" in kwargs: @@ -80,8 +78,8 @@ class PlaywrightWrapper: executable_path = Path(browser_type.executable_path) if not executable_path.exists() and "executable_path" not in self.launch_kwargs: kwargs = {} - if CONFIG.global_proxy: - kwargs["env"] = {"ALL_PROXY": CONFIG.global_proxy} + if config.proxy: + kwargs["env"] = {"ALL_PROXY": config.proxy} await _install_browsers(self.browser_type, **kwargs) if self._has_run_precheck: diff --git a/metagpt/tools/web_browser_engine_selenium.py b/metagpt/tools/web_browser_engine_selenium.py index 70b651935..02dd5c173 100644 --- a/metagpt/tools/web_browser_engine_selenium.py +++ b/metagpt/tools/web_browser_engine_selenium.py @@ -1,7 +1,5 @@ #!/usr/bin/env python -""" -@Modified By: mashenquan, 2023/8/20. Remove global configuration `CONFIG`, enable configuration support for business isolation. -""" +# -*- coding: utf-8 -*- from __future__ import annotations @@ -17,7 +15,7 @@ from selenium.webdriver.support.wait import WebDriverWait from webdriver_manager.core.download_manager import WDMDownloadManager from webdriver_manager.core.http import WDMHttpClient -from metagpt.config import CONFIG +from metagpt.config2 import config from metagpt.utils.parse_html import WebPage @@ -35,18 +33,16 @@ class SeleniumWrapper: def __init__( self, - browser_type: Literal["chrome", "firefox", "edge", "ie"] | None = None, + browser_type: Literal["chrome", "firefox", "edge", "ie"] = "chrome", launch_kwargs: dict | None = None, *, loop: asyncio.AbstractEventLoop | None = None, executor: futures.Executor | None = None, ) -> None: - if browser_type is None: - browser_type = CONFIG.selenium_browser_type self.browser_type = browser_type launch_kwargs = launch_kwargs or {} - if CONFIG.global_proxy and "proxy-server" not in launch_kwargs: - launch_kwargs["proxy-server"] = CONFIG.global_proxy + if config.proxy and "proxy-server" not in launch_kwargs: + launch_kwargs["proxy-server"] = config.proxy self.executable_path = launch_kwargs.pop("executable_path", None) self.launch_args = [f"--{k}={v}" for k, v in launch_kwargs.items()] @@ -97,8 +93,8 @@ _webdriver_manager_types = { class WDMHttpProxyClient(WDMHttpClient): def get(self, url, **kwargs): - if "proxies" not in kwargs and CONFIG.global_proxy: - kwargs["proxies"] = {"all_proxy": CONFIG.global_proxy} + if "proxies" not in kwargs and config.proxy: + kwargs["proxies"] = {"all_proxy": config.proxy} return super().get(url, **kwargs) diff --git a/metagpt/utils/common.py b/metagpt/utils/common.py index ed73cb061..543c627a3 100644 --- a/metagpt/utils/common.py +++ b/metagpt/utils/common.py @@ -29,7 +29,7 @@ from typing import Any, List, Tuple, Union import aiofiles import loguru from pydantic_core import to_jsonable_python -from tenacity import RetryCallState, _utils +from tenacity import RetryCallState, RetryError, _utils from metagpt.const import MESSAGE_ROUTE_TO_ALL from metagpt.logs import logger @@ -407,12 +407,12 @@ def any_to_str_set(val) -> set: return res -def is_subscribed(message: "Message", tags: set): +def is_send_to(message: "Message", addresses: set): """Return whether it's consumer""" if MESSAGE_ROUTE_TO_ALL in message.send_to: return True - for i in tags: + for i in addresses: if i in message.send_to: return True return False @@ -531,7 +531,7 @@ def role_raise_decorator(func): self.rc.memory.delete(self.latest_observed_msg) # raise again to make it captured outside raise Exception(format_trackback_info(limit=None)) - except Exception: + except Exception as e: if self.latest_observed_msg: logger.warning( "There is a exception in role's execution, in order to resume, " @@ -540,6 +540,12 @@ def role_raise_decorator(func): # remove role newest observed msg to make it observed again self.rc.memory.delete(self.latest_observed_msg) # raise again to make it captured outside + if isinstance(e, RetryError): + last_error = e.last_attempt._exception + name = any_to_str(last_error) + if re.match(r"^openai\.", name) or re.match(r"^httpx\.", name): + raise last_error + raise Exception(format_trackback_info(limit=None)) return wrapper diff --git a/metagpt/utils/cost_manager.py b/metagpt/utils/cost_manager.py index ce53f2285..7bf5154b6 100644 --- a/metagpt/utils/cost_manager.py +++ b/metagpt/utils/cost_manager.py @@ -80,3 +80,20 @@ class CostManager(BaseModel): def get_costs(self) -> Costs: """Get all costs""" return Costs(self.total_prompt_tokens, self.total_completion_tokens, self.total_cost, self.total_budget) + + +class TokenCostManager(CostManager): + """open llm model is self-host, it's free and without cost""" + + def update_cost(self, prompt_tokens, completion_tokens, model): + """ + Update the total cost, prompt tokens, and completion tokens. + + Args: + prompt_tokens (int): The number of tokens used in the prompt. + completion_tokens (int): The number of tokens used in the completion. + model (str): The model used for the API call. + """ + self.total_prompt_tokens += prompt_tokens + self.total_completion_tokens += completion_tokens + logger.info(f"prompt_tokens: {prompt_tokens}, completion_tokens: {completion_tokens}") diff --git a/metagpt/utils/dependency_file.py b/metagpt/utils/dependency_file.py index 7cf9a1d49..c8b3bc4a4 100644 --- a/metagpt/utils/dependency_file.py +++ b/metagpt/utils/dependency_file.py @@ -9,6 +9,7 @@ from __future__ import annotations import json +import re from pathlib import Path from typing import Set @@ -36,7 +37,9 @@ class DependencyFile: """Load dependencies from the file asynchronously.""" if not self._filename.exists(): return - self._dependencies = json.loads(await aread(self._filename)) + json_data = await aread(self._filename) + json_data = re.sub(r"\\+", "/", json_data) # Compatible with windows path + self._dependencies = json.loads(json_data) @handle_exception async def save(self): @@ -60,17 +63,20 @@ class DependencyFile: key = Path(filename).relative_to(root) except ValueError: key = filename - + skey = re.sub(r"\\+", "/", str(key)) # Compatible with windows path if dependencies: relative_paths = [] for i in dependencies: try: - relative_paths.append(str(Path(i).relative_to(root))) + s = str(Path(i).relative_to(root)) except ValueError: - relative_paths.append(str(i)) - self._dependencies[str(key)] = relative_paths - elif str(key) in self._dependencies: - del self._dependencies[str(key)] + s = str(i) + s = re.sub(r"\\+", "/", s) # Compatible with windows path + relative_paths.append(s) + + self._dependencies[skey] = relative_paths + elif skey in self._dependencies: + del self._dependencies[skey] if persist: await self.save() diff --git a/metagpt/utils/embedding.py b/metagpt/utils/embedding.py new file mode 100644 index 000000000..21d62948c --- /dev/null +++ b/metagpt/utils/embedding.py @@ -0,0 +1,16 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/1/4 20:58 +@Author : alexanderwu +@File : embedding.py +""" +from langchain_community.embeddings import OpenAIEmbeddings + +from metagpt.config2 import config + + +def get_embedding(): + llm = config.get_openai_llm() + embedding = OpenAIEmbeddings(openai_api_key=llm.api_key, openai_api_base=llm.base_url) + return embedding diff --git a/metagpt/utils/file_repository.py b/metagpt/utils/file_repository.py index 0ddca414d..d2a06963a 100644 --- a/metagpt/utils/file_repository.py +++ b/metagpt/utils/file_repository.py @@ -16,7 +16,6 @@ from typing import Dict, List, Set import aiofiles -from metagpt.config import CONFIG from metagpt.logs import logger from metagpt.schema import Document from metagpt.utils.common import aread @@ -46,7 +45,7 @@ class FileRepository: # Initializing self.workdir.mkdir(parents=True, exist_ok=True) - async def save(self, filename: Path | str, content, dependencies: List[str] = None): + async def save(self, filename: Path | str, content, dependencies: List[str] = None) -> Document: """Save content to a file and update its dependencies. :param filename: The filename or path within the repository. @@ -55,6 +54,7 @@ class FileRepository: """ pathname = self.workdir / filename pathname.parent.mkdir(parents=True, exist_ok=True) + content = content if content else "" # avoid `argument must be str, not None` to make it continue async with aiofiles.open(str(pathname), mode="w") as writer: await writer.write(content) logger.info(f"save to: {str(pathname)}") @@ -64,6 +64,8 @@ class FileRepository: await dependency_file.update(pathname, set(dependencies)) logger.info(f"update dependency: {str(pathname)}:{dependencies}") + return Document(root_path=str(self._relative_path), filename=str(filename), content=content) + async def get_dependency(self, filename: Path | str) -> Set[str]: """Get the dependencies of a file. @@ -99,21 +101,28 @@ class FileRepository: path_name = self.workdir / filename if not path_name.exists(): return None + if not path_name.is_file(): + return None doc.content = await aread(path_name) return doc - async def get_all(self) -> List[Document]: + async def get_all(self, filter_ignored=True) -> List[Document]: """Get the content of all files in the repository. :return: List of Document instances representing files. """ docs = [] - for root, dirs, files in os.walk(str(self.workdir)): - for file in files: - file_path = Path(root) / file - relative_path = file_path.relative_to(self.workdir) - doc = await self.get(relative_path) + if filter_ignored: + for f in self.all_files: + doc = await self.get(f) docs.append(doc) + else: + for root, dirs, files in os.walk(str(self.workdir)): + for file in files: + file_path = Path(root) / file + relative_path = file_path.relative_to(self.workdir) + doc = await self.get(relative_path) + docs.append(doc) return docs @property @@ -182,10 +191,20 @@ class FileRepository: """ current_time = datetime.now().strftime("%Y%m%d%H%M%S") return current_time - # guid_suffix = str(uuid.uuid4())[:8] - # return f"{current_time}x{guid_suffix}" - async def save_doc(self, doc: Document, with_suffix: str = None, dependencies: List[str] = None): + async def save_doc(self, doc: Document, dependencies: List[str] = None): + """Save content to a file and update its dependencies. + + :param doc: The Document instance to be saved. + :type doc: Document + :param dependencies: A list of dependencies for the saved file. + :type dependencies: List[str], optional + """ + + await self.save(filename=doc.filename, content=doc.content, dependencies=dependencies) + logger.debug(f"File Saved: {str(doc.filename)}") + + async def save_pdf(self, doc: Document, with_suffix: str = ".md", dependencies: List[str] = None): """Save a Document instance as a PDF file. This method converts the content of the Document instance to Markdown, @@ -203,70 +222,6 @@ class FileRepository: await self.save(filename=str(filename), content=json_to_markdown(m), dependencies=dependencies) logger.debug(f"File Saved: {str(filename)}") - @staticmethod - async def get_file(filename: Path | str, relative_path: Path | str = ".") -> Document | None: - """Retrieve a specific file from the file repository. - - :param filename: The name or path of the file to retrieve. - :type filename: Path or str - :param relative_path: The relative path within the file repository. - :type relative_path: Path or str, optional - :return: The document representing the file, or None if not found. - :rtype: Document or None - """ - file_repo = CONFIG.git_repo.new_file_repository(relative_path=relative_path) - return await file_repo.get(filename=filename) - - @staticmethod - async def get_all_files(relative_path: Path | str = ".") -> List[Document]: - """Retrieve all files from the file repository. - - :param relative_path: The relative path within the file repository. - :type relative_path: Path or str, optional - :return: A list of documents representing all files in the repository. - :rtype: List[Document] - """ - file_repo = CONFIG.git_repo.new_file_repository(relative_path=relative_path) - return await file_repo.get_all() - - @staticmethod - async def save_file(filename: Path | str, content, dependencies: List[str] = None, relative_path: Path | str = "."): - """Save a file to the file repository. - - :param filename: The name or path of the file to save. - :type filename: Path or str - :param content: The content of the file. - :param dependencies: A list of dependencies for the file. - :type dependencies: List[str], optional - :param relative_path: The relative path within the file repository. - :type relative_path: Path or str, optional - """ - file_repo = CONFIG.git_repo.new_file_repository(relative_path=relative_path) - return await file_repo.save(filename=filename, content=content, dependencies=dependencies) - - @staticmethod - async def save_as( - doc: Document, with_suffix: str = None, dependencies: List[str] = None, relative_path: Path | str = "." - ): - """Save a Document instance with optional modifications. - - This static method creates a new FileRepository, saves the Document instance - with optional modifications (such as a suffix), and logs the saved file. - - :param doc: The Document instance to be saved. - :type doc: Document - :param with_suffix: An optional suffix to append to the saved file's name. - :type with_suffix: str, optional - :param dependencies: A list of dependencies for the saved file. - :type dependencies: List[str], optional - :param relative_path: The relative path within the file repository. - :type relative_path: Path or str, optional - :return: A boolean indicating whether the save operation was successful. - :rtype: bool - """ - file_repo = CONFIG.git_repo.new_file_repository(relative_path=relative_path) - return await file_repo.save_doc(doc=doc, with_suffix=with_suffix, dependencies=dependencies) - async def delete(self, filename: Path | str): """Delete a file from the file repository. @@ -283,8 +238,3 @@ class FileRepository: dependency_file = await self._git_repo.get_dependency() await dependency_file.update(filename=pathname, dependencies=None) logger.info(f"remove dependency key: {str(pathname)}") - - @staticmethod - async def delete_file(filename: Path | str, relative_path: Path | str = "."): - file_repo = CONFIG.git_repo.new_file_repository(relative_path=relative_path) - await file_repo.delete(filename=filename) diff --git a/metagpt/utils/git_repository.py b/metagpt/utils/git_repository.py index e9855df05..16f675175 100644 --- a/metagpt/utils/git_repository.py +++ b/metagpt/utils/git_repository.py @@ -107,7 +107,10 @@ class GitRepository: def delete_repository(self): """Delete the entire repository directory.""" if self.is_valid: - shutil.rmtree(self._repository.working_dir) + try: + shutil.rmtree(self._repository.working_dir) + except Exception as e: + logger.exception(f"Failed delete git repo:{self.workdir}, error:{e}") @property def changed_files(self) -> Dict[str, str]: @@ -198,11 +201,21 @@ class GitRepository: new_path = self.workdir.parent / new_dir_name if new_path.exists(): logger.info(f"Delete directory {str(new_path)}") - shutil.rmtree(new_path) + try: + shutil.rmtree(new_path) + except Exception as e: + logger.warning(f"rm {str(new_path)} error: {e}") + if new_path.exists(): # Recheck for windows os + logger.warning(f"Failed to delete directory {str(new_path)}") + return try: shutil.move(src=str(self.workdir), dst=str(new_path)) except Exception as e: logger.warning(f"Move {str(self.workdir)} to {str(new_path)} error: {e}") + finally: + if not new_path.exists(): # Recheck for windows os + logger.warning(f"Failed to move {str(self.workdir)} to {str(new_path)}") + return logger.info(f"Rename directory {str(self.workdir)} to {str(new_path)}") self._repository = Repo(new_path) self._gitignore_rules = parse_gitignore(full_path=str(new_path / ".gitignore")) diff --git a/metagpt/utils/human_interaction.py b/metagpt/utils/human_interaction.py new file mode 100644 index 000000000..3b245cac8 --- /dev/null +++ b/metagpt/utils/human_interaction.py @@ -0,0 +1,107 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +# @Desc : human interaction to get required type text + +import json +from typing import Any, Tuple, Type + +from pydantic import BaseModel + +from metagpt.logs import logger +from metagpt.utils.common import import_class + + +class HumanInteraction(object): + stop_list = ("q", "quit", "exit") + + def multilines_input(self, prompt: str = "Enter: ") -> str: + logger.warning("Enter your content, use Ctrl-D or Ctrl-Z ( windows ) to save it.") + logger.info(f"{prompt}\n") + lines = [] + while True: + try: + line = input() + lines.append(line) + except EOFError: + break + return "".join(lines) + + def check_input_type(self, input_str: str, req_type: Type) -> Tuple[bool, Any]: + check_ret = True + if req_type == str: + # required_type = str, just return True + return check_ret, input_str + try: + input_str = input_str.strip() + data = json.loads(input_str) + except Exception: + return False, None + + actionnode_class = import_class("ActionNode", "metagpt.actions.action_node") # avoid circular import + tmp_key = "tmp" + tmp_cls = actionnode_class.create_model_class(class_name=tmp_key.upper(), mapping={tmp_key: (req_type, ...)}) + try: + _ = tmp_cls(**{tmp_key: data}) + except Exception: + check_ret = False + return check_ret, data + + def input_until_valid(self, prompt: str, req_type: Type) -> Any: + # check the input with req_type until it's ok + while True: + input_content = self.multilines_input(prompt) + check_ret, structure_content = self.check_input_type(input_content, req_type) + if check_ret: + break + else: + logger.error(f"Input content can't meet required_type: {req_type}, please Re-Enter.") + return structure_content + + def input_num_until_valid(self, num_max: int) -> int: + while True: + input_num = input("Enter the num of the interaction key: ") + input_num = input_num.strip() + if input_num in self.stop_list: + return input_num + try: + input_num = int(input_num) + if 0 <= input_num < num_max: + return input_num + except Exception: + pass + + def interact_with_instruct_content( + self, instruct_content: BaseModel, mapping: dict = dict(), interact_type: str = "review" + ) -> dict[str, Any]: + assert interact_type in ["review", "revise"] + assert instruct_content + instruct_content_dict = instruct_content.model_dump() + num_fields_map = dict(zip(range(0, len(instruct_content_dict)), instruct_content_dict.keys())) + logger.info( + f"\n{interact_type.upper()} interaction\n" + f"Interaction data: {num_fields_map}\n" + f"Enter the num to interact with corresponding field or `q`/`quit`/`exit` to stop interaction.\n" + f"Enter the field content until it meet field required type.\n" + ) + + interact_contents = {} + while True: + input_num = self.input_num_until_valid(len(instruct_content_dict)) + if input_num in self.stop_list: + logger.warning("Stop human interaction") + break + + field = num_fields_map.get(input_num) + logger.info(f"You choose to interact with field: {field}, and do a `{interact_type}` operation.") + + if interact_type == "review": + prompt = "Enter your review comment: " + req_type = str + else: + prompt = "Enter your revise content: " + req_type = mapping.get(field)[0] # revise need input content match the required_type + + field_content = self.input_until_valid(prompt=prompt, req_type=req_type) + interact_contents[field] = field_content + + return interact_contents diff --git a/metagpt/utils/make_sk_kernel.py b/metagpt/utils/make_sk_kernel.py index e0272ea13..283a682d6 100644 --- a/metagpt/utils/make_sk_kernel.py +++ b/metagpt/utils/make_sk_kernel.py @@ -13,20 +13,20 @@ from semantic_kernel.connectors.ai.open_ai.services.open_ai_chat_completion impo OpenAIChatCompletion, ) -from metagpt.config import CONFIG +from metagpt.config2 import config def make_sk_kernel(): kernel = sk.Kernel() - if CONFIG.OPENAI_API_TYPE == "azure": + if llm := config.get_azure_llm(): kernel.add_chat_service( "chat_completion", - AzureChatCompletion(CONFIG.DEPLOYMENT_NAME, CONFIG.OPENAI_BASE_URL, CONFIG.OPENAI_API_KEY), + AzureChatCompletion(llm.model, llm.base_url, llm.api_key), ) - else: + elif llm := config.get_openai_llm(): kernel.add_chat_service( "chat_completion", - OpenAIChatCompletion(CONFIG.OPENAI_API_MODEL, CONFIG.OPENAI_API_KEY), + OpenAIChatCompletion(llm.model, llm.api_key), ) return kernel diff --git a/metagpt/utils/mermaid.py b/metagpt/utils/mermaid.py index 235b4979c..e49fdea5d 100644 --- a/metagpt/utils/mermaid.py +++ b/metagpt/utils/mermaid.py @@ -4,7 +4,6 @@ @Time : 2023/7/4 10:53 @Author : alexanderwu alitrack @File : mermaid.py -@Modified By: mashenquan, 2023/8/20. Remove global configuration `CONFIG`, enable configuration support for business isolation. """ import asyncio import os @@ -12,12 +11,12 @@ from pathlib import Path import aiofiles -from metagpt.config import CONFIG +from metagpt.config2 import config from metagpt.logs import logger from metagpt.utils.common import check_cmd_exists -async def mermaid_to_file(mermaid_code, output_file_without_suffix, width=2048, height=2048) -> int: +async def mermaid_to_file(engine, mermaid_code, output_file_without_suffix, width=2048, height=2048) -> int: """suffix: png/svg/pdf :param mermaid_code: mermaid code @@ -35,9 +34,8 @@ async def mermaid_to_file(mermaid_code, output_file_without_suffix, width=2048, await f.write(mermaid_code) # tmp.write_text(mermaid_code, encoding="utf-8") - engine = CONFIG.mermaid_engine.lower() if engine == "nodejs": - if check_cmd_exists(CONFIG.mmdc) != 0: + if check_cmd_exists(config.mmdc) != 0: logger.warning( "RUN `npm install -g @mermaid-js/mermaid-cli` to install mmdc," "or consider changing MERMAID_ENGINE to `playwright`, `pyppeteer`, or `ink`." @@ -49,11 +47,11 @@ async def mermaid_to_file(mermaid_code, output_file_without_suffix, width=2048, # Call the `mmdc` command to convert the Mermaid code to a PNG logger.info(f"Generating {output_file}..") - if CONFIG.puppeteer_config: + if config.puppeteer_config: commands = [ - CONFIG.mmdc, + config.mmdc, "-p", - CONFIG.puppeteer_config, + config.puppeteer_config, "-i", str(tmp), "-o", @@ -64,7 +62,7 @@ async def mermaid_to_file(mermaid_code, output_file_without_suffix, width=2048, str(height), ] else: - commands = [CONFIG.mmdc, "-i", str(tmp), "-o", output_file, "-w", str(width), "-H", str(height)] + commands = [config.mmdc, "-i", str(tmp), "-o", output_file, "-w", str(width), "-H", str(height)] process = await asyncio.create_subprocess_shell( " ".join(commands), stdout=asyncio.subprocess.PIPE, stderr=asyncio.subprocess.PIPE ) diff --git a/metagpt/utils/mmdc_pyppeteer.py b/metagpt/utils/mmdc_pyppeteer.py index 7125cafc5..d80098b7d 100644 --- a/metagpt/utils/mmdc_pyppeteer.py +++ b/metagpt/utils/mmdc_pyppeteer.py @@ -10,7 +10,7 @@ from urllib.parse import urljoin from pyppeteer import launch -from metagpt.config import CONFIG +from metagpt.config2 import config from metagpt.logs import logger @@ -30,10 +30,10 @@ async def mermaid_to_file(mermaid_code, output_file_without_suffix, width=2048, suffixes = ["png", "svg", "pdf"] __dirname = os.path.dirname(os.path.abspath(__file__)) - if CONFIG.pyppeteer_executable_path: + if config.pyppeteer_executable_path: browser = await launch( headless=True, - executablePath=CONFIG.pyppeteer_executable_path, + executablePath=config.pyppeteer_executable_path, args=["--disable-extensions", "--no-sandbox"], ) else: diff --git a/metagpt/utils/project_repo.py b/metagpt/utils/project_repo.py new file mode 100644 index 000000000..72bca7ea0 --- /dev/null +++ b/metagpt/utils/project_repo.py @@ -0,0 +1,139 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/1/8 +@Author : mashenquan +@File : project_repo.py +@Desc : Wrapper for GitRepository and FileRepository of project. + Implementation of Chapter 4.6 of https://deepwisdom.feishu.cn/wiki/CUK4wImd7id9WlkQBNscIe9cnqh +""" +from __future__ import annotations + +from pathlib import Path + +from metagpt.const import ( + CLASS_VIEW_FILE_REPO, + CODE_PLAN_AND_CHANGE_FILE_REPO, + CODE_PLAN_AND_CHANGE_PDF_FILE_REPO, + CODE_SUMMARIES_FILE_REPO, + CODE_SUMMARIES_PDF_FILE_REPO, + COMPETITIVE_ANALYSIS_FILE_REPO, + DATA_API_DESIGN_FILE_REPO, + DOCS_FILE_REPO, + GRAPH_REPO_FILE_REPO, + PRD_PDF_FILE_REPO, + PRDS_FILE_REPO, + REQUIREMENT_FILENAME, + RESOURCES_FILE_REPO, + SD_OUTPUT_FILE_REPO, + SEQ_FLOW_FILE_REPO, + SYSTEM_DESIGN_FILE_REPO, + SYSTEM_DESIGN_PDF_FILE_REPO, + TASK_FILE_REPO, + TASK_PDF_FILE_REPO, + TEST_CODES_FILE_REPO, + TEST_OUTPUTS_FILE_REPO, +) +from metagpt.utils.file_repository import FileRepository +from metagpt.utils.git_repository import GitRepository + + +class DocFileRepositories(FileRepository): + prd: FileRepository + system_design: FileRepository + task: FileRepository + code_summary: FileRepository + graph_repo: FileRepository + class_view: FileRepository + code_plan_and_change: FileRepository + + def __init__(self, git_repo): + super().__init__(git_repo=git_repo, relative_path=DOCS_FILE_REPO) + + self.prd = git_repo.new_file_repository(relative_path=PRDS_FILE_REPO) + self.system_design = git_repo.new_file_repository(relative_path=SYSTEM_DESIGN_FILE_REPO) + self.task = git_repo.new_file_repository(relative_path=TASK_FILE_REPO) + self.code_summary = git_repo.new_file_repository(relative_path=CODE_SUMMARIES_FILE_REPO) + self.graph_repo = git_repo.new_file_repository(relative_path=GRAPH_REPO_FILE_REPO) + self.class_view = git_repo.new_file_repository(relative_path=CLASS_VIEW_FILE_REPO) + self.code_plan_and_change = git_repo.new_file_repository(relative_path=CODE_PLAN_AND_CHANGE_FILE_REPO) + + +class ResourceFileRepositories(FileRepository): + competitive_analysis: FileRepository + data_api_design: FileRepository + seq_flow: FileRepository + system_design: FileRepository + prd: FileRepository + api_spec_and_task: FileRepository + code_summary: FileRepository + sd_output: FileRepository + code_plan_and_change: FileRepository + + def __init__(self, git_repo): + super().__init__(git_repo=git_repo, relative_path=RESOURCES_FILE_REPO) + + self.competitive_analysis = git_repo.new_file_repository(relative_path=COMPETITIVE_ANALYSIS_FILE_REPO) + self.data_api_design = git_repo.new_file_repository(relative_path=DATA_API_DESIGN_FILE_REPO) + self.seq_flow = git_repo.new_file_repository(relative_path=SEQ_FLOW_FILE_REPO) + self.system_design = git_repo.new_file_repository(relative_path=SYSTEM_DESIGN_PDF_FILE_REPO) + self.prd = git_repo.new_file_repository(relative_path=PRD_PDF_FILE_REPO) + self.api_spec_and_task = git_repo.new_file_repository(relative_path=TASK_PDF_FILE_REPO) + self.code_summary = git_repo.new_file_repository(relative_path=CODE_SUMMARIES_PDF_FILE_REPO) + self.sd_output = git_repo.new_file_repository(relative_path=SD_OUTPUT_FILE_REPO) + self.code_plan_and_change = git_repo.new_file_repository(relative_path=CODE_PLAN_AND_CHANGE_PDF_FILE_REPO) + + +class ProjectRepo(FileRepository): + def __init__(self, root: str | Path | GitRepository): + if isinstance(root, str) or isinstance(root, Path): + git_repo_ = GitRepository(local_path=Path(root)) + elif isinstance(root, GitRepository): + git_repo_ = root + else: + raise ValueError("Invalid root") + super().__init__(git_repo=git_repo_, relative_path=Path(".")) + self._git_repo = git_repo_ + self.docs = DocFileRepositories(self._git_repo) + self.resources = ResourceFileRepositories(self._git_repo) + self.tests = self._git_repo.new_file_repository(relative_path=TEST_CODES_FILE_REPO) + self.test_outputs = self._git_repo.new_file_repository(relative_path=TEST_OUTPUTS_FILE_REPO) + self._srcs_path = None + + @property + async def requirement(self): + return await self.docs.get(filename=REQUIREMENT_FILENAME) + + @property + def git_repo(self) -> GitRepository: + return self._git_repo + + @property + def workdir(self) -> Path: + return Path(self.git_repo.workdir) + + @property + def srcs(self) -> FileRepository: + if not self._srcs_path: + raise ValueError("Call with_srcs first.") + return self._git_repo.new_file_repository(self._srcs_path) + + def code_files_exists(self) -> bool: + git_workdir = self.git_repo.workdir + src_workdir = git_workdir / git_workdir.name + if not src_workdir.exists(): + return False + code_files = self.with_src_path(path=git_workdir / git_workdir.name).srcs.all_files + if not code_files: + return False + + def with_src_path(self, path: str | Path) -> ProjectRepo: + try: + self._srcs_path = Path(path).relative_to(self.workdir) + except ValueError: + self._srcs_path = Path(path) + return self + + @property + def src_relative_path(self) -> Path | None: + return self._srcs_path diff --git a/metagpt/utils/redis.py b/metagpt/utils/redis.py index 10f33285c..7a640563a 100644 --- a/metagpt/utils/redis.py +++ b/metagpt/utils/redis.py @@ -12,26 +12,25 @@ from datetime import timedelta import aioredis # https://aioredis.readthedocs.io/en/latest/getting-started/ -from metagpt.config import CONFIG +from metagpt.configs.redis_config import RedisConfig from metagpt.logs import logger class Redis: - def __init__(self): + def __init__(self, config: RedisConfig = None): + self.config = config self._client = None async def _connect(self, force=False): if self._client and not force: return True - if not self.is_configured: - return False try: self._client = await aioredis.from_url( - f"redis://{CONFIG.REDIS_HOST}:{CONFIG.REDIS_PORT}", - username=CONFIG.REDIS_USER, - password=CONFIG.REDIS_PASSWORD, - db=CONFIG.REDIS_DB, + self.config.to_url(), + username=self.config.username, + password=self.config.password, + db=self.config.db, ) return True except Exception as e: @@ -62,18 +61,3 @@ class Redis: return await self._client.close() self._client = None - - @property - def is_valid(self) -> bool: - return self._client is not None - - @property - def is_configured(self) -> bool: - return bool( - CONFIG.REDIS_HOST - and CONFIG.REDIS_HOST != "YOUR_REDIS_HOST" - and CONFIG.REDIS_PORT - and CONFIG.REDIS_PORT != "YOUR_REDIS_PORT" - and CONFIG.REDIS_DB is not None - and CONFIG.REDIS_PASSWORD is not None - ) diff --git a/metagpt/utils/repair_llm_raw_output.py b/metagpt/utils/repair_llm_raw_output.py index a96c3dce0..82b2dd5b1 100644 --- a/metagpt/utils/repair_llm_raw_output.py +++ b/metagpt/utils/repair_llm_raw_output.py @@ -9,7 +9,7 @@ from typing import Callable, Union import regex as re from tenacity import RetryCallState, retry, stop_after_attempt, wait_fixed -from metagpt.config import CONFIG +from metagpt.config2 import config from metagpt.logs import logger from metagpt.utils.custom_decoder import CustomDecoder @@ -120,6 +120,15 @@ def repair_json_format(output: str) -> str: elif output.startswith("{") and output.endswith("]"): output = output[:-1] + "}" + # remove `#` in output json str, usually appeared in `glm-4` + arr = output.split("\n") + new_arr = [] + for line in arr: + idx = line.find("#") + if idx >= 0: + line = line[:idx] + new_arr.append(line) + output = "\n".join(new_arr) return output @@ -152,7 +161,7 @@ def repair_llm_raw_output(output: str, req_keys: list[str], repair_type: RepairT target: { xxx } output: { xxx }] """ - if not CONFIG.repair_llm_output: + if not config.repair_llm_output: return output # do the repairation usually for non-openai models @@ -168,15 +177,17 @@ def repair_invalid_json(output: str, error: str) -> str: example 1. json.decoder.JSONDecodeError: Expecting ',' delimiter: line 154 column 1 (char 2765) example 2. xxx.JSONDecodeError: Expecting property name enclosed in double quotes: line 14 column 1 (char 266) """ - pattern = r"line ([0-9]+)" + pattern = r"line ([0-9]+) column ([0-9]+)" matches = re.findall(pattern, error, re.DOTALL) if len(matches) > 0: - line_no = int(matches[0]) - 1 + line_no = int(matches[0][0]) - 1 + col_no = int(matches[0][1]) - 1 # due to CustomDecoder can handle `"": ''` or `'': ""`, so convert `"""` -> `"`, `'''` -> `'` output = output.replace('"""', '"').replace("'''", '"') arr = output.split("\n") + rline = arr[line_no] # raw line line = arr[line_no].strip() # different general problems if line.endswith("],"): @@ -187,9 +198,12 @@ def repair_invalid_json(output: str, error: str) -> str: new_line = line.replace("}", "") elif line.endswith("},") and output.endswith("},"): new_line = line[:-1] - elif '",' not in line and "," not in line: + elif (rline[col_no] in ["'", '"']) and (line.startswith('"') or line.startswith("'")) and "," not in line: + # problem, `"""` or `'''` without `,` + new_line = f",{line}" + elif '",' not in line and "," not in line and '"' not in line: new_line = f'{line}",' - elif "," not in line: + elif not line.endswith(","): # problem, miss char `,` at the end. new_line = f"{line}," elif "," in line and len(line) == 1: @@ -231,7 +245,7 @@ def run_after_exp_and_passon_next_retry(logger: "loguru.Logger") -> Callable[["R func_param_output = retry_state.kwargs.get("output", "") exp_str = str(retry_state.outcome.exception()) - fix_str = "try to fix it, " if CONFIG.repair_llm_output else "" + fix_str = "try to fix it, " if config.repair_llm_output else "" logger.warning( f"parse json from content inside [CONTENT][/CONTENT] failed at retry " f"{retry_state.attempt_number}, {fix_str}exp: {exp_str}" @@ -244,7 +258,7 @@ def run_after_exp_and_passon_next_retry(logger: "loguru.Logger") -> Callable[["R @retry( - stop=stop_after_attempt(3 if CONFIG.repair_llm_output else 0), + stop=stop_after_attempt(3 if config.repair_llm_output else 0), wait=wait_fixed(1), after=run_after_exp_and_passon_next_retry(logger), ) diff --git a/metagpt/utils/s3.py b/metagpt/utils/s3.py index 2a2c1a31c..c0afbb2f5 100644 --- a/metagpt/utils/s3.py +++ b/metagpt/utils/s3.py @@ -8,7 +8,7 @@ from typing import Optional import aioboto3 import aiofiles -from metagpt.config import CONFIG +from metagpt.config2 import S3Config from metagpt.const import BASE64_FORMAT from metagpt.logs import logger @@ -16,13 +16,14 @@ from metagpt.logs import logger class S3: """A class for interacting with Amazon S3 storage.""" - def __init__(self): + def __init__(self, config: S3Config): self.session = aioboto3.Session() + self.config = config self.auth_config = { "service_name": "s3", - "aws_access_key_id": CONFIG.S3_ACCESS_KEY, - "aws_secret_access_key": CONFIG.S3_SECRET_KEY, - "endpoint_url": CONFIG.S3_ENDPOINT_URL, + "aws_access_key_id": config.access_key, + "aws_secret_access_key": config.secret_key, + "endpoint_url": config.endpoint, } async def upload_file( @@ -139,8 +140,8 @@ class S3: data = base64.b64decode(data) if format == BASE64_FORMAT else data.encode(encoding="utf-8") await file.write(data) - bucket = CONFIG.S3_BUCKET - object_pathname = CONFIG.S3_BUCKET or "system" + bucket = self.config.bucket + object_pathname = self.config.bucket or "system" object_pathname += f"/{object_name}" object_pathname = os.path.normpath(object_pathname) await self.upload_file(bucket=bucket, local_path=str(pathname), object_name=object_pathname) @@ -151,20 +152,3 @@ class S3: logger.exception(f"{e}, stack:{traceback.format_exc()}") pathname.unlink(missing_ok=True) return None - - @property - def is_valid(self): - return self.is_configured - - @property - def is_configured(self) -> bool: - return bool( - CONFIG.S3_ACCESS_KEY - and CONFIG.S3_ACCESS_KEY != "YOUR_S3_ACCESS_KEY" - and CONFIG.S3_SECRET_KEY - and CONFIG.S3_SECRET_KEY != "YOUR_S3_SECRET_KEY" - and CONFIG.S3_ENDPOINT_URL - and CONFIG.S3_ENDPOINT_URL != "YOUR_S3_ENDPOINT_URL" - and CONFIG.S3_BUCKET - and CONFIG.S3_BUCKET != "YOUR_S3_BUCKET" - ) diff --git a/metagpt/utils/token_counter.py b/metagpt/utils/token_counter.py index a1b74a074..94506e373 100644 --- a/metagpt/utils/token_counter.py +++ b/metagpt/utils/token_counter.py @@ -4,10 +4,11 @@ @Time : 2023/5/18 00:40 @Author : alexanderwu @File : token_counter.py -ref1: https://github.com/openai/openai-cookbook/blob/main/examples/How_to_count_tokens_with_tiktoken.ipynb -ref2: https://github.com/Significant-Gravitas/Auto-GPT/blob/master/autogpt/llm/token_counter.py -ref3: https://github.com/hwchase17/langchain/blob/master/langchain/chat_models/openai.py -ref4: https://ai.google.dev/models/gemini +ref1: https://openai.com/pricing +ref2: https://github.com/openai/openai-cookbook/blob/main/examples/How_to_count_tokens_with_tiktoken.ipynb +ref3: https://github.com/Significant-Gravitas/Auto-GPT/blob/master/autogpt/llm/token_counter.py +ref4: https://github.com/hwchase17/langchain/blob/master/langchain/chat_models/openai.py +ref5: https://ai.google.dev/models/gemini """ import tiktoken @@ -25,9 +26,13 @@ TOKEN_COSTS = { "gpt-4-32k": {"prompt": 0.06, "completion": 0.12}, "gpt-4-32k-0314": {"prompt": 0.06, "completion": 0.12}, "gpt-4-0613": {"prompt": 0.06, "completion": 0.12}, + "gpt-4-turbo-preview": {"prompt": 0.01, "completion": 0.03}, + "gpt-4-0125-preview": {"prompt": 0.01, "completion": 0.03}, "gpt-4-1106-preview": {"prompt": 0.01, "completion": 0.03}, + "gpt-4-1106-vision-preview": {"prompt": 0.01, "completion": 0.03}, "text-embedding-ada-002": {"prompt": 0.0004, "completion": 0.0}, - "chatglm_turbo": {"prompt": 0.0, "completion": 0.00069}, # 32k version, prompt + completion tokens=0.005￥/k-tokens + "glm-3-turbo": {"prompt": 0.0, "completion": 0.0007}, # 128k version, prompt + completion tokens=0.005￥/k-tokens + "glm-4": {"prompt": 0.0, "completion": 0.014}, # 128k version, prompt + completion tokens=0.1￥/k-tokens "gemini-pro": {"prompt": 0.00025, "completion": 0.0005}, } @@ -46,7 +51,10 @@ TOKEN_MAX = { "gpt-4-32k": 32768, "gpt-4-32k-0314": 32768, "gpt-4-0613": 8192, + "gpt-4-turbo-preview": 128000, + "gpt-4-0125-preview": 128000, "gpt-4-1106-preview": 128000, + "gpt-4-1106-vision-preview": 128000, "text-embedding-ada-002": 8192, "chatglm_turbo": 32768, "gemini-pro": 32768, @@ -71,7 +79,10 @@ def count_message_tokens(messages, model="gpt-3.5-turbo-0613"): "gpt-4-32k-0314", "gpt-4-0613", "gpt-4-32k-0613", + "gpt-4-turbo-preview", + "gpt-4-0125-preview", "gpt-4-1106-preview", + "gpt-4-1106-vision-preview", }: tokens_per_message = 3 # # every reply is primed with <|start|>assistant<|message|> tokens_per_name = 1 diff --git a/metagpt/utils/yaml_model.py b/metagpt/utils/yaml_model.py new file mode 100644 index 000000000..8f2d22c3d --- /dev/null +++ b/metagpt/utils/yaml_model.py @@ -0,0 +1,48 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/1/4 10:18 +@Author : alexanderwu +@File : YamlModel.py +""" +from pathlib import Path +from typing import Dict, Optional + +import yaml +from pydantic import BaseModel, model_validator + + +class YamlModel(BaseModel): + """Base class for yaml model""" + + extra_fields: Optional[Dict[str, str]] = None + + @classmethod + def read_yaml(cls, file_path: Path, encoding: str = "utf-8") -> Dict: + """Read yaml file and return a dict""" + if not file_path.exists(): + return {} + with open(file_path, "r", encoding=encoding) as file: + return yaml.safe_load(file) + + @classmethod + def from_yaml_file(cls, file_path: Path) -> "YamlModel": + """Read yaml file and return a YamlModel instance""" + return cls(**cls.read_yaml(file_path)) + + def to_yaml_file(self, file_path: Path, encoding: str = "utf-8") -> None: + """Dump YamlModel instance to yaml file""" + with open(file_path, "w", encoding=encoding) as file: + yaml.dump(self.model_dump(), file) + + +class YamlModelWithoutDefault(YamlModel): + """YamlModel without default values""" + + @model_validator(mode="before") + @classmethod + def check_not_default_config(cls, values): + """Check if there is any default config in config.yaml""" + if any(["YOUR" in v for v in values]): + raise ValueError("Please set your config in config.yaml") + return values diff --git a/requirements.txt b/requirements.txt index 016c2f5d5..66b3c9fc0 100644 --- a/requirements.txt +++ b/requirements.txt @@ -14,7 +14,7 @@ lancedb==0.4.0 langchain==0.0.352 loguru==0.6.0 meilisearch==0.21.0 -numpy==1.24.3 +numpy>=1.24.3 openai==1.6.0 openpyxl beautifulsoup4==4.12.2 @@ -49,7 +49,7 @@ aioredis~=2.0.1 # Used by metagpt/utils/redis.py websocket-client==1.6.2 aiofiles==23.2.1 gitpython==3.1.40 -zhipuai==1.0.7 +zhipuai==2.0.1 rich==13.6.0 nbclient==0.9.0 nbformat==5.9.2 diff --git a/setup.py b/setup.py index 0439d6cd4..d1445e3f8 100644 --- a/setup.py +++ b/setup.py @@ -46,8 +46,8 @@ extras_require["test"] = [ "chromadb==0.4.14", "gradio==3.0.0", "grpcio-status==1.48.2", - "mock==5.1.0", "pylint==3.0.3", + "pybrowsers", ] extras_require["pyppeteer"] = [ @@ -58,7 +58,7 @@ extras_require["dev"] = (["pylint~=3.0.3", "black~=23.3.0", "isort~=5.12.0", "pr setup( name="metagpt", - version="0.6.0", + version="0.7.0", description="The Multi-Agent Framework", long_description=long_description, long_description_content_type="text/markdown", diff --git a/tests/conftest.py b/tests/conftest.py index 7dec506bb..c69fca15e 100644 --- a/tests/conftest.py +++ b/tests/conftest.py @@ -12,14 +12,19 @@ import logging import os import re import uuid +from typing import Callable import pytest -from metagpt.config import CONFIG, Config from metagpt.const import DEFAULT_WORKSPACE_ROOT, TEST_DATA_PATH +from metagpt.context import Context as MetagptContext from metagpt.llm import LLM from metagpt.logs import logger from metagpt.utils.git_repository import GitRepository +from metagpt.utils.project_repo import ProjectRepo +from tests.mock.mock_aiohttp import MockAioResponse +from tests.mock.mock_curl_cffi import MockCurlCffiResponse +from tests.mock.mock_httplib2 import MockHttplib2Response from tests.mock.mock_llm import MockLLM RSP_CACHE_NEW = {} # used globally for producing new and useful only response cache @@ -30,7 +35,6 @@ ALLOW_OPENAI_API_CALL = int( @pytest.fixture(scope="session") def rsp_cache(): - # model_version = CONFIG.openai_api_model rsp_cache_file_path = TEST_DATA_PATH / "rsp_cache.json" # read repo-provided new_rsp_cache_file_path = TEST_DATA_PATH / "rsp_cache_new.json" # exporting a new copy if os.path.exists(rsp_cache_file_path): @@ -76,7 +80,7 @@ def llm_mock(rsp_cache, mocker, request): class Context: def __init__(self): self._llm_ui = None - self._llm_api = LLM(provider=CONFIG.get_default_llm_provider_enum()) + self._llm_api = LLM() @property def llm_api(self): @@ -90,9 +94,9 @@ class Context: @pytest.fixture(scope="package") def llm_api(): logger.info("Setting up the test") - _context = Context() + g_context = Context() - yield _context.llm_api + yield g_context.llm_api logger.info("Tearing down the test") @@ -125,7 +129,7 @@ def proxy(): server = await asyncio.start_server(handle_client, "127.0.0.1", 0) return server, "http://{}:{}".format(*server.sockets[0].getsockname()) - return proxy_func() + return proxy_func # see https://github.com/Delgan/loguru/issues/59#issuecomment-466591978 @@ -139,23 +143,25 @@ def loguru_caplog(caplog): yield caplog -# init & dispose git repo -@pytest.fixture(scope="function", autouse=True) -def setup_and_teardown_git_repo(request): - CONFIG.git_repo = GitRepository(local_path=DEFAULT_WORKSPACE_ROOT / f"unittest/{uuid.uuid4().hex}") - CONFIG.git_reinit = True +@pytest.fixture(scope="function") +def context(request): + ctx = MetagptContext() + ctx.git_repo = GitRepository(local_path=DEFAULT_WORKSPACE_ROOT / f"unittest/{uuid.uuid4().hex}") + ctx.repo = ProjectRepo(ctx.git_repo) # Destroy git repo at the end of the test session. def fin(): - CONFIG.git_repo.delete_repository() + if ctx.git_repo: + ctx.git_repo.delete_repository() # Register the function for destroying the environment. request.addfinalizer(fin) + return ctx @pytest.fixture(scope="session", autouse=True) def init_config(): - Config() + pass @pytest.fixture(scope="function") @@ -165,39 +171,63 @@ def new_filename(mocker): yield mocker +@pytest.fixture(scope="session") +def search_rsp_cache(): + rsp_cache_file_path = TEST_DATA_PATH / "search_rsp_cache.json" # read repo-provided + if os.path.exists(rsp_cache_file_path): + with open(rsp_cache_file_path, "r") as f1: + rsp_cache_json = json.load(f1) + else: + rsp_cache_json = {} + yield rsp_cache_json + with open(rsp_cache_file_path, "w") as f2: + json.dump(rsp_cache_json, f2, indent=4, ensure_ascii=False) + + @pytest.fixture def aiohttp_mocker(mocker): - class MockAioResponse: - async def json(self, *args, **kwargs): - return self._json - - def set_json(self, json): - self._json = json - - response = MockAioResponse() - - class MockCTXMng: - async def __aenter__(self): - return response - - async def __aexit__(self, *args, **kwargs): - pass - - def __await__(self): - yield - return response - - def mock_request(self, method, url, **kwargs): - return MockCTXMng() + MockResponse = type("MockResponse", (MockAioResponse,), {}) def wrap(method): def run(self, url, **kwargs): - return mock_request(self, method, url, **kwargs) + return MockResponse(self, method, url, **kwargs) return run - mocker.patch("aiohttp.ClientSession.request", mock_request) + mocker.patch("aiohttp.ClientSession.request", MockResponse) for i in ["get", "post", "delete", "patch"]: mocker.patch(f"aiohttp.ClientSession.{i}", wrap(i)) + yield MockResponse - yield response + +@pytest.fixture +def curl_cffi_mocker(mocker): + MockResponse = type("MockResponse", (MockCurlCffiResponse,), {}) + + def request(self, *args, **kwargs): + return MockResponse(self, *args, **kwargs) + + mocker.patch("curl_cffi.requests.Session.request", request) + yield MockResponse + + +@pytest.fixture +def httplib2_mocker(mocker): + MockResponse = type("MockResponse", (MockHttplib2Response,), {}) + + def request(self, *args, **kwargs): + return MockResponse(self, *args, **kwargs) + + mocker.patch("httplib2.Http.request", request) + yield MockResponse + + +@pytest.fixture +def search_engine_mocker(aiohttp_mocker, curl_cffi_mocker, httplib2_mocker, search_rsp_cache): + # aiohttp_mocker: serpapi/serper + # httplib2_mocker: google + # curl_cffi_mocker: ddg + check_funcs: dict[tuple[str, str], Callable[[dict], str]] = {} + aiohttp_mocker.rsp_cache = httplib2_mocker.rsp_cache = curl_cffi_mocker.rsp_cache = search_rsp_cache + aiohttp_mocker.check_funcs = httplib2_mocker.check_funcs = curl_cffi_mocker.check_funcs = check_funcs + yield check_funcs diff --git a/tests/data/audio/hello.mp3 b/tests/data/audio/hello.mp3 new file mode 100644 index 000000000..7b3aab0a4 Binary files /dev/null and b/tests/data/audio/hello.mp3 differ diff --git a/tests/data/demo_project/dependencies.json b/tests/data/demo_project/dependencies.json index cfcf6c165..738e5d9be 100644 --- a/tests/data/demo_project/dependencies.json +++ b/tests/data/demo_project/dependencies.json @@ -1 +1 @@ -{"docs/system_design/20231221155954.json": ["docs/prds/20231221155954.json"], "docs/tasks/20231221155954.json": ["docs/system_design/20231221155954.json"], "game_2048/game.py": ["docs/tasks/20231221155954.json", "docs/system_design/20231221155954.json"], "game_2048/main.py": ["docs/tasks/20231221155954.json", "docs/system_design/20231221155954.json"], "resources/code_summaries/20231221155954.md": ["docs/tasks/20231221155954.json", "game_2048/game.py", "docs/system_design/20231221155954.json", "game_2048/main.py"], "docs/code_summaries/20231221155954.json": ["docs/tasks/20231221155954.json", "game_2048/game.py", "docs/system_design/20231221155954.json", "game_2048/main.py"], "tests/test_main.py": ["game_2048/main.py"], "tests/test_game.py": ["game_2048/game.py"], "test_outputs/test_main.py.json": ["game_2048/main.py", "tests/test_main.py"], "test_outputs/test_game.py.json": ["game_2048/game.py", "tests/test_game.py"]} \ No newline at end of file +{"docs/system_design/20231221155954.json": ["docs/prd/20231221155954.json"], "docs/task/20231221155954.json": ["docs/system_design/20231221155954.json"], "game_2048/game.py": ["docs/task/20231221155954.json", "docs/system_design/20231221155954.json"], "game_2048/main.py": ["docs/task/20231221155954.json", "docs/system_design/20231221155954.json"], "resources/code_summary/20231221155954.md": ["docs/task/20231221155954.json", "game_2048/game.py", "docs/system_design/20231221155954.json", "game_2048/main.py"], "docs/code_summary/20231221155954.json": ["docs/task/20231221155954.json", "game_2048/game.py", "docs/system_design/20231221155954.json", "game_2048/main.py"], "tests/test_main.py": ["game_2048/main.py"], "tests/test_game.py": ["game_2048/game.py"], "test_outputs/test_main.py.json": ["game_2048/main.py", "tests/test_main.py"], "test_outputs/test_game.py.json": ["game_2048/game.py", "tests/test_game.py"]} \ No newline at end of file diff --git a/tests/data/incremental_dev_project/Gomoku.zip b/tests/data/incremental_dev_project/Gomoku.zip new file mode 100644 index 000000000..23649565a Binary files /dev/null and b/tests/data/incremental_dev_project/Gomoku.zip differ diff --git a/tests/data/incremental_dev_project/dice_simulator_new.zip b/tests/data/incremental_dev_project/dice_simulator_new.zip new file mode 100644 index 000000000..4752ab4c5 Binary files /dev/null and b/tests/data/incremental_dev_project/dice_simulator_new.zip differ diff --git a/tests/data/incremental_dev_project/mock.py b/tests/data/incremental_dev_project/mock.py new file mode 100644 index 000000000..f2eb71359 --- /dev/null +++ b/tests/data/incremental_dev_project/mock.py @@ -0,0 +1,466 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/01/17 +@Author : mannaandpoem +@File : mock.py +""" +NEW_REQUIREMENT_SAMPLE = """ +Adding graphical interface functionality to enhance the user experience in the number-guessing game. The existing number-guessing game currently relies on command-line input for numbers. The goal is to introduce a graphical interface to improve the game's usability and visual appeal +""" + +PRD_SAMPLE = """ +## Language + +en_us + +## Programming Language + +Python + +## Original Requirements + +Make a simple number guessing game + +## Product Goals + +- Ensure a user-friendly interface for the game +- Provide a challenging yet enjoyable game experience +- Design the game to be easily extendable for future features + +## User Stories + +- As a player, I want to guess numbers and receive feedback on whether my guess is too high or too low +- As a player, I want to be able to set the difficulty level by choosing the range of possible numbers +- As a player, I want to see my previous guesses to strategize my next guess +- As a player, I want to know how many attempts it took me to guess the number once I get it right + +## Competitive Analysis + +- Guess The Number Game A: Basic text interface, no difficulty levels +- Number Master B: Has difficulty levels, but cluttered interface +- Quick Guess C: Sleek design, but lacks performance tracking +- NumGuess D: Good performance tracking, but not mobile-friendly +- GuessIt E: Mobile-friendly, but too many ads +- Perfect Guess F: Offers hints, but the hints are not very helpful +- SmartGuesser G: Has a learning mode, but lacks a competitive edge + +## Competitive Quadrant Chart + +quadrantChart + title "User Engagement and Game Complexity" + x-axis "Low Complexity" --> "High Complexity" + y-axis "Low Engagement" --> "High Engagement" + quadrant-1 "Too Simple" + quadrant-2 "Niche Appeal" + quadrant-3 "Complex & Unengaging" + quadrant-4 "Sweet Spot" + "Guess The Number Game A": [0.2, 0.4] + "Number Master B": [0.5, 0.3] + "Quick Guess C": [0.6, 0.7] + "NumGuess D": [0.4, 0.6] + "GuessIt E": [0.7, 0.5] + "Perfect Guess F": [0.6, 0.4] + "SmartGuesser G": [0.8, 0.6] + "Our Target Product": [0.5, 0.8] + +## Requirement Analysis + +The game should be simple yet engaging, allowing players of different skill levels to enjoy it. It should provide immediate feedback and track the player's performance. The game should also be designed with a clean and intuitive interface, and it should be easy to add new features in the future. + +## Requirement Pool + +- ['P0', 'Implement the core game logic to randomly select a number and allow the user to guess it'] +- ['P0', 'Design a user interface that displays the game status and results clearly'] +- ['P1', 'Add difficulty levels by varying the range of possible numbers'] +- ['P1', 'Keep track of and display the number of attempts for each game session'] +- ['P2', "Store and show the history of the player's guesses during a game session"] + +## UI Design draft + +The UI will feature a clean and minimalist design with a number input field, submit button, and messages area to provide feedback. There will be options to select the difficulty level and a display showing the number of attempts and history of past guesses. + +## Anything UNCLEAR""" + +DESIGN_SAMPLE = """ +## Implementation approach + +We will create a Python-based number guessing game with a simple command-line interface. For the user interface, we will use the built-in 'input' and 'print' functions for interaction. The random library will be used for generating random numbers. We will structure the code to be modular and easily extendable, separating the game logic from the user interface. + +## File list + +- main.py +- game.py +- ui.py + +## Data structures and interfaces + + +classDiagram + class Game { + -int secret_number + -int min_range + -int max_range + -list attempts + +__init__(difficulty: str) + +start_game() + +check_guess(guess: int) str + +get_attempts() int + +get_history() list + } + class UI { + +start() + +display_message(message: str) + +get_user_input(prompt: str) str + +show_attempts(attempts: int) + +show_history(history: list) + +select_difficulty() str + } + class Main { + +main() + } + Main --> UI + UI --> Game + + +## Program call flow + + +sequenceDiagram + participant M as Main + participant UI as UI + participant G as Game + M->>UI: start() + UI->>UI: select_difficulty() + UI-->>G: __init__(difficulty) + G->>G: start_game() + loop Game Loop + UI->>UI: get_user_input("Enter your guess:") + UI-->>G: check_guess(guess) + G->>UI: display_message(feedback) + G->>UI: show_attempts(attempts) + G->>UI: show_history(history) + end + G->>UI: display_message("Correct! Game over.") + UI->>M: main() # Game session ends + + +## Anything UNCLEAR + +The requirement analysis suggests the need for a clean and intuitive interface. Since we are using a command-line interface, we need to ensure that the text-based UI is as user-friendly as possible. Further clarification on whether a graphical user interface (GUI) is expected in the future would be helpful for planning the extendability of the game.""" + +TASKS_SAMPLE = """ +## Required Python packages + +- random==2.2.1 + +## Required Other language third-party packages + +- No third-party dependencies required + +## Logic Analysis + +- ['game.py', 'Contains Game class with methods __init__, start_game, check_guess, get_attempts, get_history and uses random library for generating secret_number'] +- ['ui.py', 'Contains UI class with methods start, display_message, get_user_input, show_attempts, show_history, select_difficulty and interacts with Game class'] +- ['main.py', 'Contains Main class with method main that initializes UI class and starts the game loop'] + +## Task list + +- game.py +- ui.py +- main.py + +## Full API spec + + + +## Shared Knowledge + +`game.py` contains the core game logic and is used by `ui.py` to interact with the user. `main.py` serves as the entry point to start the game. + +## Anything UNCLEAR + +The requirement analysis suggests the need for a clean and intuitive interface. Since we are using a command-line interface, we need to ensure that the text-based UI is as user-friendly as possible. Further clarification on whether a graphical user interface (GUI) is expected in the future would be helpful for planning the extendability of the game.""" + +OLD_CODE_SAMPLE = """ +--- game.py +```## game.py + +import random + +class Game: + def __init__(self, difficulty: str = 'medium'): + self.min_range, self.max_range = self._set_difficulty(difficulty) + self.secret_number = random.randint(self.min_range, self.max_range) + self.attempts = [] + + def _set_difficulty(self, difficulty: str): + difficulties = { + 'easy': (1, 10), + 'medium': (1, 100), + 'hard': (1, 1000) + } + return difficulties.get(difficulty, (1, 100)) + + def start_game(self): + self.secret_number = random.randint(self.min_range, self.max_range) + self.attempts = [] + + def check_guess(self, guess: int) -> str: + self.attempts.append(guess) + if guess < self.secret_number: + return "It's higher." + elif guess > self.secret_number: + return "It's lower." + else: + return "Correct! Game over." + + def get_attempts(self) -> int: + return len(self.attempts) + + def get_history(self) -> list: + return self.attempts``` + +--- ui.py +```## ui.py + +from game import Game + +class UI: + def start(self): + difficulty = self.select_difficulty() + game = Game(difficulty) + game.start_game() + self.display_welcome_message(game) + + feedback = "" + while feedback != "Correct! Game over.": + guess = self.get_user_input("Enter your guess: ") + if self.is_valid_guess(guess): + feedback = game.check_guess(int(guess)) + self.display_message(feedback) + self.show_attempts(game.get_attempts()) + self.show_history(game.get_history()) + else: + self.display_message("Please enter a valid number.") + + def display_welcome_message(self, game): + print("Welcome to the Number Guessing Game!") + print(f"Guess the number between {game.min_range} and {game.max_range}.") + + def is_valid_guess(self, guess): + return guess.isdigit() + + def display_message(self, message: str): + print(message) + + def get_user_input(self, prompt: str) -> str: + return input(prompt) + + def show_attempts(self, attempts: int): + print(f"Number of attempts: {attempts}") + + def show_history(self, history: list): + print("Guess history:") + for guess in history: + print(guess) + + def select_difficulty(self) -> str: + while True: + difficulty = input("Select difficulty (easy, medium, hard): ").lower() + if difficulty in ['easy', 'medium', 'hard']: + return difficulty + else: + self.display_message("Invalid difficulty. Please choose 'easy', 'medium', or 'hard'.")``` + +--- main.py +```## main.py + +from ui import UI + +class Main: + def main(self): + user_interface = UI() + user_interface.start() + +if __name__ == "__main__": + main_instance = Main() + main_instance.main()``` +""" + +REFINED_PRD_JSON = { + "Language": "en_us", + "Programming Language": "Python", + "Refined Requirements": "Adding graphical interface functionality to enhance the user experience in the number-guessing game.", + "Project Name": "number_guessing_game", + "Refined Product Goals": [ + "Ensure a user-friendly interface for the game with the new graphical interface", + "Provide a challenging yet enjoyable game experience with visual enhancements", + "Design the game to be easily extendable for future features, including graphical elements", + ], + "Refined User Stories": [ + "As a player, I want to interact with a graphical interface to guess numbers and receive visual feedback on my guesses", + "As a player, I want to easily select the difficulty level through the graphical interface", + "As a player, I want to visually track my previous guesses and the number of attempts in the graphical interface", + "As a player, I want to be congratulated with a visually appealing message when I guess the number correctly", + ], + "Competitive Analysis": [ + "Guess The Number Game A: Basic text interface, no difficulty levels", + "Number Master B: Has difficulty levels, but cluttered interface", + "Quick Guess C: Sleek design, but lacks performance tracking", + "NumGuess D: Good performance tracking, but not mobile-friendly", + "GuessIt E: Mobile-friendly, but too many ads", + "Perfect Guess F: Offers hints, but the hints are not very helpful", + "SmartGuesser G: Has a learning mode, but lacks a competitive edge", + "Graphical Guess H: Graphical interface, but poor user experience due to complex design", + ], + "Competitive Quadrant Chart": 'quadrantChart\n title "User Engagement and Game Complexity with Graphical Interface"\n x-axis "Low Complexity" --> "High Complexity"\n y-axis "Low Engagement" --> "High Engagement"\n quadrant-1 "Too Simple"\n quadrant-2 "Niche Appeal"\n quadrant-3 "Complex & Unengaging"\n quadrant-4 "Sweet Spot"\n "Guess The Number Game A": [0.2, 0.4]\n "Number Master B": [0.5, 0.3]\n "Quick Guess C": [0.6, 0.7]\n "NumGuess D": [0.4, 0.6]\n "GuessIt E": [0.7, 0.5]\n "Perfect Guess F": [0.6, 0.4]\n "SmartGuesser G": [0.8, 0.6]\n "Graphical Guess H": [0.7, 0.3]\n "Our Target Product": [0.5, 0.9]', + "Refined Requirement Analysis": [ + "The game should maintain its simplicity while integrating a graphical interface for enhanced engagement.", + "Immediate visual feedback is crucial for user satisfaction in the graphical interface.", + "The interface must be intuitive, allowing for easy navigation and selection of game options.", + "The graphical design should be clean and not detract from the game's core guessing mechanic.", + ], + "Refined Requirement Pool": [ + ["P0", "Implement a graphical user interface (GUI) to replace the command-line interaction"], + [ + "P0", + "Design a user interface that displays the game status, results, and feedback clearly with graphical elements", + ], + ["P1", "Incorporate interactive elements for selecting difficulty levels"], + ["P1", "Visualize the history of the player's guesses and the number of attempts within the game session"], + ["P2", "Create animations for correct or incorrect guesses to enhance user feedback"], + ["P2", "Ensure the GUI is responsive and compatible with various screen sizes"], + ["P2", "Store and show the history of the player's guesses during a game session"], + ], + "UI Design draft": "The UI will feature a modern and minimalist design with a graphical number input field, a submit button with animations, and a dedicated area for visual feedback. It will include interactive elements to select the difficulty level and a visual display for the number of attempts and history of past guesses.", + "Anything UNCLEAR": "", +} + +REFINED_DESIGN_JSON = { + "Refined Implementation Approach": "To accommodate the new graphical user interface (GUI) requirements, we will leverage the Tkinter library, which is included with Python and supports the creation of a user-friendly GUI. The game logic will remain in Python, with Tkinter handling the rendering of the interface. We will ensure that the GUI is responsive and provides immediate visual feedback. The main game loop will be event-driven, responding to user inputs such as button clicks and difficulty selection.", + "Refined File list": ["main.py", "game.py", "ui.py", "gui.py"], + "Refined Data structures and interfaces": "\nclassDiagram\n class Game {\n -int secret_number\n -int min_range\n -int max_range\n -list attempts\n +__init__(difficulty: str)\n +start_game()\n +check_guess(guess: int) str\n +get_attempts() int\n +get_history() list\n }\n class UI {\n +start()\n +display_message(message: str)\n +get_user_input(prompt: str) str\n +show_attempts(attempts: int)\n +show_history(history: list)\n +select_difficulty() str\n }\n class GUI {\n +__init__()\n +setup_window()\n +bind_events()\n +update_feedback(message: str)\n +update_attempts(attempts: int)\n +update_history(history: list)\n +show_difficulty_selector()\n +animate_guess_result(correct: bool)\n }\n class Main {\n +main()\n }\n Main --> UI\n UI --> Game\n UI --> GUI\n GUI --> Game\n", + "Refined Program call flow": '\nsequenceDiagram\n participant M as Main\n participant UI as UI\n participant G as Game\n participant GU as GUI\n M->>UI: start()\n UI->>GU: setup_window()\n GU->>GU: bind_events()\n GU->>UI: select_difficulty()\n UI-->>G: __init__(difficulty)\n G->>G: start_game()\n loop Game Loop\n GU->>GU: show_difficulty_selector()\n GU->>UI: get_user_input("Enter your guess:")\n UI-->>G: check_guess(guess)\n G->>GU: update_feedback(feedback)\n G->>GU: update_attempts(attempts)\n G->>GU: update_history(history)\n GU->>GU: animate_guess_result(correct)\n end\n G->>GU: update_feedback("Correct! Game over.")\n GU->>M: main() # Game session ends\n', + "Anything UNCLEAR": "", +} + +REFINED_TASKS_JSON = { + "Required Python packages": ["random==2.2.1", "Tkinter==8.6"], + "Required Other language third-party packages": ["No third-party dependencies required"], + "Refined Logic Analysis": [ + [ + "game.py", + "Contains Game class with methods __init__, start_game, check_guess, get_attempts, get_history and uses random library for generating secret_number", + ], + [ + "ui.py", + "Contains UI class with methods start, display_message, get_user_input, show_attempts, show_history, select_difficulty and interacts with Game class", + ], + [ + "gui.py", + "Contains GUI class with methods __init__, setup_window, bind_events, update_feedback, update_attempts, update_history, show_difficulty_selector, animate_guess_result and interacts with Game class for GUI rendering", + ], + [ + "main.py", + "Contains Main class with method main that initializes UI class and starts the event-driven game loop", + ], + ], + "Refined Task list": ["game.py", "ui.py", "gui.py", "main.py"], + "Full API spec": "", + "Refined Shared Knowledge": "`game.py` contains the core game logic and is used by `ui.py` to interact with the user. `main.py` serves as the entry point to start the game. `gui.py` is introduced to handle the graphical user interface using Tkinter, which will interact with both `game.py` and `ui.py` for a responsive and user-friendly experience.", + "Anything UNCLEAR": "", +} + +CODE_PLAN_AND_CHANGE_SAMPLE = { + "Code Plan And Change": '\n1. Plan for gui.py: Develop the GUI using Tkinter to replace the command-line interface. Start by setting up the main window and event handling. Then, add widgets for displaying the game status, results, and feedback. Implement interactive elements for difficulty selection and visualize the guess history. Finally, create animations for guess feedback and ensure responsiveness across different screen sizes.\n```python\nclass GUI:\n- pass\n+ def __init__(self):\n+ self.setup_window()\n+\n+ def setup_window(self):\n+ # Initialize the main window using Tkinter\n+ pass\n+\n+ def bind_events(self):\n+ # Bind button clicks and other events\n+ pass\n+\n+ def update_feedback(self, message: str):\n+ # Update the feedback label with the given message\n+ pass\n+\n+ def update_attempts(self, attempts: int):\n+ # Update the attempts label with the number of attempts\n+ pass\n+\n+ def update_history(self, history: list):\n+ # Update the history view with the list of past guesses\n+ pass\n+\n+ def show_difficulty_selector(self):\n+ # Show buttons or a dropdown for difficulty selection\n+ pass\n+\n+ def animate_guess_result(self, correct: bool):\n+ # Trigger an animation for correct or incorrect guesses\n+ pass\n```\n\n2. Plan for main.py: Modify the main.py to initialize the GUI and start the event-driven game loop. Ensure that the GUI is the primary interface for user interaction.\n```python\nclass Main:\n def main(self):\n- user_interface = UI()\n- user_interface.start()\n+ graphical_user_interface = GUI()\n+ graphical_user_interface.setup_window()\n+ graphical_user_interface.bind_events()\n+ # Start the Tkinter main loop\n+ pass\n\n if __name__ == "__main__":\n main_instance = Main()\n main_instance.main()\n```\n\n3. Plan for ui.py: Refactor ui.py to work with the new GUI class. Remove command-line interactions and delegate display and input tasks to the GUI.\n```python\nclass UI:\n- def display_message(self, message: str):\n- print(message)\n+\n+ def display_message(self, message: str):\n+ # This method will now pass the message to the GUI to display\n+ pass\n\n- def get_user_input(self, prompt: str) -> str:\n- return input(prompt)\n+\n+ def get_user_input(self, prompt: str) -> str:\n+ # This method will now trigger the GUI to get user input\n+ pass\n\n- def show_attempts(self, attempts: int):\n- print(f"Number of attempts: {attempts}")\n+\n+ def show_attempts(self, attempts: int):\n+ # This method will now update the GUI with the number of attempts\n+ pass\n\n- def show_history(self, history: list):\n- print("Guess history:")\n- for guess in history:\n- print(guess)\n+\n+ def show_history(self, history: list):\n+ # This method will now update the GUI with the guess history\n+ pass\n```\n\n4. Plan for game.py: Ensure game.py remains mostly unchanged as it contains the core game logic. However, make minor adjustments if necessary to integrate with the new GUI.\n```python\nclass Game:\n # No changes required for now\n```\n' +} + +REFINED_CODE_INPUT_SAMPLE = """ +-----Now, game.py to be rewritten +```## game.py + +import random + +class Game: + def __init__(self, difficulty: str = 'medium'): + self.min_range, self.max_range = self._set_difficulty(difficulty) + self.secret_number = random.randint(self.min_range, self.max_range) + self.attempts = [] + + def _set_difficulty(self, difficulty: str): + difficulties = { + 'easy': (1, 10), + 'medium': (1, 100), + 'hard': (1, 1000) + } + return difficulties.get(difficulty, (1, 100)) + + def start_game(self): + self.secret_number = random.randint(self.min_range, self.max_range) + self.attempts = [] + + def check_guess(self, guess: int) -> str: + self.attempts.append(guess) + if guess < self.secret_number: + return "It's higher." + elif guess > self.secret_number: + return "It's lower." + else: + return "Correct! Game over." + + def get_attempts(self) -> int: + return len(self.attempts) + + def get_history(self) -> list: + return self.attempts``` +""" + +REFINED_CODE_SAMPLE = """ +## game.py + +import random + +class Game: + def __init__(self, difficulty: str = 'medium'): + # Set the difficulty level with default value 'medium' + self.min_range, self.max_range = self._set_difficulty(difficulty) + # Initialize the secret number based on the difficulty + self.secret_number = random.randint(self.min_range, self.max_range) + # Initialize the list to keep track of attempts + self.attempts = [] + + def _set_difficulty(self, difficulty: str): + # Define the range of numbers for each difficulty level + difficulties = { + 'easy': (1, 10), + 'medium': (1, 100), + 'hard': (1, 1000) + } + # Return the corresponding range for the selected difficulty, default to 'medium' if not found + return difficulties.get(difficulty, (1, 100)) + + def start_game(self): + # Reset the secret number and attempts list for a new game + self.secret_number = random.randint(self.min_range, self.max_range) + self.attempts.clear() + + def check_guess(self, guess: int) -> str: + # Add the guess to the attempts list + self.attempts.append(guess) + # Provide feedback based on the guess + if guess < self.secret_number: + return "It's higher." + elif guess > self.secret_number: + return "It's lower." + else: + return "Correct! Game over." + + def get_attempts(self) -> int: + # Return the number of attempts made + return len(self.attempts) + + def get_history(self) -> list: + # Return the list of attempts made + return self.attempts +""" diff --git a/tests/data/incremental_dev_project/number_guessing_game.zip b/tests/data/incremental_dev_project/number_guessing_game.zip new file mode 100644 index 000000000..7bbe07713 Binary files /dev/null and b/tests/data/incremental_dev_project/number_guessing_game.zip differ diff --git a/tests/data/incremental_dev_project/pygame_2048.zip b/tests/data/incremental_dev_project/pygame_2048.zip new file mode 100644 index 000000000..93e9cf0fe Binary files /dev/null and b/tests/data/incremental_dev_project/pygame_2048.zip differ diff --git a/tests/data/incremental_dev_project/readme.md b/tests/data/incremental_dev_project/readme.md new file mode 100644 index 000000000..231589028 --- /dev/null +++ b/tests/data/incremental_dev_project/readme.md @@ -0,0 +1,3 @@ +# Code archive + +This folder contains a compressed package for the test_incremental_dev.py file, which is used to demonstrate the process of incremental development. diff --git a/tests/data/incremental_dev_project/simple_add_calculator.zip b/tests/data/incremental_dev_project/simple_add_calculator.zip new file mode 100644 index 000000000..e6e73f8f9 Binary files /dev/null and b/tests/data/incremental_dev_project/simple_add_calculator.zip differ diff --git a/tests/data/incremental_dev_project/snake_game.zip b/tests/data/incremental_dev_project/snake_game.zip new file mode 100644 index 000000000..997203a18 Binary files /dev/null and b/tests/data/incremental_dev_project/snake_game.zip differ diff --git a/tests/data/incremental_dev_project/word_cloud.zip b/tests/data/incremental_dev_project/word_cloud.zip new file mode 100644 index 000000000..d8747d14d Binary files /dev/null and b/tests/data/incremental_dev_project/word_cloud.zip differ diff --git a/tests/data/ml_datasets/titanic/split_eval.csv b/tests/data/ml_datasets/titanic/split_eval.csv new file mode 100644 index 000000000..6da6ff6b3 --- /dev/null +++ b/tests/data/ml_datasets/titanic/split_eval.csv @@ -0,0 +1,180 @@ +PassengerId,Survived,Pclass,Name,Sex,Age,SibSp,Parch,Ticket,Fare,Cabin,Embarked +206,0,3,"Strom, Miss. Telma Matilda",female,2.0,0,1,347054,10.4625,G6,S +45,1,3,"Devaney, Miss. Margaret Delia",female,19.0,0,0,330958,7.8792,,Q +822,1,3,"Lulic, Mr. Nikola",male,27.0,0,0,315098,8.6625,,S +459,1,2,"Toomey, Miss. Ellen",female,50.0,0,0,F.C.C. 13531,10.5,,S +796,0,2,"Otter, Mr. Richard",male,39.0,0,0,28213,13.0,,S +119,0,1,"Baxter, Mr. Quigg Edmond",male,24.0,0,1,PC 17558,247.5208,B58 B60,C +425,0,3,"Rosblom, Mr. Viktor Richard",male,18.0,1,1,370129,20.2125,,S +679,0,3,"Goodwin, Mrs. Frederick (Augusta Tyler)",female,43.0,1,6,CA 2144,46.9,,S +270,1,1,"Bissette, Miss. Amelia",female,35.0,0,0,PC 17760,135.6333,C99,S +230,0,3,"Lefebre, Miss. Mathilde",female,,3,1,4133,25.4667,,S +690,1,1,"Madill, Miss. Georgette Alexandra",female,15.0,0,1,24160,211.3375,B5,S +321,0,3,"Dennis, Mr. Samuel",male,22.0,0,0,A/5 21172,7.25,,S +406,0,2,"Gale, Mr. Shadrach",male,34.0,1,0,28664,21.0,,S +41,0,3,"Ahlin, Mrs. Johan (Johanna Persdotter Larsson)",female,40.0,1,0,7546,9.475,,S +25,0,3,"Palsson, Miss. Torborg Danira",female,8.0,3,1,349909,21.075,,S +554,1,3,"Leeni, Mr. Fahim (""Philip Zenni"")",male,22.0,0,0,2620,7.225,,C +413,1,1,"Minahan, Miss. Daisy E",female,33.0,1,0,19928,90.0,C78,Q +513,1,1,"McGough, Mr. James Robert",male,36.0,0,0,PC 17473,26.2875,E25,S +756,1,2,"Hamalainen, Master. Viljo",male,0.67,1,1,250649,14.5,,S +392,1,3,"Jansson, Mr. Carl Olof",male,21.0,0,0,350034,7.7958,,S +602,0,3,"Slabenoff, Mr. Petco",male,,0,0,349214,7.8958,,S +326,1,1,"Young, Miss. Marie Grice",female,36.0,0,0,PC 17760,135.6333,C32,C +373,0,3,"Beavan, Mr. William Thomas",male,19.0,0,0,323951,8.05,,S +377,1,3,"Landergren, Miss. Aurora Adelia",female,22.0,0,0,C 7077,7.25,,S +201,0,3,"Vande Walle, Mr. Nestor Cyriel",male,28.0,0,0,345770,9.5,,S +512,0,3,"Webber, Mr. James",male,,0,0,SOTON/OQ 3101316,8.05,,S +601,1,2,"Jacobsohn, Mrs. Sidney Samuel (Amy Frances Christy)",female,24.0,2,1,243847,27.0,,S +631,1,1,"Barkworth, Mr. Algernon Henry Wilson",male,80.0,0,0,27042,30.0,A23,S +364,0,3,"Asim, Mr. Adola",male,35.0,0,0,SOTON/O.Q. 3101310,7.05,,S +144,0,3,"Burke, Mr. Jeremiah",male,19.0,0,0,365222,6.75,,Q +202,0,3,"Sage, Mr. Frederick",male,,8,2,CA. 2343,69.55,,S +134,1,2,"Weisz, Mrs. Leopold (Mathilde Francoise Pede)",female,29.0,1,0,228414,26.0,,S +431,1,1,"Bjornstrom-Steffansson, Mr. Mauritz Hakan",male,28.0,0,0,110564,26.55,C52,S +419,0,2,"Matthews, Mr. William John",male,30.0,0,0,28228,13.0,,S +782,1,1,"Dick, Mrs. Albert Adrian (Vera Gillespie)",female,17.0,1,0,17474,57.0,B20,S +705,0,3,"Hansen, Mr. Henrik Juul",male,26.0,1,0,350025,7.8542,,S +536,1,2,"Hart, Miss. Eva Miriam",female,7.0,0,2,F.C.C. 13529,26.25,,S +335,1,1,"Frauenthal, Mrs. Henry William (Clara Heinsheimer)",female,,1,0,PC 17611,133.65,,S +273,1,2,"Mellinger, Mrs. (Elizabeth Anne Maidment)",female,41.0,0,1,250644,19.5,,S +108,1,3,"Moss, Mr. Albert Johan",male,,0,0,312991,7.775,,S +403,0,3,"Jussila, Miss. Mari Aina",female,21.0,1,0,4137,9.825,,S +307,1,1,"Fleming, Miss. Margaret",female,,0,0,17421,110.8833,,C +218,0,2,"Jacobsohn, Mr. Sidney Samuel",male,42.0,1,0,243847,27.0,,S +789,1,3,"Dean, Master. Bertram Vere",male,1.0,1,2,C.A. 2315,20.575,,S +160,0,3,"Sage, Master. Thomas Henry",male,,8,2,CA. 2343,69.55,,S +20,1,3,"Masselmani, Mrs. Fatima",female,,0,0,2649,7.225,,C +174,0,3,"Sivola, Mr. Antti Wilhelm",male,21.0,0,0,STON/O 2. 3101280,7.925,,S +311,1,1,"Hays, Miss. Margaret Bechstein",female,24.0,0,0,11767,83.1583,C54,C +595,0,2,"Chapman, Mr. John Henry",male,37.0,1,0,SC/AH 29037,26.0,,S +592,1,1,"Stephenson, Mrs. Walter Bertram (Martha Eustis)",female,52.0,1,0,36947,78.2667,D20,C +164,0,3,"Calic, Mr. Jovo",male,17.0,0,0,315093,8.6625,,S +563,0,2,"Norman, Mr. Robert Douglas",male,28.0,0,0,218629,13.5,,S +172,0,3,"Rice, Master. Arthur",male,4.0,4,1,382652,29.125,,Q +871,0,3,"Balkic, Mr. Cerin",male,26.0,0,0,349248,7.8958,,S +176,0,3,"Klasen, Mr. Klas Albin",male,18.0,1,1,350404,7.8542,,S +434,0,3,"Kallio, Mr. Nikolai Erland",male,17.0,0,0,STON/O 2. 3101274,7.125,,S +462,0,3,"Morley, Mr. William",male,34.0,0,0,364506,8.05,,S +49,0,3,"Samaan, Mr. Youssef",male,,2,0,2662,21.6792,,C +126,1,3,"Nicola-Yarred, Master. Elias",male,12.0,1,0,2651,11.2417,,C +125,0,1,"White, Mr. Percival Wayland",male,54.0,0,1,35281,77.2875,D26,S +266,0,2,"Reeves, Mr. David",male,36.0,0,0,C.A. 17248,10.5,,S +550,1,2,"Davies, Master. John Morgan Jr",male,8.0,1,1,C.A. 33112,36.75,,S +589,0,3,"Gilinski, Mr. Eliezer",male,22.0,0,0,14973,8.05,,S +779,0,3,"Kilgannon, Mr. Thomas J",male,,0,0,36865,7.7375,,Q +179,0,2,"Hale, Mr. Reginald",male,30.0,0,0,250653,13.0,,S +107,1,3,"Salkjelsvik, Miss. Anna Kristine",female,21.0,0,0,343120,7.65,,S +624,0,3,"Hansen, Mr. Henry Damsgaard",male,21.0,0,0,350029,7.8542,,S +115,0,3,"Attalah, Miss. Malake",female,17.0,0,0,2627,14.4583,,C +42,0,2,"Turpin, Mrs. William John Robert (Dorothy Ann Wonnacott)",female,27.0,1,0,11668,21.0,,S +664,0,3,"Coleff, Mr. Peju",male,36.0,0,0,349210,7.4958,,S +661,1,1,"Frauenthal, Dr. Henry William",male,50.0,2,0,PC 17611,133.65,,S +762,0,3,"Nirva, Mr. Iisakki Antino Aijo",male,41.0,0,0,SOTON/O2 3101272,7.125,,S +580,1,3,"Jussila, Mr. Eiriik",male,32.0,0,0,STON/O 2. 3101286,7.925,,S +265,0,3,"Henry, Miss. Delia",female,,0,0,382649,7.75,,Q +757,0,3,"Carlsson, Mr. August Sigfrid",male,28.0,0,0,350042,7.7958,,S +666,0,2,"Hickman, Mr. Lewis",male,32.0,2,0,S.O.C. 14879,73.5,,S +634,0,1,"Parr, Mr. William Henry Marsh",male,,0,0,112052,0.0,,S +532,0,3,"Toufik, Mr. Nakli",male,,0,0,2641,7.2292,,C +640,0,3,"Thorneycroft, Mr. Percival",male,,1,0,376564,16.1,,S +599,0,3,"Boulos, Mr. Hanna",male,,0,0,2664,7.225,,C +220,0,2,"Harris, Mr. Walter",male,30.0,0,0,W/C 14208,10.5,,S +150,0,2,"Byles, Rev. Thomas Roussel Davids",male,42.0,0,0,244310,13.0,,S +269,1,1,"Graham, Mrs. William Thompson (Edith Junkins)",female,58.0,0,1,PC 17582,153.4625,C125,S +670,1,1,"Taylor, Mrs. Elmer Zebley (Juliet Cummins Wright)",female,,1,0,19996,52.0,C126,S +578,1,1,"Silvey, Mrs. William Baird (Alice Munger)",female,39.0,1,0,13507,55.9,E44,S +786,0,3,"Harmer, Mr. Abraham (David Lishin)",male,25.0,0,0,374887,7.25,,S +82,1,3,"Sheerlinck, Mr. Jan Baptist",male,29.0,0,0,345779,9.5,,S +400,1,2,"Trout, Mrs. William H (Jessie L)",female,28.0,0,0,240929,12.65,,S +135,0,2,"Sobey, Mr. Samuel James Hayden",male,25.0,0,0,C.A. 29178,13.0,,S +223,0,3,"Green, Mr. George Henry",male,51.0,0,0,21440,8.05,,S +693,1,3,"Lam, Mr. Ali",male,,0,0,1601,56.4958,,S +280,1,3,"Abbott, Mrs. Stanton (Rosa Hunt)",female,35.0,1,1,C.A. 2673,20.25,,S +102,0,3,"Petroff, Mr. Pastcho (""Pentcho"")",male,,0,0,349215,7.8958,,S +288,0,3,"Naidenoff, Mr. Penko",male,22.0,0,0,349206,7.8958,,S +711,1,1,"Mayne, Mlle. Berthe Antonine (""Mrs de Villiers"")",female,24.0,0,0,PC 17482,49.5042,C90,C +256,1,3,"Touma, Mrs. Darwis (Hanne Youssef Razi)",female,29.0,0,2,2650,15.2458,,C +23,1,3,"McGowan, Miss. Anna ""Annie""",female,15.0,0,0,330923,8.0292,,Q +582,1,1,"Thayer, Mrs. John Borland (Marian Longstreth Morris)",female,39.0,1,1,17421,110.8833,C68,C +564,0,3,"Simmons, Mr. John",male,,0,0,SOTON/OQ 392082,8.05,,S +405,0,3,"Oreskovic, Miss. Marija",female,20.0,0,0,315096,8.6625,,S +429,0,3,"Flynn, Mr. James",male,,0,0,364851,7.75,,Q +848,0,3,"Markoff, Mr. Marin",male,35.0,0,0,349213,7.8958,,C +726,0,3,"Oreskovic, Mr. Luka",male,20.0,0,0,315094,8.6625,,S +721,1,2,"Harper, Miss. Annie Jessie ""Nina""",female,6.0,0,1,248727,33.0,,S +637,0,3,"Leinonen, Mr. Antti Gustaf",male,32.0,0,0,STON/O 2. 3101292,7.925,,S +863,1,1,"Swift, Mrs. Frederick Joel (Margaret Welles Barron)",female,48.0,0,0,17466,25.9292,D17,S +615,0,3,"Brocklebank, Mr. William Alfred",male,35.0,0,0,364512,8.05,,S +199,1,3,"Madigan, Miss. Margaret ""Maggie""",female,,0,0,370370,7.75,,Q +787,1,3,"Sjoblom, Miss. Anna Sofia",female,18.0,0,0,3101265,7.4958,,S +156,0,1,"Williams, Mr. Charles Duane",male,51.0,0,1,PC 17597,61.3792,,C +190,0,3,"Turcin, Mr. Stjepan",male,36.0,0,0,349247,7.8958,,S +556,0,1,"Wright, Mr. George",male,62.0,0,0,113807,26.55,,S +890,1,1,"Behr, Mr. Karl Howell",male,26.0,0,0,111369,30.0,C148,C +827,0,3,"Lam, Mr. Len",male,,0,0,1601,56.4958,,S +534,1,3,"Peter, Mrs. Catherine (Catherine Rizk)",female,,0,2,2668,22.3583,,C +834,0,3,"Augustsson, Mr. Albert",male,23.0,0,0,347468,7.8542,,S +279,0,3,"Rice, Master. Eric",male,7.0,4,1,382652,29.125,,Q +189,0,3,"Bourke, Mr. John",male,40.0,1,1,364849,15.5,,Q +561,0,3,"Morrow, Mr. Thomas Rowan",male,,0,0,372622,7.75,,Q +375,0,3,"Palsson, Miss. Stina Viola",female,3.0,3,1,349909,21.075,,S +322,0,3,"Danoff, Mr. Yoto",male,27.0,0,0,349219,7.8958,,S +158,0,3,"Corn, Mr. Harry",male,30.0,0,0,SOTON/OQ 392090,8.05,,S +524,1,1,"Hippach, Mrs. Louis Albert (Ida Sophia Fischer)",female,44.0,0,1,111361,57.9792,B18,C +175,0,1,"Smith, Mr. James Clinch",male,56.0,0,0,17764,30.6958,A7,C +117,0,3,"Connors, Mr. Patrick",male,70.5,0,0,370369,7.75,,Q +810,1,1,"Chambers, Mrs. Norman Campbell (Bertha Griggs)",female,33.0,1,0,113806,53.1,E8,S +472,0,3,"Cacic, Mr. Luka",male,38.0,0,0,315089,8.6625,,S +228,0,3,"Lovell, Mr. John Hall (""Henry"")",male,20.5,0,0,A/5 21173,7.25,,S +330,1,1,"Hippach, Miss. Jean Gertrude",female,16.0,0,1,111361,57.9792,B18,C +147,1,3,"Andersson, Mr. August Edvard (""Wennerstrom"")",male,27.0,0,0,350043,7.7958,,S +98,1,1,"Greenfield, Mr. William Bertram",male,23.0,0,1,PC 17759,63.3583,D10 D12,C +493,0,1,"Molson, Mr. Harry Markland",male,55.0,0,0,113787,30.5,C30,S +73,0,2,"Hood, Mr. Ambrose Jr",male,21.0,0,0,S.O.C. 14879,73.5,,S +645,1,3,"Baclini, Miss. Eugenie",female,0.75,2,1,2666,19.2583,,C +303,0,3,"Johnson, Mr. William Cahoone Jr",male,19.0,0,0,LINE,0.0,,S +699,0,1,"Thayer, Mr. John Borland",male,49.0,1,1,17421,110.8833,C68,C +704,0,3,"Gallagher, Mr. Martin",male,25.0,0,0,36864,7.7417,,Q +639,0,3,"Panula, Mrs. Juha (Maria Emilia Ojala)",female,41.0,0,5,3101295,39.6875,,S +99,1,2,"Doling, Mrs. John T (Ada Julia Bone)",female,34.0,0,1,231919,23.0,,S +74,0,3,"Chronopoulos, Mr. Apostolos",male,26.0,1,0,2680,14.4542,,C +157,1,3,"Gilnagh, Miss. Katherine ""Katie""",female,16.0,0,0,35851,7.7333,,Q +475,0,3,"Strandberg, Miss. Ida Sofia",female,22.0,0,0,7553,9.8375,,S +240,0,2,"Hunt, Mr. George Henry",male,33.0,0,0,SCO/W 1585,12.275,,S +801,0,2,"Ponesell, Mr. Martin",male,34.0,0,0,250647,13.0,,S +829,1,3,"McCormack, Mr. Thomas Joseph",male,,0,0,367228,7.75,,Q +208,1,3,"Albimona, Mr. Nassef Cassem",male,26.0,0,0,2699,18.7875,,C +29,1,3,"O'Dwyer, Miss. Ellen ""Nellie""",female,,0,0,330959,7.8792,,Q +616,1,2,"Herman, Miss. Alice",female,24.0,1,2,220845,65.0,,S +309,0,2,"Abelson, Mr. Samuel",male,30.0,1,0,P/PP 3381,24.0,,C +382,1,3,"Nakid, Miss. Maria (""Mary"")",female,1.0,0,2,2653,15.7417,,C +703,0,3,"Barbara, Miss. Saiide",female,18.0,0,1,2691,14.4542,,C +623,1,3,"Nakid, Mr. Sahid",male,20.0,1,1,2653,15.7417,,C +26,1,3,"Asplund, Mrs. Carl Oscar (Selma Augusta Emilia Johansson)",female,38.0,1,5,347077,31.3875,,S +519,1,2,"Angle, Mrs. William A (Florence ""Mary"" Agnes Hughes)",female,36.0,1,0,226875,26.0,,S +638,0,2,"Collyer, Mr. Harvey",male,31.0,1,1,C.A. 31921,26.25,,S +360,1,3,"Mockler, Miss. Helen Mary ""Ellie""",female,,0,0,330980,7.8792,,Q +736,0,3,"Williams, Mr. Leslie",male,28.5,0,0,54636,16.1,,S +101,0,3,"Petranec, Miss. Matilda",female,28.0,0,0,349245,7.8958,,S +165,0,3,"Panula, Master. Eino Viljami",male,1.0,4,1,3101295,39.6875,,S +591,0,3,"Rintamaki, Mr. Matti",male,35.0,0,0,STON/O 2. 3101273,7.125,,S +11,1,3,"Sandstrom, Miss. Marguerite Rut",female,4.0,1,1,PP 9549,16.7,G6,S +217,1,3,"Honkanen, Miss. Eliina",female,27.0,0,0,STON/O2. 3101283,7.925,,S +734,0,2,"Berriman, Mr. William John",male,23.0,0,0,28425,13.0,,S +385,0,3,"Plotcharsky, Mr. Vasil",male,,0,0,349227,7.8958,,S +854,1,1,"Lines, Miss. Mary Conover",female,16.0,0,1,PC 17592,39.4,D28,S +860,0,3,"Razi, Mr. Raihed",male,,0,0,2629,7.2292,,C +359,1,3,"McGovern, Miss. Mary",female,,0,0,330931,7.8792,,Q +448,1,1,"Seward, Mr. Frederic Kimber",male,34.0,0,0,113794,26.55,,S +214,0,2,"Givard, Mr. Hans Kristensen",male,30.0,0,0,250646,13.0,,S +652,1,2,"Doling, Miss. Elsie",female,18.0,0,1,231919,23.0,,S +192,0,2,"Carbines, Mr. William",male,19.0,0,0,28424,13.0,,S +57,1,2,"Rugg, Miss. Emily",female,21.0,0,0,C.A. 31026,10.5,,S +868,0,1,"Roebling, Mr. Washington Augustus II",male,31.0,0,0,PC 17590,50.4958,A24,S +531,1,2,"Quick, Miss. Phyllis May",female,2.0,1,1,26360,26.0,,S +248,1,2,"Hamalainen, Mrs. William (Anna)",female,24.0,0,2,250649,14.5,,S +260,1,2,"Parrish, Mrs. (Lutie Davis)",female,50.0,0,1,230433,26.0,,S +354,0,3,"Arnold-Franchi, Mr. Josef",male,25.0,1,0,349237,17.8,,S +784,0,3,"Johnston, Mr. Andrew G",male,,1,2,W./C. 6607,23.45,,S +853,0,3,"Boulos, Miss. Nourelain",female,9.0,1,1,2678,15.2458,,C diff --git a/tests/data/ml_datasets/titanic/split_train.csv b/tests/data/ml_datasets/titanic/split_train.csv new file mode 100644 index 000000000..a48680208 --- /dev/null +++ b/tests/data/ml_datasets/titanic/split_train.csv @@ -0,0 +1,713 @@ +PassengerId,Survived,Pclass,Name,Sex,Age,SibSp,Parch,Ticket,Fare,Cabin,Embarked +409,0,3,"Birkeland, Mr. Hans Martin Monsen",male,21.0,0,0,312992,7.775,,S +481,0,3,"Goodwin, Master. Harold Victor",male,9.0,5,2,CA 2144,46.9,,S +511,1,3,"Daly, Mr. Eugene Patrick",male,29.0,0,0,382651,7.75,,Q +610,1,1,"Shutes, Miss. Elizabeth W",female,40.0,0,0,PC 17582,153.4625,C125,S +548,1,2,"Padro y Manent, Mr. Julian",male,,0,0,SC/PARIS 2146,13.8625,,C +710,1,3,"Moubarek, Master. Halim Gonios (""William George"")",male,,1,1,2661,15.2458,,C +153,0,3,"Meo, Mr. Alfonzo",male,55.5,0,0,A.5. 11206,8.05,,S +494,0,1,"Artagaveytia, Mr. Ramon",male,71.0,0,0,PC 17609,49.5042,,C +393,0,3,"Gustafsson, Mr. Johan Birger",male,28.0,2,0,3101277,7.925,,S +824,1,3,"Moor, Mrs. (Beila)",female,27.0,0,1,392096,12.475,E121,S +577,1,2,"Garside, Miss. Ethel",female,34.0,0,0,243880,13.0,,S +773,0,2,"Mack, Mrs. (Mary)",female,57.0,0,0,S.O./P.P. 3,10.5,E77,S +745,1,3,"Stranden, Mr. Juho",male,31.0,0,0,STON/O 2. 3101288,7.925,,S +328,1,2,"Ball, Mrs. (Ada E Hall)",female,36.0,0,0,28551,13.0,D,S +460,0,3,"O'Connor, Mr. Maurice",male,,0,0,371060,7.75,,Q +222,0,2,"Bracken, Mr. James H",male,27.0,0,0,220367,13.0,,S +851,0,3,"Andersson, Master. Sigvard Harald Elias",male,4.0,4,2,347082,31.275,,S +558,0,1,"Robbins, Mr. Victor",male,,0,0,PC 17757,227.525,,C +47,0,3,"Lennon, Mr. Denis",male,,1,0,370371,15.5,,Q +449,1,3,"Baclini, Miss. Marie Catherine",female,5.0,2,1,2666,19.2583,,C +371,1,1,"Harder, Mr. George Achilles",male,25.0,1,0,11765,55.4417,E50,C +196,1,1,"Lurette, Miss. Elise",female,58.0,0,0,PC 17569,146.5208,B80,C +761,0,3,"Garfirth, Mr. John",male,,0,0,358585,14.5,,S +55,0,1,"Ostby, Mr. Engelhart Cornelius",male,65.0,0,1,113509,61.9792,B30,C +573,1,1,"Flynn, Mr. John Irwin (""Irving"")",male,36.0,0,0,PC 17474,26.3875,E25,S +379,0,3,"Betros, Mr. Tannous",male,20.0,0,0,2648,4.0125,,C +198,0,3,"Olsen, Mr. Karl Siegwart Andreas",male,42.0,0,1,4579,8.4042,,S +396,0,3,"Johansson, Mr. Erik",male,22.0,0,0,350052,7.7958,,S +111,0,1,"Porter, Mr. Walter Chamberlain",male,47.0,0,0,110465,52.0,C110,S +138,0,1,"Futrelle, Mr. Jacques Heath",male,37.0,1,0,113803,53.1,C123,S +312,1,1,"Ryerson, Miss. Emily Borie",female,18.0,2,2,PC 17608,262.375,B57 B59 B63 B66,C +391,1,1,"Carter, Mr. William Ernest",male,36.0,1,2,113760,120.0,B96 B98,S +24,1,1,"Sloper, Mr. William Thompson",male,28.0,0,0,113788,35.5,A6,S +818,0,2,"Mallet, Mr. Albert",male,31.0,1,1,S.C./PARIS 2079,37.0042,,C +110,1,3,"Moran, Miss. Bertha",female,,1,0,371110,24.15,,Q +302,1,3,"McCoy, Mr. Bernard",male,,2,0,367226,23.25,,Q +104,0,3,"Johansson, Mr. Gustaf Joel",male,33.0,0,0,7540,8.6542,,S +875,1,2,"Abelson, Mrs. Samuel (Hannah Wizosky)",female,28.0,1,0,P/PP 3381,24.0,,C +62,1,1,"Icard, Miss. Amelie",female,38.0,0,0,113572,80.0,B28, +154,0,3,"van Billiard, Mr. Austin Blyler",male,40.5,0,2,A/5. 851,14.5,,S +289,1,2,"Hosono, Mr. Masabumi",male,42.0,0,0,237798,13.0,,S +245,0,3,"Attalah, Mr. Sleiman",male,30.0,0,0,2694,7.225,,C +681,0,3,"Peters, Miss. Katie",female,,0,0,330935,8.1375,,Q +797,1,1,"Leader, Dr. Alice (Farnham)",female,49.0,0,0,17465,25.9292,D17,S +226,0,3,"Berglund, Mr. Karl Ivar Sven",male,22.0,0,0,PP 4348,9.35,,S +857,1,1,"Wick, Mrs. George Dennick (Mary Hitchcock)",female,45.0,1,1,36928,164.8667,,S +621,0,3,"Yasbeck, Mr. Antoni",male,27.0,1,0,2659,14.4542,,C +451,0,2,"West, Mr. Edwy Arthur",male,36.0,1,2,C.A. 34651,27.75,,S +424,0,3,"Danbom, Mrs. Ernst Gilbert (Anna Sigrid Maria Brogren)",female,28.0,1,1,347080,14.4,,S +450,1,1,"Peuchen, Major. Arthur Godfrey",male,52.0,0,0,113786,30.5,C104,S +161,0,3,"Cribb, Mr. John Hatfield",male,44.0,0,1,371362,16.1,,S +743,1,1,"Ryerson, Miss. Susan Parker ""Suzette""",female,21.0,2,2,PC 17608,262.375,B57 B59 B63 B66,C +651,0,3,"Mitkoff, Mr. Mito",male,,0,0,349221,7.8958,,S +250,0,2,"Carter, Rev. Ernest Courtenay",male,54.0,1,0,244252,26.0,,S +540,1,1,"Frolicher, Miss. Hedwig Margaritha",female,22.0,0,2,13568,49.5,B39,C +414,0,2,"Cunningham, Mr. Alfred Fleming",male,,0,0,239853,0.0,,S +207,0,3,"Backstrom, Mr. Karl Alfred",male,32.0,1,0,3101278,15.85,,S +828,1,2,"Mallet, Master. Andre",male,1.0,0,2,S.C./PARIS 2079,37.0042,,C +484,1,3,"Turkula, Mrs. (Hedwig)",female,63.0,0,0,4134,9.5875,,S +607,0,3,"Karaic, Mr. Milan",male,30.0,0,0,349246,7.8958,,S +185,1,3,"Kink-Heilmann, Miss. Luise Gretchen",female,4.0,0,2,315153,22.025,,S +683,0,3,"Olsvigen, Mr. Thor Anderson",male,20.0,0,0,6563,9.225,,S +794,0,1,"Hoyt, Mr. William Fisher",male,,0,0,PC 17600,30.6958,,C +13,0,3,"Saundercock, Mr. William Henry",male,20.0,0,0,A/5. 2151,8.05,,S +118,0,2,"Turpin, Mr. William John Robert",male,29.0,1,0,11668,21.0,,S +483,0,3,"Rouse, Mr. Richard Henry",male,50.0,0,0,A/5 3594,8.05,,S +421,0,3,"Gheorgheff, Mr. Stanio",male,,0,0,349254,7.8958,,C +543,0,3,"Andersson, Miss. Sigrid Elisabeth",female,11.0,4,2,347082,31.275,,S +884,0,2,"Banfield, Mr. Frederick James",male,28.0,0,0,C.A./SOTON 34068,10.5,,S +877,0,3,"Gustafsson, Mr. Alfred Ossian",male,20.0,0,0,7534,9.8458,,S +109,0,3,"Rekic, Mr. Tido",male,38.0,0,0,349249,7.8958,,S +603,0,1,"Harrington, Mr. Charles H",male,,0,0,113796,42.4,,S +575,0,3,"Rush, Mr. Alfred George John",male,16.0,0,0,A/4. 20589,8.05,,S +253,0,1,"Stead, Mr. William Thomas",male,62.0,0,0,113514,26.55,C87,S +712,0,1,"Klaber, Mr. Herman",male,,0,0,113028,26.55,C124,S +397,0,3,"Olsson, Miss. Elina",female,31.0,0,0,350407,7.8542,,S +194,1,2,"Navratil, Master. Michel M",male,3.0,1,1,230080,26.0,F2,S +567,0,3,"Stoytcheff, Mr. Ilia",male,19.0,0,0,349205,7.8958,,S +204,0,3,"Youseff, Mr. Gerious",male,45.5,0,0,2628,7.225,,C +491,0,3,"Hagland, Mr. Konrad Mathias Reiersen",male,,1,0,65304,19.9667,,S +815,0,3,"Tomlin, Mr. Ernest Portage",male,30.5,0,0,364499,8.05,,S +219,1,1,"Bazzani, Miss. Albina",female,32.0,0,0,11813,76.2917,D15,C +446,1,1,"Dodge, Master. Washington",male,4.0,0,2,33638,81.8583,A34,S +490,1,3,"Coutts, Master. Eden Leslie ""Neville""",male,9.0,1,1,C.A. 37671,15.9,,S +112,0,3,"Zabour, Miss. Hileni",female,14.5,1,0,2665,14.4542,,C +731,1,1,"Allen, Miss. Elisabeth Walton",female,29.0,0,0,24160,211.3375,B5,S +106,0,3,"Mionoff, Mr. Stoytcho",male,28.0,0,0,349207,7.8958,,S +480,1,3,"Hirvonen, Miss. Hildur E",female,2.0,0,1,3101298,12.2875,,S +278,0,2,"Parkes, Mr. Francis ""Frank""",male,,0,0,239853,0.0,,S +70,0,3,"Kink, Mr. Vincenz",male,26.0,2,0,315151,8.6625,,S +86,1,3,"Backstrom, Mrs. Karl Alfred (Maria Mathilda Gustafsson)",female,33.0,3,0,3101278,15.85,,S +795,0,3,"Dantcheff, Mr. Ristiu",male,25.0,0,0,349203,7.8958,,S +162,1,2,"Watt, Mrs. James (Elizabeth ""Bessie"" Inglis Milne)",female,40.0,0,0,C.A. 33595,15.75,,S +816,0,1,"Fry, Mr. Richard",male,,0,0,112058,0.0,B102,S +517,1,2,"Lemore, Mrs. (Amelia Milley)",female,34.0,0,0,C.A. 34260,10.5,F33,S +300,1,1,"Baxter, Mrs. James (Helene DeLaudeniere Chaput)",female,50.0,0,1,PC 17558,247.5208,B58 B60,C +455,0,3,"Peduzzi, Mr. Joseph",male,,0,0,A/5 2817,8.05,,S +60,0,3,"Goodwin, Master. William Frederick",male,11.0,5,2,CA 2144,46.9,,S +880,1,1,"Potter, Mrs. Thomas Jr (Lily Alexenia Wilson)",female,56.0,0,1,11767,83.1583,C50,C +43,0,3,"Kraeff, Mr. Theodor",male,,0,0,349253,7.8958,,C +500,0,3,"Svensson, Mr. Olof",male,24.0,0,0,350035,7.7958,,S +236,0,3,"Harknett, Miss. Alice Phoebe",female,,0,0,W./C. 6609,7.55,,S +255,0,3,"Rosblom, Mrs. Viktor (Helena Wilhelmina)",female,41.0,0,2,370129,20.2125,,S +346,1,2,"Brown, Miss. Amelia ""Mildred""",female,24.0,0,0,248733,13.0,F33,S +105,0,3,"Gustafsson, Mr. Anders Vilhelm",male,37.0,2,0,3101276,7.925,,S +316,1,3,"Nilsson, Miss. Helmina Josefina",female,26.0,0,0,347470,7.8542,,S +873,0,1,"Carlsson, Mr. Frans Olof",male,33.0,0,0,695,5.0,B51 B53 B55,S +4,1,1,"Futrelle, Mrs. Jacques Heath (Lily May Peel)",female,35.0,1,0,113803,53.1,C123,S +805,1,3,"Hedman, Mr. Oskar Arvid",male,27.0,0,0,347089,6.975,,S +225,1,1,"Hoyt, Mr. Frederick Maxfield",male,38.0,1,0,19943,90.0,C93,S +772,0,3,"Jensen, Mr. Niels Peder",male,48.0,0,0,350047,7.8542,,S +539,0,3,"Risien, Mr. Samuel Beard",male,,0,0,364498,14.5,,S +249,1,1,"Beckwith, Mr. Richard Leonard",male,37.0,1,1,11751,52.5542,D35,S +32,1,1,"Spencer, Mrs. William Augustus (Marie Eugenie)",female,,1,0,PC 17569,146.5208,B78,C +268,1,3,"Persson, Mr. Ernst Ulrik",male,25.0,1,0,347083,7.775,,S +544,1,2,"Beane, Mr. Edward",male,32.0,1,0,2908,26.0,,S +685,0,2,"Brown, Mr. Thomas William Solomon",male,60.0,1,1,29750,39.0,,S +608,1,1,"Daniel, Mr. Robert Williams",male,27.0,0,0,113804,30.5,,S +749,0,1,"Marvin, Mr. Daniel Warner",male,19.0,1,0,113773,53.1,D30,S +234,1,3,"Asplund, Miss. Lillian Gertrud",female,5.0,4,2,347077,31.3875,,S +641,0,3,"Jensen, Mr. Hans Peder",male,20.0,0,0,350050,7.8542,,S +707,1,2,"Kelly, Mrs. Florence ""Fannie""",female,45.0,0,0,223596,13.5,,S +611,0,3,"Andersson, Mrs. Anders Johan (Alfrida Konstantia Brogren)",female,39.0,1,5,347082,31.275,,S +647,0,3,"Cor, Mr. Liudevit",male,19.0,0,0,349231,7.8958,,S +148,0,3,"Ford, Miss. Robina Maggie ""Ruby""",female,9.0,2,2,W./C. 6608,34.375,,S +574,1,3,"Kelly, Miss. Mary",female,,0,0,14312,7.75,,Q +809,0,2,"Meyer, Mr. August",male,39.0,0,0,248723,13.0,,S +535,0,3,"Cacic, Miss. Marija",female,30.0,0,0,315084,8.6625,,S +588,1,1,"Frolicher-Stehli, Mr. Maxmillian",male,60.0,1,1,13567,79.2,B41,C +331,1,3,"McCoy, Miss. Agnes",female,,2,0,367226,23.25,,Q +569,0,3,"Doharr, Mr. Tannous",male,,0,0,2686,7.2292,,C +725,1,1,"Chambers, Mr. Norman Campbell",male,27.0,1,0,113806,53.1,E8,S +100,0,2,"Kantor, Mr. Sinai",male,34.0,1,0,244367,26.0,,S +708,1,1,"Calderhead, Mr. Edward Pennington",male,42.0,0,0,PC 17476,26.2875,E24,S +277,0,3,"Lindblom, Miss. Augusta Charlotta",female,45.0,0,0,347073,7.75,,S +418,1,2,"Silven, Miss. Lyyli Karoliina",female,18.0,0,2,250652,13.0,,S +463,0,1,"Gee, Mr. Arthur H",male,47.0,0,0,111320,38.5,E63,S +665,1,3,"Lindqvist, Mr. Eino William",male,20.0,1,0,STON/O 2. 3101285,7.925,,S +718,1,2,"Troutt, Miss. Edwina Celia ""Winnie""",female,27.0,0,0,34218,10.5,E101,S +850,1,1,"Goldenberg, Mrs. Samuel L (Edwiga Grabowska)",female,,1,0,17453,89.1042,C92,C +516,0,1,"Walker, Mr. William Anderson",male,47.0,0,0,36967,34.0208,D46,S +633,1,1,"Stahelin-Maeglin, Dr. Max",male,32.0,0,0,13214,30.5,B50,C +538,1,1,"LeRoy, Miss. Bertha",female,30.0,0,0,PC 17761,106.425,,C +151,0,2,"Bateman, Rev. Robert James",male,51.0,0,0,S.O.P. 1166,12.525,,S +79,1,2,"Caldwell, Master. Alden Gates",male,0.83,0,2,248738,29.0,,S +10,1,2,"Nasser, Mrs. Nicholas (Adele Achem)",female,14.0,1,0,237736,30.0708,,C +143,1,3,"Hakkarainen, Mrs. Pekka Pietari (Elin Matilda Dolck)",female,24.0,1,0,STON/O2. 3101279,15.85,,S +76,0,3,"Moen, Mr. Sigurd Hansen",male,25.0,0,0,348123,7.65,F G73,S +254,0,3,"Lobb, Mr. William Arthur",male,30.0,1,0,A/5. 3336,16.1,,S +30,0,3,"Todoroff, Mr. Lalio",male,,0,0,349216,7.8958,,S +170,0,3,"Ling, Mr. Lee",male,28.0,0,0,1601,56.4958,,S +747,0,3,"Abbott, Mr. Rossmore Edward",male,16.0,1,1,C.A. 2673,20.25,,S +212,1,2,"Cameron, Miss. Clear Annie",female,35.0,0,0,F.C.C. 13528,21.0,,S +636,1,2,"Davis, Miss. Mary",female,28.0,0,0,237668,13.0,,S +689,0,3,"Fischer, Mr. Eberhard Thelander",male,18.0,0,0,350036,7.7958,,S +600,1,1,"Duff Gordon, Sir. Cosmo Edmund (""Mr Morgan"")",male,49.0,1,0,PC 17485,56.9292,A20,C +423,0,3,"Zimmerman, Mr. Leo",male,29.0,0,0,315082,7.875,,S +59,1,2,"West, Miss. Constance Mirium",female,5.0,1,2,C.A. 34651,27.75,,S +504,0,3,"Laitinen, Miss. Kristina Sofia",female,37.0,0,0,4135,9.5875,,S +352,0,1,"Williams-Lambert, Mr. Fletcher Fellows",male,,0,0,113510,35.0,C128,S +542,0,3,"Andersson, Miss. Ingeborg Constanzia",female,9.0,4,2,347082,31.275,,S +89,1,1,"Fortune, Miss. Mabel Helen",female,23.0,3,2,19950,263.0,C23 C25 C27,S +433,1,2,"Louch, Mrs. Charles Alexander (Alice Adelaide Slow)",female,42.0,1,0,SC/AH 3085,26.0,,S +566,0,3,"Davies, Mr. Alfred J",male,24.0,2,0,A/4 48871,24.15,,S +502,0,3,"Canavan, Miss. Mary",female,21.0,0,0,364846,7.75,,Q +128,1,3,"Madsen, Mr. Fridtjof Arne",male,24.0,0,0,C 17369,7.1417,,S +688,0,3,"Dakic, Mr. Branko",male,19.0,0,0,349228,10.1708,,S +329,1,3,"Goldsmith, Mrs. Frank John (Emily Alice Brown)",female,31.0,1,1,363291,20.525,,S +845,0,3,"Culumovic, Mr. Jeso",male,17.0,0,0,315090,8.6625,,S +886,0,3,"Rice, Mrs. William (Margaret Norton)",female,39.0,0,5,382652,29.125,,Q +581,1,2,"Christy, Miss. Julie Rachel",female,25.0,1,1,237789,30.0,,S +568,0,3,"Palsson, Mrs. Nils (Alma Cornelia Berglund)",female,29.0,0,4,349909,21.075,,S +152,1,1,"Pears, Mrs. Thomas (Edith Wearne)",female,22.0,1,0,113776,66.6,C2,S +342,1,1,"Fortune, Miss. Alice Elizabeth",female,24.0,3,2,19950,263.0,C23 C25 C27,S +272,1,3,"Tornquist, Mr. William Henry",male,25.0,0,0,LINE,0.0,,S +737,0,3,"Ford, Mrs. Edward (Margaret Ann Watson)",female,48.0,1,3,W./C. 6608,34.375,,S +700,0,3,"Humblen, Mr. Adolf Mathias Nicolai Olsen",male,42.0,0,0,348121,7.65,F G63,S +291,1,1,"Barber, Miss. Ellen ""Nellie""",female,26.0,0,0,19877,78.85,,S +141,0,3,"Boulos, Mrs. Joseph (Sultana)",female,,0,2,2678,15.2458,,C +261,0,3,"Smith, Mr. Thomas",male,,0,0,384461,7.75,,Q +163,0,3,"Bengtsson, Mr. John Viktor",male,26.0,0,0,347068,7.775,,S +232,0,3,"Larsson, Mr. Bengt Edvin",male,29.0,0,0,347067,7.775,,S +802,1,2,"Collyer, Mrs. Harvey (Charlotte Annie Tate)",female,31.0,1,1,C.A. 31921,26.25,,S +844,0,3,"Lemberopolous, Mr. Peter L",male,34.5,0,0,2683,6.4375,,C +691,1,1,"Dick, Mr. Albert Adrian",male,31.0,1,0,17474,57.0,B20,S +649,0,3,"Willey, Mr. Edward",male,,0,0,S.O./P.P. 751,7.55,,S +137,1,1,"Newsom, Miss. Helen Monypeny",female,19.0,0,2,11752,26.2833,D47,S +570,1,3,"Jonsson, Mr. Carl",male,32.0,0,0,350417,7.8542,,S +862,0,2,"Giles, Mr. Frederick Edward",male,21.0,1,0,28134,11.5,,S +445,1,3,"Johannesen-Bratthammer, Mr. Bernt",male,,0,0,65306,8.1125,,S +697,0,3,"Kelly, Mr. James",male,44.0,0,0,363592,8.05,,S +674,1,2,"Wilhelms, Mr. Charles",male,31.0,0,0,244270,13.0,,S +748,1,2,"Sinkkonen, Miss. Anna",female,30.0,0,0,250648,13.0,,S +367,1,1,"Warren, Mrs. Frank Manley (Anna Sophia Atkinson)",female,60.0,1,0,110813,75.25,D37,C +626,0,1,"Sutton, Mr. Frederick",male,61.0,0,0,36963,32.3208,D50,S +741,1,1,"Hawksford, Mr. Walter James",male,,0,0,16988,30.0,D45,S +821,1,1,"Hays, Mrs. Charles Melville (Clara Jennings Gregg)",female,52.0,1,1,12749,93.5,B69,S +282,0,3,"Olsson, Mr. Nils Johan Goransson",male,28.0,0,0,347464,7.8542,,S +546,0,1,"Nicholson, Mr. Arthur Ernest",male,64.0,0,0,693,26.0,,S +237,0,2,"Hold, Mr. Stephen",male,44.0,1,0,26707,26.0,,S +16,1,2,"Hewlett, Mrs. (Mary D Kingcome) ",female,55.0,0,0,248706,16.0,,S +565,0,3,"Meanwell, Miss. (Marion Ogden)",female,,0,0,SOTON/O.Q. 392087,8.05,,S +798,1,3,"Osman, Mrs. Mara",female,31.0,0,0,349244,8.6833,,S +740,0,3,"Nankoff, Mr. Minko",male,,0,0,349218,7.8958,,S +549,0,3,"Goldsmith, Mr. Frank John",male,33.0,1,1,363291,20.525,,S +663,0,1,"Colley, Mr. Edward Pomeroy",male,47.0,0,0,5727,25.5875,E58,S +482,0,2,"Frost, Mr. Anthony Wood ""Archie""",male,,0,0,239854,0.0,,S +113,0,3,"Barton, Mr. David John",male,22.0,0,0,324669,8.05,,S +458,1,1,"Kenyon, Mrs. Frederick R (Marion)",female,,1,0,17464,51.8625,D21,S +842,0,2,"Mudd, Mr. Thomas Charles",male,16.0,0,0,S.O./P.P. 3,10.5,,S +518,0,3,"Ryan, Mr. Patrick",male,,0,0,371110,24.15,,Q +553,0,3,"O'Brien, Mr. Timothy",male,,0,0,330979,7.8292,,Q +388,1,2,"Buss, Miss. Kate",female,36.0,0,0,27849,13.0,,S +514,1,1,"Rothschild, Mrs. Martin (Elizabeth L. Barrett)",female,54.0,1,0,PC 17603,59.4,,C +560,1,3,"de Messemaeker, Mrs. Guillaume Joseph (Emma)",female,36.0,1,0,345572,17.4,,S +701,1,1,"Astor, Mrs. John Jacob (Madeleine Talmadge Force)",female,18.0,1,0,PC 17757,227.525,C62 C64,C +241,0,3,"Zabour, Miss. Thamine",female,,1,0,2665,14.4542,,C +428,1,2,"Phillips, Miss. Kate Florence (""Mrs Kate Louise Phillips Marshall"")",female,19.0,0,0,250655,26.0,,S +593,0,3,"Elsbury, Mr. William James",male,47.0,0,0,A/5 3902,7.25,,S +116,0,3,"Pekoniemi, Mr. Edvard",male,21.0,0,0,STON/O 2. 3101294,7.925,,S +686,0,2,"Laroche, Mr. Joseph Philippe Lemercier",male,25.0,1,2,SC/Paris 2123,41.5792,,C +155,0,3,"Olsen, Mr. Ole Martin",male,,0,0,Fa 265302,7.3125,,S +308,1,1,"Penasco y Castellana, Mrs. Victor de Satode (Maria Josefa Perez de Soto y Vallejo)",female,17.0,1,0,PC 17758,108.9,C65,C +765,0,3,"Eklund, Mr. Hans Linus",male,16.0,0,0,347074,7.775,,S +597,1,2,"Leitch, Miss. Jessie Wills",female,,0,0,248727,33.0,,S +242,1,3,"Murphy, Miss. Katherine ""Kate""",female,,1,0,367230,15.5,,Q +823,0,1,"Reuchlin, Jonkheer. John George",male,38.0,0,0,19972,0.0,,S +380,0,3,"Gustafsson, Mr. Karl Gideon",male,19.0,0,0,347069,7.775,,S +336,0,3,"Denkoff, Mr. Mitto",male,,0,0,349225,7.8958,,S +488,0,1,"Kent, Mr. Edward Austin",male,58.0,0,0,11771,29.7,B37,C +672,0,1,"Davidson, Mr. Thornton",male,31.0,1,0,F.C. 12750,52.0,B71,S +791,0,3,"Keane, Mr. Andrew ""Andy""",male,,0,0,12460,7.75,,Q +340,0,1,"Blackwell, Mr. Stephen Weart",male,45.0,0,0,113784,35.5,T,S +879,0,3,"Laleff, Mr. Kristo",male,,0,0,349217,7.8958,,S +464,0,2,"Milling, Mr. Jacob Christian",male,48.0,0,0,234360,13.0,,S +717,1,1,"Endres, Miss. Caroline Louise",female,38.0,0,0,PC 17757,227.525,C45,C +343,0,2,"Collander, Mr. Erik Gustaf",male,28.0,0,0,248740,13.0,,S +276,1,1,"Andrews, Miss. Kornelia Theodosia",female,63.0,1,0,13502,77.9583,D7,S +530,0,2,"Hocking, Mr. Richard George",male,23.0,2,1,29104,11.5,,S +861,0,3,"Hansen, Mr. Claus Peter",male,41.0,2,0,350026,14.1083,,S +8,0,3,"Palsson, Master. Gosta Leonard",male,2.0,3,1,349909,21.075,,S +841,0,3,"Alhomaki, Mr. Ilmari Rudolf",male,20.0,0,0,SOTON/O2 3101287,7.925,,S +231,1,1,"Harris, Mrs. Henry Birkhardt (Irene Wallach)",female,35.0,1,0,36973,83.475,C83,S +338,1,1,"Burns, Miss. Elizabeth Margaret",female,41.0,0,0,16966,134.5,E40,C +286,0,3,"Stankovic, Mr. Ivan",male,33.0,0,0,349239,8.6625,,C +381,1,1,"Bidois, Miss. Rosalie",female,42.0,0,0,PC 17757,227.525,,C +468,0,1,"Smart, Mr. John Montgomery",male,56.0,0,0,113792,26.55,,S +838,0,3,"Sirota, Mr. Maurice",male,,0,0,392092,8.05,,S +742,0,1,"Cavendish, Mr. Tyrell William",male,36.0,1,0,19877,78.85,C46,S +617,0,3,"Danbom, Mr. Ernst Gilbert",male,34.0,1,1,347080,14.4,,S +485,1,1,"Bishop, Mr. Dickinson H",male,25.0,1,0,11967,91.0792,B49,C +437,0,3,"Ford, Miss. Doolina Margaret ""Daisy""",female,21.0,2,2,W./C. 6608,34.375,,S +885,0,3,"Sutehall, Mr. Henry Jr",male,25.0,0,0,SOTON/OQ 392076,7.05,,S +28,0,1,"Fortune, Mr. Charles Alexander",male,19.0,3,2,19950,263.0,C23 C25 C27,S +751,1,2,"Wells, Miss. Joan",female,4.0,1,1,29103,23.0,,S +97,0,1,"Goldschmidt, Mr. George B",male,71.0,0,0,PC 17754,34.6542,A5,C +6,0,3,"Moran, Mr. James",male,,0,0,330877,8.4583,,Q +271,0,1,"Cairns, Mr. Alexander",male,,0,0,113798,31.0,,S +301,1,3,"Kelly, Miss. Anna Katherine ""Annie Kate""",female,,0,0,9234,7.75,,Q +366,0,3,"Adahl, Mr. Mauritz Nils Martin",male,30.0,0,0,C 7076,7.25,,S +200,0,2,"Yrois, Miss. Henriette (""Mrs Harbeck"")",female,24.0,0,0,248747,13.0,,S +776,0,3,"Myhrman, Mr. Pehr Fabian Oliver Malkolm",male,18.0,0,0,347078,7.75,,S +178,0,1,"Isham, Miss. Ann Elizabeth",female,50.0,0,0,PC 17595,28.7125,C49,C +728,1,3,"Mannion, Miss. Margareth",female,,0,0,36866,7.7375,,Q +167,1,1,"Chibnall, Mrs. (Edith Martha Bowerman)",female,,0,1,113505,55.0,E33,S +869,0,3,"van Melkebeke, Mr. Philemon",male,,0,0,345777,9.5,,S +313,0,2,"Lahtinen, Mrs. William (Anna Sylfven)",female,26.0,1,1,250651,26.0,,S +285,0,1,"Smith, Mr. Richard William",male,,0,0,113056,26.0,A19,S +495,0,3,"Stanley, Mr. Edward Roland",male,21.0,0,0,A/4 45380,8.05,,S +33,1,3,"Glynn, Miss. Mary Agatha",female,,0,0,335677,7.75,,Q +417,1,2,"Drew, Mrs. James Vivian (Lulu Thorne Christian)",female,34.0,1,1,28220,32.5,,S +887,0,2,"Montvila, Rev. Juozas",male,27.0,0,0,211536,13.0,,S +559,1,1,"Taussig, Mrs. Emil (Tillie Mandelbaum)",female,39.0,1,1,110413,79.65,E67,S +806,0,3,"Johansson, Mr. Karl Johan",male,31.0,0,0,347063,7.775,,S +294,0,3,"Haas, Miss. Aloisia",female,24.0,0,0,349236,8.85,,S +209,1,3,"Carr, Miss. Helen ""Ellen""",female,16.0,0,0,367231,7.75,,Q +85,1,2,"Ilett, Miss. Bertha",female,17.0,0,0,SO/C 14885,10.5,,S +38,0,3,"Cann, Mr. Ernest Charles",male,21.0,0,0,A./5. 2152,8.05,,S +7,0,1,"McCarthy, Mr. Timothy J",male,54.0,0,0,17463,51.8625,E46,S +426,0,3,"Wiseman, Mr. Phillippe",male,,0,0,A/4. 34244,7.25,,S +790,0,1,"Guggenheim, Mr. Benjamin",male,46.0,0,0,PC 17593,79.2,B82 B84,C +389,0,3,"Sadlier, Mr. Matthew",male,,0,0,367655,7.7292,,Q +258,1,1,"Cherry, Miss. Gladys",female,30.0,0,0,110152,86.5,B77,S +643,0,3,"Skoog, Miss. Margit Elizabeth",female,2.0,3,2,347088,27.9,,S +355,0,3,"Yousif, Mr. Wazli",male,,0,0,2647,7.225,,C +830,1,1,"Stone, Mrs. George Nelson (Martha Evelyn)",female,62.0,0,0,113572,80.0,B28, +781,1,3,"Ayoub, Miss. Banoura",female,13.0,0,0,2687,7.2292,,C +267,0,3,"Panula, Mr. Ernesti Arvid",male,16.0,4,1,3101295,39.6875,,S +506,0,1,"Penasco y Castellana, Mr. Victor de Satode",male,18.0,1,0,PC 17758,108.9,C65,C +52,0,3,"Nosworthy, Mr. Richard Cater",male,21.0,0,0,A/4. 39886,7.8,,S +401,1,3,"Niskanen, Mr. Juha",male,39.0,0,0,STON/O 2. 3101289,7.925,,S +533,0,3,"Elias, Mr. Joseph Jr",male,17.0,1,1,2690,7.2292,,C +283,0,3,"de Pelsmaeker, Mr. Alfons",male,16.0,0,0,345778,9.5,,S +442,0,3,"Hampe, Mr. Leon",male,20.0,0,0,345769,9.5,,S +361,0,3,"Skoog, Mr. Wilhelm",male,40.0,1,4,347088,27.9,,S +840,1,1,"Marechal, Mr. Pierre",male,,0,0,11774,29.7,C47,C +509,0,3,"Olsen, Mr. Henry Margido",male,28.0,0,0,C 4001,22.525,,S +121,0,2,"Hickman, Mr. Stanley George",male,21.0,2,0,S.O.C. 14879,73.5,,S +320,1,1,"Spedden, Mrs. Frederic Oakley (Margaretta Corning Stone)",female,40.0,1,1,16966,134.5,E34,C +858,1,1,"Daly, Mr. Peter Denis ",male,51.0,0,0,113055,26.55,E17,S +501,0,3,"Calic, Mr. Petar",male,17.0,0,0,315086,8.6625,,S +91,0,3,"Christmann, Mr. Emil",male,29.0,0,0,343276,8.05,,S +727,1,2,"Renouf, Mrs. Peter Henry (Lillian Jefferys)",female,30.0,3,0,31027,21.0,,S +671,1,2,"Brown, Mrs. Thomas William Solomon (Elizabeth Catherine Ford)",female,40.0,1,1,29750,39.0,,S +456,1,3,"Jalsevac, Mr. Ivan",male,29.0,0,0,349240,7.8958,,C +427,1,2,"Clarke, Mrs. Charles V (Ada Maria Winfield)",female,28.0,1,0,2003,26.0,,S +63,0,1,"Harris, Mr. Henry Birkhardt",male,45.0,1,0,36973,83.475,C83,S +51,0,3,"Panula, Master. Juha Niilo",male,7.0,4,1,3101295,39.6875,,S +454,1,1,"Goldenberg, Mr. Samuel L",male,49.0,1,0,17453,89.1042,C92,C +394,1,1,"Newell, Miss. Marjorie",female,23.0,1,0,35273,113.275,D36,C +188,1,1,"Romaine, Mr. Charles Hallace (""Mr C Rolmane"")",male,45.0,0,0,111428,26.55,,S +368,1,3,"Moussa, Mrs. (Mantoura Boulos)",female,,0,0,2626,7.2292,,C +759,0,3,"Theobald, Mr. Thomas Leonard",male,34.0,0,0,363294,8.05,,S +804,1,3,"Thomas, Master. Assad Alexander",male,0.42,0,1,2625,8.5167,,C +510,1,3,"Lang, Mr. Fang",male,26.0,0,0,1601,56.4958,,S +788,0,3,"Rice, Master. George Hugh",male,8.0,4,1,382652,29.125,,Q +298,0,1,"Allison, Miss. Helen Loraine",female,2.0,1,2,113781,151.55,C22 C26,S +92,0,3,"Andreasson, Mr. Paul Edvin",male,20.0,0,0,347466,7.8542,,S +754,0,3,"Jonkoff, Mr. Lalio",male,23.0,0,0,349204,7.8958,,S +547,1,2,"Beane, Mrs. Edward (Ethel Clarke)",female,19.0,1,0,2908,26.0,,S +492,0,3,"Windelov, Mr. Einar",male,21.0,0,0,SOTON/OQ 3101317,7.25,,S +2,1,1,"Cumings, Mrs. John Bradley (Florence Briggs Thayer)",female,38.0,1,0,PC 17599,71.2833,C85,C +777,0,3,"Tobin, Mr. Roger",male,,0,0,383121,7.75,F38,Q +473,1,2,"West, Mrs. Edwy Arthur (Ada Mary Worth)",female,33.0,1,2,C.A. 34651,27.75,,S +252,0,3,"Strom, Mrs. Wilhelm (Elna Matilda Persson)",female,29.0,1,1,347054,10.4625,G6,S +93,0,1,"Chaffee, Mr. Herbert Fuller",male,46.0,1,0,W.E.P. 5734,61.175,E31,S +635,0,3,"Skoog, Miss. Mabel",female,9.0,3,2,347088,27.9,,S +44,1,2,"Laroche, Miss. Simonne Marie Anne Andree",female,3.0,1,2,SC/Paris 2123,41.5792,,C +835,0,3,"Allum, Mr. Owen George",male,18.0,0,0,2223,8.3,,S +48,1,3,"O'Driscoll, Miss. Bridget",female,,0,0,14311,7.75,,Q +891,0,3,"Dooley, Mr. Patrick",male,32.0,0,0,370376,7.75,,Q +264,0,1,"Harrison, Mr. William",male,40.0,0,0,112059,0.0,B94,S +356,0,3,"Vanden Steen, Mr. Leo Peter",male,28.0,0,0,345783,9.5,,S +528,0,1,"Farthing, Mr. John",male,,0,0,PC 17483,221.7792,C95,S +339,1,3,"Dahl, Mr. Karl Edwart",male,45.0,0,0,7598,8.05,,S +780,1,1,"Robert, Mrs. Edward Scott (Elisabeth Walton McMillan)",female,43.0,0,1,24160,211.3375,B3,S +21,0,2,"Fynney, Mr. Joseph J",male,35.0,0,0,239865,26.0,,S +723,0,2,"Gillespie, Mr. William Henry",male,34.0,0,0,12233,13.0,,S +677,0,3,"Sawyer, Mr. Frederick Charles",male,24.5,0,0,342826,8.05,,S +349,1,3,"Coutts, Master. William Loch ""William""",male,3.0,1,1,C.A. 37671,15.9,,S +817,0,3,"Heininen, Miss. Wendla Maria",female,23.0,0,0,STON/O2. 3101290,7.925,,S +334,0,3,"Vander Planke, Mr. Leo Edmondus",male,16.0,2,0,345764,18.0,,S +470,1,3,"Baclini, Miss. Helene Barbara",female,0.75,2,1,2666,19.2583,,C +130,0,3,"Ekstrom, Mr. Johan",male,45.0,0,0,347061,6.975,,S +191,1,2,"Pinsky, Mrs. (Rosa)",female,32.0,0,0,234604,13.0,,S +760,1,1,"Rothes, the Countess. of (Lucy Noel Martha Dyer-Edwards)",female,33.0,0,0,110152,86.5,B77,S +520,0,3,"Pavlovic, Mr. Stefo",male,32.0,0,0,349242,7.8958,,S +67,1,2,"Nye, Mrs. (Elizabeth Ramell)",female,29.0,0,0,C.A. 29395,10.5,F33,S +487,1,1,"Hoyt, Mrs. Frederick Maxfield (Jane Anne Forby)",female,35.0,1,0,19943,90.0,C93,S +19,0,3,"Vander Planke, Mrs. Julius (Emelia Maria Vandemoortele)",female,31.0,1,0,345763,18.0,,S +702,1,1,"Silverthorne, Mr. Spencer Victor",male,35.0,0,0,PC 17475,26.2875,E24,S +826,0,3,"Flynn, Mr. John",male,,0,0,368323,6.95,,Q +333,0,1,"Graham, Mr. George Edward",male,38.0,0,1,PC 17582,153.4625,C91,S +855,0,2,"Carter, Mrs. Ernest Courtenay (Lilian Hughes)",female,44.0,1,0,244252,26.0,,S +441,1,2,"Hart, Mrs. Benjamin (Esther Ada Bloomfield)",female,45.0,1,1,F.C.C. 13529,26.25,,S +775,1,2,"Hocking, Mrs. Elizabeth (Eliza Needs)",female,54.0,1,3,29105,23.0,,S +675,0,2,"Watson, Mr. Ennis Hastings",male,,0,0,239856,0.0,,S +552,0,2,"Sharp, Mr. Percival James R",male,27.0,0,0,244358,26.0,,S +56,1,1,"Woolner, Mr. Hugh",male,,0,0,19947,35.5,C52,S +653,0,3,"Kalvik, Mr. Johannes Halvorsen",male,21.0,0,0,8475,8.4333,,S +849,0,2,"Harper, Rev. John",male,28.0,0,1,248727,33.0,,S +730,0,3,"Ilmakangas, Miss. Pieta Sofia",female,25.0,1,0,STON/O2. 3101271,7.925,,S +233,0,2,"Sjostedt, Mr. Ernst Adolf",male,59.0,0,0,237442,13.5,,S +660,0,1,"Newell, Mr. Arthur Webster",male,58.0,0,2,35273,113.275,D48,C +243,0,2,"Coleridge, Mr. Reginald Charles",male,29.0,0,0,W./C. 14263,10.5,,S +36,0,1,"Holverson, Mr. Alexander Oskar",male,42.0,1,0,113789,52.0,,S +541,1,1,"Crosby, Miss. Harriet R",female,36.0,0,2,WE/P 5735,71.0,B22,S +719,0,3,"McEvoy, Mr. Michael",male,,0,0,36568,15.5,,Q +752,1,3,"Moor, Master. Meier",male,6.0,0,1,392096,12.475,E121,S +888,1,1,"Graham, Miss. Margaret Edith",female,19.0,0,0,112053,30.0,B42,S +122,0,3,"Moore, Mr. Leonard Charles",male,,0,0,A4. 54510,8.05,,S +411,0,3,"Sdycoff, Mr. Todor",male,,0,0,349222,7.8958,,S +353,0,3,"Elias, Mr. Tannous",male,15.0,1,1,2695,7.2292,,C +34,0,2,"Wheadon, Mr. Edward H",male,66.0,0,0,C.A. 24579,10.5,,S +180,0,3,"Leonard, Mr. Lionel",male,36.0,0,0,LINE,0.0,,S +646,1,1,"Harper, Mr. Henry Sleeper",male,48.0,1,0,PC 17572,76.7292,D33,C +819,0,3,"Holm, Mr. John Fredrik Alexander",male,43.0,0,0,C 7075,6.45,,S +22,1,2,"Beesley, Mr. Lawrence",male,34.0,0,0,248698,13.0,D56,S +412,0,3,"Hart, Mr. Henry",male,,0,0,394140,6.8583,,Q +422,0,3,"Charters, Mr. David",male,21.0,0,0,A/5. 13032,7.7333,,Q +584,0,1,"Ross, Mr. John Hugo",male,36.0,0,0,13049,40.125,A10,C +729,0,2,"Bryhl, Mr. Kurt Arnold Gottfrid",male,25.0,1,0,236853,26.0,,S +813,0,2,"Slemen, Mr. Richard James",male,35.0,0,0,28206,10.5,,S +562,0,3,"Sivic, Mr. Husein",male,40.0,0,0,349251,7.8958,,S +332,0,1,"Partner, Mr. Austen",male,45.5,0,0,113043,28.5,C124,S +341,1,2,"Navratil, Master. Edmond Roger",male,2.0,1,1,230080,26.0,F2,S +247,0,3,"Lindahl, Miss. Agda Thorilda Viktoria",female,25.0,0,0,347071,7.775,,S +127,0,3,"McMahon, Mr. Martin",male,,0,0,370372,7.75,,Q +324,1,2,"Caldwell, Mrs. Albert Francis (Sylvia Mae Harbaugh)",female,22.0,1,1,248738,29.0,,S +398,0,2,"McKane, Mr. Peter David",male,46.0,0,0,28403,26.0,,S +46,0,3,"Rogers, Mr. William John",male,,0,0,S.C./A.4. 23567,8.05,,S +65,0,1,"Stewart, Mr. Albert A",male,,0,0,PC 17605,27.7208,,C +262,1,3,"Asplund, Master. Edvin Rojj Felix",male,3.0,4,2,347077,31.3875,,S +372,0,3,"Wiklund, Mr. Jakob Alfred",male,18.0,1,0,3101267,6.4958,,S +376,1,1,"Meyer, Mrs. Edgar Joseph (Leila Saks)",female,,1,0,PC 17604,82.1708,,C +676,0,3,"Edvardsson, Mr. Gustaf Hjalmar",male,18.0,0,0,349912,7.775,,S +471,0,3,"Keefe, Mr. Arthur",male,,0,0,323592,7.25,,S +210,1,1,"Blank, Mr. Henry",male,40.0,0,0,112277,31.0,A31,C +733,0,2,"Knight, Mr. Robert J",male,,0,0,239855,0.0,,S +81,0,3,"Waelens, Mr. Achille",male,22.0,0,0,345767,9.0,,S +609,1,2,"Laroche, Mrs. Joseph (Juliette Marie Louise Lafargue)",female,22.0,1,2,SC/Paris 2123,41.5792,,C +874,0,3,"Vander Cruyssen, Mr. Victor",male,47.0,0,0,345765,9.0,,S +435,0,1,"Silvey, Mr. William Baird",male,50.0,1,0,13507,55.9,E44,S +767,0,1,"Brewe, Dr. Arthur Jackson",male,,0,0,112379,39.6,,C +768,0,3,"Mangan, Miss. Mary",female,30.5,0,0,364850,7.75,,Q +168,0,3,"Skoog, Mrs. William (Anna Bernhardina Karlsson)",female,45.0,1,4,347088,27.9,,S +709,1,1,"Cleaver, Miss. Alice",female,22.0,0,0,113781,151.55,,S +327,0,3,"Nysveen, Mr. Johan Hansen",male,61.0,0,0,345364,6.2375,,S +843,1,1,"Serepeca, Miss. Augusta",female,30.0,0,0,113798,31.0,,C +211,0,3,"Ali, Mr. Ahmed",male,24.0,0,0,SOTON/O.Q. 3101311,7.05,,S +159,0,3,"Smiljanic, Mr. Mile",male,,0,0,315037,8.6625,,S +378,0,1,"Widener, Mr. Harry Elkins",male,27.0,0,2,113503,211.5,C82,C +778,1,3,"Emanuel, Miss. Virginia Ethel",female,5.0,0,0,364516,12.475,,S +457,0,1,"Millet, Mr. Francis Davis",male,65.0,0,0,13509,26.55,E38,S +769,0,3,"Moran, Mr. Daniel J",male,,1,0,371110,24.15,,Q +362,0,2,"del Carlo, Mr. Sebastiano",male,29.0,1,0,SC/PARIS 2167,27.7208,,C +655,0,3,"Hegarty, Miss. Hanora ""Nora""",female,18.0,0,0,365226,6.75,,Q +698,1,3,"Mullens, Miss. Katherine ""Katie""",female,,0,0,35852,7.7333,,Q +444,1,2,"Reynaldo, Ms. Encarnacion",female,28.0,0,0,230434,13.0,,S +203,0,3,"Johanson, Mr. Jakob Alfred",male,34.0,0,0,3101264,6.4958,,S +606,0,3,"Lindell, Mr. Edvard Bengtsson",male,36.0,1,0,349910,15.55,,S +673,0,2,"Mitchell, Mr. Henry Michael",male,70.0,0,0,C.A. 24580,10.5,,S +846,0,3,"Abbing, Mr. Anthony",male,42.0,0,0,C.A. 5547,7.55,,S +374,0,1,"Ringhini, Mr. Sante",male,22.0,0,0,PC 17760,135.6333,,C +667,0,2,"Butler, Mr. Reginald Fenton",male,25.0,0,0,234686,13.0,,S +61,0,3,"Sirayanian, Mr. Orsen",male,22.0,0,0,2669,7.2292,,C +642,1,1,"Sagesser, Mlle. Emma",female,24.0,0,0,PC 17477,69.3,B35,C +469,0,3,"Scanlan, Mr. James",male,,0,0,36209,7.725,,Q +792,0,2,"Gaskell, Mr. Alfred",male,16.0,0,0,239865,26.0,,S +465,0,3,"Maisner, Mr. Simon",male,,0,0,A/S 2816,8.05,,S +551,1,1,"Thayer, Mr. John Borland Jr",male,17.0,0,2,17421,110.8833,C70,C +523,0,3,"Lahoud, Mr. Sarkis",male,,0,0,2624,7.225,,C +369,1,3,"Jermyn, Miss. Annie",female,,0,0,14313,7.75,,Q +864,0,3,"Sage, Miss. Dorothy Edith ""Dolly""",female,,8,2,CA. 2343,69.55,,S +839,1,3,"Chip, Mr. Chang",male,32.0,0,0,1601,56.4958,,S +590,0,3,"Murdlin, Mr. Joseph",male,,0,0,A./5. 3235,8.05,,S +9,1,3,"Johnson, Mrs. Oscar W (Elisabeth Vilhelmina Berg)",female,27.0,0,2,347742,11.1333,,S +505,1,1,"Maioni, Miss. Roberta",female,16.0,0,0,110152,86.5,B79,S +572,1,1,"Appleton, Mrs. Edward Dale (Charlotte Lamson)",female,53.0,2,0,11769,51.4792,C101,S +235,0,2,"Leyson, Mr. Robert William Norman",male,24.0,0,0,C.A. 29566,10.5,,S +345,0,2,"Fox, Mr. Stanley Hubert",male,36.0,0,0,229236,13.0,,S +714,0,3,"Larsson, Mr. August Viktor",male,29.0,0,0,7545,9.4833,,S +477,0,2,"Renouf, Mr. Peter Henry",male,34.0,1,0,31027,21.0,,S +587,0,2,"Jarvis, Mr. John Denzil",male,47.0,0,0,237565,15.0,,S +630,0,3,"O'Connell, Mr. Patrick D",male,,0,0,334912,7.7333,,Q +133,0,3,"Robins, Mrs. Alexander A (Grace Charity Laury)",female,47.0,1,0,A/5. 3337,14.5,,S +27,0,3,"Emir, Mr. Farred Chehab",male,,0,0,2631,7.225,,C +612,0,3,"Jardin, Mr. Jose Neto",male,,0,0,SOTON/O.Q. 3101305,7.05,,S +292,1,1,"Bishop, Mrs. Dickinson H (Helen Walton)",female,19.0,1,0,11967,91.0792,B49,C +293,0,2,"Levy, Mr. Rene Jacques",male,36.0,0,0,SC/Paris 2163,12.875,D,C +40,1,3,"Nicola-Yarred, Miss. Jamila",female,14.0,1,0,2651,11.2417,,C +205,1,3,"Cohen, Mr. Gurshon ""Gus""",male,18.0,0,0,A/5 3540,8.05,,S +832,1,2,"Richards, Master. George Sibley",male,0.83,1,1,29106,18.75,,S +716,0,3,"Soholt, Mr. Peter Andreas Lauritz Andersen",male,19.0,0,0,348124,7.65,F G73,S +596,0,3,"Van Impe, Mr. Jean Baptiste",male,36.0,1,1,345773,24.15,,S +344,0,2,"Sedgwick, Mr. Charles Frederick Waddington",male,25.0,0,0,244361,13.0,,S +687,0,3,"Panula, Mr. Jaako Arnold",male,14.0,4,1,3101295,39.6875,,S +662,0,3,"Badt, Mr. Mohamed",male,40.0,0,0,2623,7.225,,C +66,1,3,"Moubarek, Master. Gerios",male,,1,1,2661,15.2458,,C +820,0,3,"Skoog, Master. Karl Thorsten",male,10.0,3,2,347088,27.9,,S +865,0,2,"Gill, Mr. John William",male,24.0,0,0,233866,13.0,,S +323,1,2,"Slayter, Miss. Hilda Mary",female,30.0,0,0,234818,12.35,,Q +358,0,2,"Funk, Miss. Annie Clemmer",female,38.0,0,0,237671,13.0,,S +129,1,3,"Peter, Miss. Anna",female,,1,1,2668,22.3583,F E69,C +166,1,3,"Goldsmith, Master. Frank John William ""Frankie""",male,9.0,0,2,363291,20.525,,S +799,0,3,"Ibrahim Shawah, Mr. Yousseff",male,30.0,0,0,2685,7.2292,,C +770,0,3,"Gronnestad, Mr. Daniel Danielsen",male,32.0,0,0,8471,8.3625,,S +785,0,3,"Ali, Mr. William",male,25.0,0,0,SOTON/O.Q. 3101312,7.05,,S +399,0,2,"Pain, Dr. Alfred",male,23.0,0,0,244278,10.5,,S +746,0,1,"Crosby, Capt. Edward Gifford",male,70.0,1,1,WE/P 5735,71.0,B22,S +498,0,3,"Shellard, Mr. Frederick William",male,,0,0,C.A. 6212,15.1,,S +297,0,3,"Hanna, Mr. Mansour",male,23.5,0,0,2693,7.2292,,C +295,0,3,"Mineff, Mr. Ivan",male,24.0,0,0,349233,7.8958,,S +545,0,1,"Douglas, Mr. Walter Donald",male,50.0,1,0,PC 17761,106.425,C86,C +755,1,2,"Herman, Mrs. Samuel (Jane Laver)",female,48.0,1,2,220845,65.0,,S +305,0,3,"Williams, Mr. Howard Hugh ""Harry""",male,,0,0,A/5 2466,8.05,,S +682,1,1,"Hassab, Mr. Hammad",male,27.0,0,0,PC 17572,76.7292,D49,C +124,1,2,"Webber, Miss. Susan",female,32.5,0,0,27267,13.0,E101,S +499,0,1,"Allison, Mrs. Hudson J C (Bessie Waldo Daniels)",female,25.0,1,2,113781,151.55,C22 C26,S +870,1,3,"Johnson, Master. Harold Theodor",male,4.0,1,1,347742,11.1333,,S +72,0,3,"Goodwin, Miss. Lillian Amy",female,16.0,5,2,CA 2144,46.9,,S +120,0,3,"Andersson, Miss. Ellis Anna Maria",female,2.0,4,2,347082,31.275,,S +325,0,3,"Sage, Mr. George John Jr",male,,8,2,CA. 2343,69.55,,S +383,0,3,"Tikkanen, Mr. Juho",male,32.0,0,0,STON/O 2. 3101293,7.925,,S +628,1,1,"Longley, Miss. Gretchen Fiske",female,21.0,0,0,13502,77.9583,D9,S +744,0,3,"McNamee, Mr. Neal",male,24.0,1,0,376566,16.1,,S +684,0,3,"Goodwin, Mr. Charles Edward",male,14.0,5,2,CA 2144,46.9,,S +598,0,3,"Johnson, Mr. Alfred",male,49.0,0,0,LINE,0.0,,S +866,1,2,"Bystrom, Mrs. (Karolina)",female,42.0,0,0,236852,13.0,,S +53,1,1,"Harper, Mrs. Henry Sleeper (Myna Haxtun)",female,49.0,1,0,PC 17572,76.7292,D33,C +732,0,3,"Hassan, Mr. Houssein G N",male,11.0,0,0,2699,18.7875,,C +306,1,1,"Allison, Master. Hudson Trevor",male,0.92,1,2,113781,151.55,C22 C26,S +140,0,1,"Giglio, Mr. Victor",male,24.0,0,0,PC 17593,79.2,B86,C +814,0,3,"Andersson, Miss. Ebba Iris Alfrida",female,6.0,4,2,347082,31.275,,S +310,1,1,"Francatelli, Miss. Laura Mabel",female,30.0,0,0,PC 17485,56.9292,E36,C +71,0,2,"Jenkin, Mr. Stephen Curnow",male,32.0,0,0,C.A. 33111,10.5,,S +529,0,3,"Salonen, Mr. Johan Werner",male,39.0,0,0,3101296,7.925,,S +466,0,3,"Goncalves, Mr. Manuel Estanslas",male,38.0,0,0,SOTON/O.Q. 3101306,7.05,,S +319,1,1,"Wick, Miss. Mary Natalie",female,31.0,0,2,36928,164.8667,C7,S +259,1,1,"Ward, Miss. Anna",female,35.0,0,0,PC 17755,512.3292,,C +114,0,3,"Jussila, Miss. Katriina",female,20.0,1,0,4136,9.825,,S +625,0,3,"Bowen, Mr. David John ""Dai""",male,21.0,0,0,54636,16.1,,S +555,1,3,"Ohman, Miss. Velin",female,22.0,0,0,347085,7.775,,S +357,1,1,"Bowerman, Miss. Elsie Edith",female,22.0,0,1,113505,55.0,E33,S +837,0,3,"Pasic, Mr. Jakob",male,21.0,0,0,315097,8.6625,,S +84,0,1,"Carrau, Mr. Francisco M",male,28.0,0,0,113059,47.1,,S +184,1,2,"Becker, Master. Richard F",male,1.0,2,1,230136,39.0,F4,S +183,0,3,"Asplund, Master. Clarence Gustaf Hugo",male,9.0,4,2,347077,31.3875,,S +145,0,2,"Andrew, Mr. Edgardo Samuel",male,18.0,0,0,231945,11.5,,S +859,1,3,"Baclini, Mrs. Solomon (Latifa Qurban)",female,24.0,0,3,2666,19.2583,,C +299,1,1,"Saalfeld, Mr. Adolphe",male,,0,0,19988,30.5,C106,S +658,0,3,"Bourke, Mrs. John (Catherine)",female,32.0,1,1,364849,15.5,,Q +507,1,2,"Quick, Mrs. Frederick Charles (Jane Richards)",female,33.0,0,2,26360,26.0,,S +692,1,3,"Karun, Miss. Manca",female,4.0,0,1,349256,13.4167,,C +88,0,3,"Slocovski, Mr. Selman Francis",male,,0,0,SOTON/OQ 392086,8.05,,S +314,0,3,"Hendekovic, Mr. Ignjac",male,28.0,0,0,349243,7.8958,,S +800,0,3,"Van Impe, Mrs. Jean Baptiste (Rosalie Paula Govaert)",female,30.0,1,1,345773,24.15,,S +614,0,3,"Horgan, Mr. John",male,,0,0,370377,7.75,,Q +12,1,1,"Bonnell, Miss. Elizabeth",female,58.0,0,0,113783,26.55,C103,S +771,0,3,"Lievens, Mr. Rene Aime",male,24.0,0,0,345781,9.5,,S +365,0,3,"O'Brien, Mr. Thomas",male,,1,0,370365,15.5,,Q +876,1,3,"Najib, Miss. Adele Kiamie ""Jane""",female,15.0,0,0,2667,7.225,,C +195,1,1,"Brown, Mrs. James Joseph (Margaret Tobin)",female,44.0,0,0,PC 17610,27.7208,B4,C +594,0,3,"Bourke, Miss. Mary",female,,0,2,364848,7.75,,Q +654,1,3,"O'Leary, Miss. Hanora ""Norah""",female,,0,0,330919,7.8292,,Q +402,0,3,"Adams, Mr. John",male,26.0,0,0,341826,8.05,,S +83,1,3,"McDermott, Miss. Brigdet Delia",female,,0,0,330932,7.7875,,Q +669,0,3,"Cook, Mr. Jacob",male,43.0,0,0,A/5 3536,8.05,,S +878,0,3,"Petroff, Mr. Nedelio",male,19.0,0,0,349212,7.8958,,S +833,0,3,"Saad, Mr. Amin",male,,0,0,2671,7.2292,,C +75,1,3,"Bing, Mr. Lee",male,32.0,0,0,1601,56.4958,,S +722,0,3,"Jensen, Mr. Svend Lauritz",male,17.0,1,0,350048,7.0542,,S +251,0,3,"Reed, Mr. James George",male,,0,0,362316,7.25,,S +238,1,2,"Collyer, Miss. Marjorie ""Lottie""",female,8.0,0,2,C.A. 31921,26.25,,S +146,0,2,"Nicholls, Mr. Joseph Charles",male,19.0,1,1,C.A. 33112,36.75,,S +808,0,3,"Pettersson, Miss. Ellen Natalia",female,18.0,0,0,347087,7.775,,S +131,0,3,"Drazenoic, Mr. Jozef",male,33.0,0,0,349241,7.8958,,C +576,0,3,"Patchett, Mr. George",male,19.0,0,0,358585,14.5,,S +515,0,3,"Coleff, Mr. Satio",male,24.0,0,0,349209,7.4958,,S +847,0,3,"Sage, Mr. Douglas Bullen",male,,8,2,CA. 2343,69.55,,S +648,1,1,"Simonius-Blumer, Col. Oberst Alfons",male,56.0,0,0,13213,35.5,A26,C +443,0,3,"Petterson, Mr. Johan Emil",male,25.0,1,0,347076,7.775,,S +478,0,3,"Braund, Mr. Lewis Richard",male,29.0,1,0,3460,7.0458,,S +537,0,1,"Butt, Major. Archibald Willingham",male,45.0,0,0,113050,26.55,B38,S +169,0,1,"Baumann, Mr. John D",male,,0,0,PC 17318,25.925,,S +149,0,2,"Navratil, Mr. Michel (""Louis M Hoffman"")",male,36.5,0,2,230080,26.0,F2,S +290,1,3,"Connolly, Miss. Kate",female,22.0,0,0,370373,7.75,,Q +15,0,3,"Vestrom, Miss. Hulda Amanda Adolfina",female,14.0,0,0,350406,7.8542,,S +386,0,2,"Davies, Mr. Charles Henry",male,18.0,0,0,S.O.C. 14879,73.5,,S +811,0,3,"Alexander, Mr. William",male,26.0,0,0,3474,7.8875,,S +78,0,3,"Moutal, Mr. Rahamin Haim",male,,0,0,374746,8.05,,S +738,1,1,"Lesurer, Mr. Gustave J",male,35.0,0,0,PC 17755,512.3292,B101,C +452,0,3,"Hagland, Mr. Ingvald Olai Olsen",male,,1,0,65303,19.9667,,S +35,0,1,"Meyer, Mr. Edgar Joseph",male,28.0,1,0,PC 17604,82.1708,,C +347,1,2,"Smith, Miss. Marion Elsie",female,40.0,0,0,31418,13.0,,S +436,1,1,"Carter, Miss. Lucile Polk",female,14.0,1,2,113760,120.0,B96 B98,S +390,1,2,"Lehmann, Miss. Bertha",female,17.0,0,0,SC 1748,12.0,,C +657,0,3,"Radeff, Mr. Alexander",male,,0,0,349223,7.8958,,S +695,0,1,"Weir, Col. John",male,60.0,0,0,113800,26.55,,S +586,1,1,"Taussig, Miss. Ruth",female,18.0,0,2,110413,79.65,E68,S +384,1,1,"Holverson, Mrs. Alexander Oskar (Mary Aline Towner)",female,35.0,1,0,113789,52.0,,S +58,0,3,"Novel, Mr. Mansouer",male,28.5,0,0,2697,7.2292,,C +246,0,1,"Minahan, Dr. William Edward",male,44.0,2,0,19928,90.0,C78,Q +557,1,1,"Duff Gordon, Lady. (Lucille Christiana Sutherland) (""Mrs Morgan"")",female,48.0,1,0,11755,39.6,A16,C +605,1,1,"Homer, Mr. Harry (""Mr E Haven"")",male,35.0,0,0,111426,26.55,,C +350,0,3,"Dimic, Mr. Jovan",male,42.0,0,0,315088,8.6625,,S +659,0,2,"Eitemiller, Mr. George Floyd",male,23.0,0,0,29751,13.0,,S +415,1,3,"Sundman, Mr. Johan Julian",male,44.0,0,0,STON/O 2. 3101269,7.925,,S +713,1,1,"Taylor, Mr. Elmer Zebley",male,48.0,1,0,19996,52.0,C126,S +474,1,2,"Jerwan, Mrs. Amin S (Marie Marthe Thuillard)",female,23.0,0,0,SC/AH Basle 541,13.7917,D,C +139,0,3,"Osen, Mr. Olaf Elon",male,16.0,0,0,7534,9.2167,,S +224,0,3,"Nenkoff, Mr. Christo",male,,0,0,349234,7.8958,,S +221,1,3,"Sunderland, Mr. Victor Francis",male,16.0,0,0,SOTON/OQ 392089,8.05,,S +68,0,3,"Crease, Mr. Ernest James",male,19.0,0,0,S.P. 3464,8.1583,,S +622,1,1,"Kimball, Mr. Edwin Nelson Jr",male,42.0,1,0,11753,52.5542,D19,S +467,0,2,"Campbell, Mr. William",male,,0,0,239853,0.0,,S +525,0,3,"Kassem, Mr. Fared",male,,0,0,2700,7.2292,,C +17,0,3,"Rice, Master. Eugene",male,2.0,4,1,382652,29.125,,Q +430,1,3,"Pickard, Mr. Berk (Berk Trembisky)",male,32.0,0,0,SOTON/O.Q. 392078,8.05,E10,S +90,0,3,"Celotti, Mr. Francesco",male,24.0,0,0,343275,8.05,,S +486,0,3,"Lefebre, Miss. Jeannie",female,,3,1,4133,25.4667,,S +831,1,3,"Yasbeck, Mrs. Antoni (Selini Alexander)",female,15.0,1,0,2659,14.4542,,C +440,0,2,"Kvillner, Mr. Johan Henrik Johannesson",male,31.0,0,0,C.A. 18723,10.5,,S +244,0,3,"Maenpaa, Mr. Matti Alexanteri",male,22.0,0,0,STON/O 2. 3101275,7.125,,S +882,0,3,"Markun, Mr. Johann",male,33.0,0,0,349257,7.8958,,S +287,1,3,"de Mulder, Mr. Theodore",male,30.0,0,0,345774,9.5,,S +735,0,2,"Troupiansky, Mr. Moses Aaron",male,23.0,0,0,233639,13.0,,S +620,0,2,"Gavey, Mr. Lawrence",male,26.0,0,0,31028,10.5,,S +296,0,1,"Lewy, Mr. Ervin G",male,,0,0,PC 17612,27.7208,,C +187,1,3,"O'Brien, Mrs. Thomas (Johanna ""Hannah"" Godfrey)",female,,1,0,370365,15.5,,Q +629,0,3,"Bostandyeff, Mr. Guentcho",male,26.0,0,0,349224,7.8958,,S +123,0,2,"Nasser, Mr. Nicholas",male,32.5,1,0,237736,30.0708,,C +678,1,3,"Turja, Miss. Anna Sofia",female,18.0,0,0,4138,9.8417,,S +263,0,1,"Taussig, Mr. Emil",male,52.0,1,1,110413,79.65,E67,S +439,0,1,"Fortune, Mr. Mark",male,64.0,1,4,19950,263.0,C23 C25 C27,S +410,0,3,"Lefebre, Miss. Ida",female,,3,1,4133,25.4667,,S +497,1,1,"Eustis, Miss. Elizabeth Mussey",female,54.0,1,0,36947,78.2667,D20,C +522,0,3,"Vovk, Mr. Janko",male,22.0,0,0,349252,7.8958,,S +766,1,1,"Hogeboom, Mrs. John C (Anna Andrews)",female,51.0,1,0,13502,77.9583,D11,S +408,1,2,"Richards, Master. William Rowe",male,3.0,1,1,29106,18.75,,S +420,0,3,"Van Impe, Miss. Catharina",female,10.0,0,2,345773,24.15,,S +453,0,1,"Foreman, Mr. Benjamin Laventall",male,30.0,0,0,113051,27.75,C111,C +447,1,2,"Mellinger, Miss. Madeleine Violet",female,13.0,0,1,250644,19.5,,S +197,0,3,"Mernagh, Mr. Robert",male,,0,0,368703,7.75,,Q +227,1,2,"Mellors, Mr. William John",male,19.0,0,0,SW/PP 751,10.5,,S +852,0,3,"Svensson, Mr. Johan",male,74.0,0,0,347060,7.775,,S +763,1,3,"Barah, Mr. Hanna Assi",male,20.0,0,0,2663,7.2292,,C +257,1,1,"Thorne, Mrs. Gertrude Maybelle",female,,0,0,PC 17585,79.2,,C +407,0,3,"Widegren, Mr. Carl/Charles Peter",male,51.0,0,0,347064,7.75,,S +103,0,1,"White, Mr. Richard Frasar",male,21.0,0,1,35281,77.2875,D26,S +315,0,2,"Hart, Mr. Benjamin",male,43.0,1,1,F.C.C. 13529,26.25,,S +77,0,3,"Staneff, Mr. Ivan",male,,0,0,349208,7.8958,,S +632,0,3,"Lundahl, Mr. Johan Svensson",male,51.0,0,0,347743,7.0542,,S +750,0,3,"Connaghton, Mr. Michael",male,31.0,0,0,335097,7.75,,Q +627,0,2,"Kirkland, Rev. Charles Leonard",male,57.0,0,0,219533,12.35,,Q +96,0,3,"Shorney, Mr. Charles Joseph",male,,0,0,374910,8.05,,S +171,0,1,"Van der hoef, Mr. Wyckoff",male,61.0,0,0,111240,33.5,B19,S +881,1,2,"Shelley, Mrs. William (Imanita Parrish Hall)",female,25.0,0,1,230433,26.0,,S +95,0,3,"Coxon, Mr. Daniel",male,59.0,0,0,364500,7.25,,S +215,0,3,"Kiernan, Mr. Philip",male,,1,0,367229,7.75,,Q +39,0,3,"Vander Planke, Miss. Augusta Maria",female,18.0,2,0,345764,18.0,,S +774,0,3,"Elias, Mr. Dibo",male,,0,0,2674,7.225,,C +37,1,3,"Mamee, Mr. Hanna",male,,0,0,2677,7.2292,,C +181,0,3,"Sage, Miss. Constance Gladys",female,,8,2,CA. 2343,69.55,,S +177,0,3,"Lefebre, Master. Henry Forbes",male,,3,1,4133,25.4667,,S +812,0,3,"Lester, Mr. James",male,39.0,0,0,A/4 48871,24.15,,S +496,0,3,"Yousseff, Mr. Gerious",male,,0,0,2627,14.4583,,C +503,0,3,"O'Sullivan, Miss. Bridget Mary",female,,0,0,330909,7.6292,,Q +216,1,1,"Newell, Miss. Madeleine",female,31.0,1,0,35273,113.275,D36,C +395,1,3,"Sandstrom, Mrs. Hjalmar (Agnes Charlotta Bengtsson)",female,24.0,0,2,PP 9549,16.7,G6,S +720,0,3,"Johnson, Mr. Malkolm Joackim",male,33.0,0,0,347062,7.775,,S +213,0,3,"Perkin, Mr. John Henry",male,22.0,0,0,A/5 21174,7.25,,S +644,1,3,"Foo, Mr. Choong",male,,0,0,1601,56.4958,,S +583,0,2,"Downton, Mr. William James",male,54.0,0,0,28403,26.0,,S +132,0,3,"Coelho, Mr. Domingos Fernandeo",male,20.0,0,0,SOTON/O.Q. 3101307,7.05,,S +363,0,3,"Barbara, Mrs. (Catherine David)",female,45.0,0,1,2691,14.4542,,C +461,1,1,"Anderson, Mr. Harry",male,48.0,0,0,19952,26.55,E12,S +186,0,1,"Rood, Mr. Hugh Roscoe",male,,0,0,113767,50.0,A32,S +14,0,3,"Andersson, Mr. Anders Johan",male,39.0,1,5,347082,31.275,,S +1,0,3,"Braund, Mr. Owen Harris",male,22.0,1,0,A/5 21171,7.25,,S +694,0,3,"Saad, Mr. Khalil",male,25.0,0,0,2672,7.225,,C +476,0,1,"Clifford, Mr. George Quincy",male,,0,0,110465,52.0,A14,S +348,1,3,"Davison, Mrs. Thomas Henry (Mary E Finck)",female,,1,0,386525,16.1,,S +489,0,3,"Somerton, Mr. Francis William",male,30.0,0,0,A.5. 18509,8.05,,S +69,1,3,"Andersson, Miss. Erna Alexandra",female,17.0,4,2,3101281,7.925,,S +883,0,3,"Dahlberg, Miss. Gerda Ulrika",female,22.0,0,0,7552,10.5167,,S +18,1,2,"Williams, Mr. Charles Eugene",male,,0,0,244373,13.0,,S +31,0,1,"Uruchurtu, Don. Manuel E",male,40.0,0,0,PC 17601,27.7208,,C +619,1,2,"Becker, Miss. Marion Louise",female,4.0,2,1,230136,39.0,F4,S +526,0,3,"Farrell, Mr. James",male,40.5,0,0,367232,7.75,,Q +585,0,3,"Paulner, Mr. Uscher",male,,0,0,3411,8.7125,,C +274,0,1,"Natsch, Mr. Charles H",male,37.0,0,1,PC 17596,29.7,C118,C +715,0,2,"Greenberg, Mr. Samuel",male,52.0,0,0,250647,13.0,,S +438,1,2,"Richards, Mrs. Sidney (Emily Hocking)",female,24.0,2,3,29106,18.75,,S +193,1,3,"Andersen-Jensen, Miss. Carla Christine Nielsine",female,19.0,1,0,350046,7.8542,,S +275,1,3,"Healy, Miss. Hanora ""Nora""",female,,0,0,370375,7.75,,Q +173,1,3,"Johnson, Miss. Eleanor Ileen",female,1.0,1,1,347742,11.1333,,S +807,0,1,"Andrews, Mr. Thomas Jr",male,39.0,0,0,112050,0.0,A36,S +680,1,1,"Cardeza, Mr. Thomas Drake Martinez",male,36.0,0,1,PC 17755,512.3292,B51 B53 B55,C +304,1,2,"Keane, Miss. Nora A",female,,0,0,226593,12.35,E101,Q +370,1,1,"Aubart, Mme. Leontine Pauline",female,24.0,0,0,PC 17477,69.3,B35,C +239,0,2,"Pengelly, Mr. Frederick William",male,19.0,0,0,28665,10.5,,S +825,0,3,"Panula, Master. Urho Abraham",male,2.0,4,1,3101295,39.6875,,S +284,1,3,"Dorking, Mr. Edward Arthur",male,19.0,0,0,A/5. 10482,8.05,,S +182,0,2,"Pernot, Mr. Rene",male,,0,0,SC/PARIS 2131,15.05,,C +64,0,3,"Skoog, Master. Harald",male,4.0,3,2,347088,27.9,,S +404,0,3,"Hakkarainen, Mr. Pekka Pietari",male,28.0,1,0,STON/O2. 3101279,15.85,,S +479,0,3,"Karlsson, Mr. Nils August",male,22.0,0,0,350060,7.5208,,S +618,0,3,"Lobb, Mrs. William Arthur (Cordelia K Stanlick)",female,26.0,1,0,A/5. 3336,16.1,,S +3,1,3,"Heikkinen, Miss. Laina",female,26.0,0,0,STON/O2. 3101282,7.925,,S +337,0,1,"Pears, Mr. Thomas Clinton",male,29.0,1,0,113776,66.6,C2,S +764,1,1,"Carter, Mrs. William Ernest (Lucile Polk)",female,36.0,1,2,113760,120.0,B96 B98,S +696,0,2,"Chapman, Mr. Charles Henry",male,52.0,0,0,248731,13.5,,S +783,0,1,"Long, Mr. Milton Clyde",male,29.0,0,0,113501,30.0,D6,S +318,0,2,"Moraweck, Dr. Ernest",male,54.0,0,0,29011,14.0,,S +706,0,2,"Morley, Mr. Henry Samuel (""Mr Henry Marshall"")",male,39.0,0,0,250655,26.0,,S +432,1,3,"Thorneycroft, Mrs. Percival (Florence Kate White)",female,,1,0,376564,16.1,,S +50,0,3,"Arnold-Franchi, Mrs. Josef (Josefine Franchi)",female,18.0,1,0,349237,17.8,,S +136,0,2,"Richard, Mr. Emile",male,23.0,0,0,SC/PARIS 2133,15.0458,,C +889,0,3,"Johnston, Miss. Catherine Helen ""Carrie""",female,,1,2,W./C. 6607,23.45,,S +604,0,3,"Torber, Mr. Ernst William",male,44.0,0,0,364511,8.05,,S +5,0,3,"Allen, Mr. William Henry",male,35.0,0,0,373450,8.05,,S +613,1,3,"Murphy, Miss. Margaret Jane",female,,1,0,367230,15.5,,Q +724,0,2,"Hodges, Mr. Henry Price",male,50.0,0,0,250643,13.0,,S +758,0,2,"Bailey, Mr. Percy Andrew",male,18.0,0,0,29108,11.5,,S +142,1,3,"Nysten, Miss. Anna Sofia",female,22.0,0,0,347081,7.75,,S +416,0,3,"Meek, Mrs. Thomas (Annie Louise Rowley)",female,,0,0,343095,8.05,,S +668,0,3,"Rommetvedt, Mr. Knud Paust",male,,0,0,312993,7.775,,S +387,0,3,"Goodwin, Master. Sidney Leonard",male,1.0,5,2,CA 2144,46.9,,S +87,0,3,"Ford, Mr. William Neal",male,16.0,1,3,W./C. 6608,34.375,,S +94,0,3,"Dean, Mr. Bertram Frank",male,26.0,1,2,C.A. 2315,20.575,,S +650,1,3,"Stanley, Miss. Amy Zillah Elsie",female,23.0,0,0,CA. 2314,7.55,,S +508,1,1,"Bradley, Mr. George (""George Arthur Brayton"")",male,,0,0,111427,26.55,,S +571,1,2,"Harris, Mr. George",male,62.0,0,0,S.W./PP 752,10.5,,S +317,1,2,"Kantor, Mrs. Sinai (Miriam Sternin)",female,24.0,1,0,244367,26.0,,S +229,0,2,"Fahlstrom, Mr. Arne Jonas",male,18.0,0,0,236171,13.0,,S +656,0,2,"Hickman, Mr. Leonard Mark",male,24.0,2,0,S.O.C. 14879,73.5,,S +281,0,3,"Duane, Mr. Frank",male,65.0,0,0,336439,7.75,,Q +753,0,3,"Vande Velde, Mr. Johannes Joseph",male,33.0,0,0,345780,9.5,,S +803,1,1,"Carter, Master. William Thornton II",male,11.0,1,2,113760,120.0,B96 B98,S +527,1,2,"Ridsdale, Miss. Lucy",female,50.0,0,0,W./C. 14258,10.5,,S +739,0,3,"Ivanoff, Mr. Kanio",male,,0,0,349201,7.8958,,S +579,0,3,"Caram, Mrs. Joseph (Maria Elias)",female,,1,0,2689,14.4583,,C +54,1,2,"Faunthorpe, Mrs. Lizzie (Elizabeth Anne Wilkinson)",female,29.0,1,0,2926,26.0,,S +867,1,2,"Duran y More, Miss. Asuncion",female,27.0,1,0,SC/PARIS 2149,13.8583,,C +351,0,3,"Odahl, Mr. Nils Martin",male,23.0,0,0,7267,9.225,,S +80,1,3,"Dowdell, Miss. Elizabeth",female,30.0,0,0,364516,12.475,,S +856,1,3,"Aks, Mrs. Sam (Leah Rosen)",female,18.0,0,1,392091,9.35,,S +872,1,1,"Beckwith, Mrs. Richard Leonard (Sallie Monypeny)",female,47.0,1,1,11751,52.5542,D35,S +836,1,1,"Compton, Miss. Sara Rebecca",female,39.0,1,1,PC 17756,83.1583,E49,C +793,0,3,"Sage, Miss. Stella Anna",female,,8,2,CA. 2343,69.55,,S +521,1,1,"Perreault, Miss. Anne",female,30.0,0,0,12749,93.5,B73,S diff --git a/tests/data/openai/embedding.json b/tests/data/openai/embedding.json new file mode 100644 index 000000000..249c78ecf --- /dev/null +++ b/tests/data/openai/embedding.json @@ -0,0 +1 @@ +{"object": "list", "data": [{"object": "embedding", "index": 0, "embedding": [-0.01999368, -0.02016083, 0.013037679, -0.011751912, -0.02810687, 0.0056188027, -0.011726197, -0.01088402, 0.01021542, -0.010967594, 0.0113276085, -0.0106332945, -0.012806241, -0.021626605, -0.00513664, 0.0023031305, 0.021343736, -0.0029026193, 0.009951838, -0.013114825, -0.0057730945, 0.0065799137, -0.016084947, -0.027309695, -0.011906204, 0.0066474164, 0.02921263, -0.013436267, -0.009096803, -0.00037287248, 0.033378515, -0.022912372, -0.036027197, -0.0077338894, -0.02307952, -0.011784056, -0.018527905, -0.0094182445, 0.02557391, -0.011276178, 0.017820733, 0.023670973, 0.0017293568, -0.0031501297, -0.0016192631, 0.01044043, 0.009071087, -0.014670604, -0.017820733, 0.010646152, 0.018656481, -0.010691154, -0.022410922, -0.017692156, -0.024391003, 0.010993309, -0.01088402, 0.0073545882, -0.000542433, -0.0028238662, 0.008569638, 0.0073481593, -0.027438272, 0.0018209678, -0.001176477, 0.00038673467, 0.010376141, 0.02259093, 0.03656722, 0.0010904913, 0.012581232, -0.0006497142, 0.010929021, 0.0024638514, 0.0135777015, 0.01380914, 0.009270381, 0.008486063, 0.01402772, -0.0069495714, 0.012414082, -0.032658488, 0.018785058, 0.013217687, 0.020842286, 0.016753547, -0.026743958, 0.021446597, -0.021112297, -0.008666071, 0.011591191, 0.025149606, 0.00043796445, 0.015699217, -0.0024863523, 0.020996578, 0.010954737, 0.022063766, 0.010948308, -0.024223853, -0.011944777, 0.0033365658, -0.010061128, -0.000753781, -0.019389369, -0.013616274, 0.004429468, -0.004220531, 0.0024043845, 0.016059233, -0.020276548, 0.024609584, 0.0053873644, -0.050222065, -0.0033654957, 0.0017872164, 0.009836119, -0.014567742, -0.0073545882, -0.0033719244, 0.009006799, 0.014850611, 0.033327084, -0.019157931, 0.026846819, 0.0026358226, -0.009758973, -0.023979558, -0.013783424, -0.00845392, 0.039473053, 0.007766034, 0.01595637, 0.0010205777, -0.022346634, 0.03584719, -0.018373612, 0.0006288205, -0.0005010474, -0.0043651797, 0.010871162, 0.035075728, -0.015673501, 0.004352322, -0.0023240242, 0.01530063, -0.0027097543, 0.00085583876, 0.015506352, -0.0063581187, 0.022655217, -0.004063024, 0.013886286, -0.0037640834, -0.010350426, 0.0055705863, 0.00433625, 0.014169155, 0.011700481, -0.011218319, 0.00019658175, 0.008691786, -0.0035390742, -0.028312594, 0.0018241822, 0.043330353, 0.022410922, 0.03155273, -0.0077403183, -0.025381044, -0.028724039, 0.021755181, -0.024995314, 0.008903937, -0.017036416, -0.009411816, -0.001938294, 0.0108904485, -0.017962167, -0.002576356, -0.024185281, 0.0028126156, 0.020726567, 0.010479002, -0.010118987, -0.018836489, 0.019350797, -0.028415455, 0.007746747, 0.006814566, 0.008762503, 0.032504193, 0.014207727, -0.01402772, -0.67847365, -0.033507094, -0.00029090483, -0.008595354, -0.0021568744, 0.0083382, -0.014002005, 0.0021022293, -0.014837753, 0.0018707912, -0.010787587, 0.004738052, -0.012041209, 0.0019800814, 0.019299366, -0.016252097, 0.020456556, -0.0022050908, -0.0056959484, 0.0107168695, -0.009379672, 0.012439798, 0.01794931, 0.0020475842, 0.008357487, -0.008209623, 0.02529104, 0.007328873, -0.012960534, 0.04567045, -0.04353608, 0.017242137, 0.015570641, -0.021999476, 0.05909386, 0.00601739, -0.019183647, 0.028364023, 0.016714973, 0.038341578, -0.035410028, -0.0023593828, 0.00016162495, 0.0012110319, -0.0011740661, -0.008209623, 0.011366182, -0.00021054437, 0.0005291735, -0.020199403, 0.0016039945, 0.015596356, 0.022449495, 0.0065574124, 0.00563166, -0.006898141, 0.0336871, 0.005004849, 0.010041841, 0.01855362, 0.0098489765, -0.0027499346, -0.0060334625, -0.01933794, 0.0019045427, 0.025805347, -0.012086212, 0.008003901, 0.01971081, -0.018913636, 0.009681826, -0.00043836626, -0.009816833, 0.01838647, -0.007361017, 0.021716608, 0.01756358, -0.0052555734, -0.01821932, 0.0291612, -0.0086339265, -0.0009860228, 0.000753781, -0.008762503, 0.033224225, -0.013796282, -0.016097805, 0.016599255, 0.010839017, 0.0010358462, 1.2945566e-05, 0.0053809355, 0.0031726304, -0.010922592, 0.0065349117, 0.0069174273, -0.019350797, -0.01044043, 0.016097805, -0.015197768, -0.008183908, -0.0035551463, 0.00601739, -0.004018022, -0.0124526555, 0.011160459, -0.012311221, 0.01065901, 0.043047484, -0.0035647894, -0.0030810195, -0.008588925, -0.0006641791, -0.0033654957, 0.029109769, -0.032761347, 0.01170691, 0.008048902, 0.010138274, 0.0050562792, 0.027952578, -0.015827794, 0.016059233, -0.007361017, 0.0027354697, 0.0075474535, -0.0108004445, -0.008434633, -0.01143047, 0.0044198246, 0.007881753, 0.0012696951, -0.007129579, -0.0007541828, -0.002476709, -0.0018836489, 0.02214091, -0.017357856, 0.006145967, -0.011803343, -0.014387735, -0.0062809726, 0.005397008, -0.015082049, 0.011494759, -0.012111926, -0.009733258, -0.004146599, -0.0012745167, 0.0041048117, -0.002721005, -0.026319655, -0.01115403, 0.03600148, 0.010080415, -0.0119319195, -0.013744852, -0.003100306, -0.018527905, -0.010138274, -0.0057120207, 0.0032208469, 0.0034651426, -0.0043716086, -0.011764769, -0.027515417, -0.017717872, 0.02016083, 0.0027226121, -0.015660644, -0.016072089, -0.017422145, 0.012066925, -0.0007702549, -0.0059016715, 0.0012351401, -0.01247837, -0.03695295, -0.0040790965, -0.016663542, 0.014349162, -0.0026968967, 0.007200296, 0.0045869746, 0.023709547, -0.00392159, -0.015390634, 0.032889925, -0.01766644, -0.020340838, -0.009533964, 0.010376141, 0.007258156, 0.0049502035, 0.012999106, -0.008762503, -0.016444962, 0.022372348, 0.044487543, 0.005975603, -0.016573539, 0.0058245254, 0.01722928, -0.009823261, 0.022449495, -0.022552356, 0.0006256061, -0.019260792, 0.00878179, 0.01650925, 0.0128833875, 0.004670549, -0.036104344, 0.004284819, -0.008685357, 0.024661014, 0.003854087, 0.022192342, 0.002960479, -0.0014167547, -0.0043491074, -0.00043515183, 0.0028029724, 0.0015139908, 0.00999684, -0.0006637773, 0.004946989, 0.042198878, -0.012999106, -0.014310589, -0.013108396, -0.0018482903, -0.0017052487, -0.0036708652, 0.00083173066, 0.0026952894, 0.035075728, -0.018013598, 0.023298101, 0.0064770523, -0.027155403, -0.0037351537, 0.049013443, -0.009636825, 0.019286508, 0.035384312, 0.02766971, -0.002584392, 0.0052748597, 0.0011652265, -0.025689628, -0.0066795605, -0.039138753, 0.02032798, 0.0145806, -0.0039280187, 0.0020250834, 0.0035808615, 0.031989887, 0.009701113, 0.02800401, 0.0016988199, 0.010350426, 0.01563493, 0.024159566, -0.007958899, -0.012330507, -0.01982653, -0.0025136748, 0.022295203, -0.0044133957, 0.0011660301, -0.0061041797, 0.002116694, 0.01595637, 0.0051848562, 0.009173949, 0.010067557, -0.0036547931, 0.013371979, -0.0017181064, -0.019453658, 0.019787956, 0.01049186, -0.0046287617, -0.015917799, -0.028801184, -0.0035197877, -0.012864101, 0.015583498, 0.0028174373, 0.028904047, 0.005593087, -0.007946041, -0.019196505, -0.0043651797, 0.012414082, -0.0017438218, 0.0126262335, 0.01590494, 0.009302526, 0.013783424, -0.01016399, -0.011623335, 0.011057598, 0.00336871, 0.005290932, -0.016252097, -0.0001511781, 0.009861834, -0.007598884, -0.0291612, -0.019260792, 0.0017952524, -0.012966962, 0.0030954846, -0.019839387, -0.019325081, 0.039627343, 0.0039698062, -0.006820995, -0.01496633, -0.027695425, -0.001907757, 0.048782006, 0.012941247, 0.0066152723, -0.010794016, 0.0019865104, -0.0013034465, -0.013963431, -0.031938456, 0.002452601, -0.01960795, -0.026203936, -0.0030617332, 0.007798178, -0.0039248043, 0.023156667, 0.00045443833, -0.00050546724, 0.0014416665, 0.0066024144, -0.004538758, 0.0023931342, -0.00081405137, 0.010427572, 0.009270381, 0.0062166844, -0.005538442, 0.026242508, 0.02689825, -0.003815514, -0.027926862, -0.01121189, 0.02738684, 0.0055480856, -0.009778259, -0.024558153, 0.012182644, -0.0078046066, 0.00070235034, 0.014207727, -0.0007156098, 0.04127313, 0.046261903, 0.009964695, -0.027052542, 0.000965129, 0.018695055, -0.0133205475, 0.012182644, -0.014194869, 0.0061170375, 0.034741428, -0.009746116, -0.025998212, -0.00040782927, 0.018245036, 0.016496394, -0.027078256, -0.012992677, -0.013011964, -0.052716453, -0.0011025454, -0.0029829799, -0.010530434, -0.011970492, 0.003944091, -0.020109398, 0.003384782, -0.0041176695, -0.02043084, -0.005049851, -0.0053166472, -0.022539498, -0.023902413, 0.0006392674, -0.0011202247, 0.018026456, -0.0049502035, -0.013204829, -0.0028110086, 0.041118834, 0.0005388168, -0.0005552907, -0.0032787062, -0.028595462, -0.021678034, -0.0025570695, -0.004953418, -0.008216052, -0.002645466, -0.0017213208, 0.032812778, 0.00955325, 0.006030248, 0.007444592, 0.00091932353, 0.0029942302, -0.0022822367, 0.023735262, -0.032118466, 0.013114825, 0.020070825, -0.024429576, -0.0045869746, -0.00455483, 0.013616274, -0.004345893, 0.017987883, 0.031064136, -0.05251073, 0.0071810097, 0.006435265, -0.012523373, 0.009411816, -0.0057313074, 0.0128833875, 0.003860516, -0.009386101, 0.010626866, -0.010755442, -0.029392637, 0.013384837, -0.030421251, 0.0063581187, 0.031321287, 0.01888792, 0.021163728, -0.01474775, -0.010376141, 0.004130527, -0.019170789, -0.00850535, 0.010350426, -0.0031244142, 0.010009698, -0.027541133, -0.0048312703, -0.015364918, 0.013005535, 0.0010085236, -0.021215158, -0.029624077, 0.0015147944, -0.0013934502, -0.01960795, -0.021189444, -0.032787062, -0.0092382375, 0.012227646, -0.003899089, 0.020070825, -0.0065188394, 0.01690784, -0.0012054067, 0.023542397, -0.01828361, -0.03422712, -0.01530063, 0.0027001111, 0.03103842, 0.023876697, 0.012375509, -0.022513783, 0.02512389, 0.0033558523, -0.02021226, -0.005577015, -0.018257894, -0.0109804515, -0.03417569, 0.008518208, 0.009051801, 0.018566478, -0.0074960226, -0.01551921, -0.017370714, -0.0097654015, -0.0041015972, -0.004821627, -0.009206093, -0.012934818, 0.0047573387, 0.0021504457, -0.015017761, 0.017074987, -0.0040276656, 0.008601783, 0.02921263, 0.0013902357, 0.019029355, 0.029135484, 0.020366551, -0.008003901, 0.005483797, -0.014130581, 0.008704644, -0.017345, -0.039473053, 0.014567742, -0.017332142, 0.030986989, 0.023825265, 0.030986989, -0.004847342, 0.015249198, -0.001099331, 0.016714973, -0.010832588, -0.024506722, 0.008871794, -0.021279447, -0.025213894, -0.032349903, -0.016149236, -0.044873275, 0.003699795, 0.016329244, -0.013500555, 0.0127998125, -0.012767668, -0.013963431, 0.0050466363, 0.016740689, 0.05073637, 0.009456818, 0.020790854, 0.006329189, -0.001030221, -0.019132216, 0.016046375, -0.018605052, -0.020353695, 0.019183647, 0.018360754, 0.011925491, 0.0006927071, 0.017460719, -0.002438136, -0.011726197, -0.02738684, 0.02307952, 0.0077403183, -0.012934818, -0.01253623, -0.00049100234, -0.014657746, 0.017062131, -0.01937651, -0.018540762, 0.008518208, -0.013989147, -0.024146708, 0.035410028, 0.0012648734, 0.030524112, 0.0101125585, -0.000100802135, -0.007991043, 0.0023674187, 0.019003639, 0.005081995, 0.0033044217, 0.0007702549, -0.011713339, 0.0045773312, -0.008344629, 0.0056284457, -0.019183647, 0.0074574496, 0.003429784, 0.02523961, -0.012491228, -0.022603787, -0.024172423, 0.003060126, 0.021112297, 0.0011587976, -0.002344918, -0.0133205475, -0.03157844, -0.016586397, 0.024866737, -0.015750648, 0.0067952797, -0.008183908, -0.0153134875, 0.0037640834, 0.003407283, -0.023516681, -0.0075860266, -0.023490967, -0.011282607, -0.013371979, 0.003799442, 0.016264955, 0.02622965, 0.016046375, 0.020173687, 0.016496394, -0.00045966177, 0.023015233, 0.0050594937, 0.012819099, -0.015390634, -0.0048794863, 0.0027049328, -0.03533288, -0.0043169633, -0.014953473, -0.0035519318, -0.025046745, 0.023683831, 0.025998212, -0.012291934, 0.014837753, 0.011481901, 0.040013075, -0.013886286, -0.021009436, -0.022436637, 0.025535336, -0.008093905, 0.011751912, 0.008955369, 0.0065027676, -0.018862205, -0.011173317, -0.009662541, 0.002531354, -0.025226751, -0.02275808, -0.0060945363, 0.026435373, -0.014182012, -0.020019395, -0.022950944, 0.013564844, -0.0056413035, 0.01838647, 0.00068828725, 0.004766982, 0.01518491, 0.02495674, 0.010408285, -0.0050980668, 0.007746747, -0.043998953, -0.014760607, -0.0047862683, -0.02666681, -0.008132477, -0.018630767, -0.008222481, 0.01369342, -0.027155403, -0.051893562, 0.0008445883, -0.01744786, -0.018373612, -0.021215158, -0.006444908, 0.0065477695, -0.0012954104, -0.022410922, 0.015982086, 0.007624599, 0.014824895, -0.008653213, -0.011436899, 0.010388999, 0.006891712, -0.008100334, 0.005644518, -0.0046930504, 0.00038974817, 0.020610848, 0.01563493, -0.010684725, 0.030524112, -0.013873428, 0.013166256, -0.018013598, 0.008511779, 0.008820363, 0.013912001, 0.0032545982, -0.008344629, -0.023400962, 0.012343365, 0.021652319, 0.02016083, -0.009302526, 0.023349533, -0.016817834, 0.022449495, -0.009450389, 0.013847712, -0.006454551, -0.012433369, 0.0084603485, -0.02529104, -0.036387213, 0.018913636, -0.03340423, -0.010041841, 0.002576356, 0.006454551, -0.018206464, 0.014156297, 0.04353608, -0.018129317, 0.02512389, 0.0030954846, 0.0074638785, -0.024352431, -0.0062713292, -0.0023111666, 0.0013500556, -0.014503454, 0.004622333, 0.003429784, -0.013031251, -0.009122518, -0.009077516, -0.0005717646, 0.001050311, -0.0011162066, -0.0028961906, -0.0073803035, 0.0033076361, -0.0013580916, -0.0042719613, -0.016740689, -0.0060977507, 0.011816201, 0.002783686, 0.009257523, 0.24110706, -0.019530803, 0.019080784, 0.031141281, -0.009926123, 0.007997472, -0.008704644, -0.013166256, 0.0015895297, 0.004783054, -0.0006718133, -0.001288178, -0.02766971, -0.0012037995, -0.0015871188, -0.002460637, -0.014452023, -0.007798178, -0.028415455, -0.04312463, -0.010704012, -0.025779633, -0.008473205, -6.790458e-05, 0.010832588, -0.0057055918, -0.013731994, 0.011256891, 0.031321287, 0.017589295, -0.010286137, -0.020366551, 0.0053037894, 0.00023967504, -0.010562577, -0.007836751, -0.0045805457, 0.007978185, 0.023092378, 0.042173162, -0.013294833, -0.0066088433, 0.012819099, -0.0016425676, -0.007759605, 0.010408285, -0.010408285, -0.020958005, -0.001645782, 0.018875062, -0.013204829, 0.018836489, 0.018103601, 0.039190184, -0.019067928, 0.005435581, 0.0010888841, 0.0035165732, -0.0037705123, 0.015416348, -0.006814566, 0.020469414, -0.009488962, 0.0027660066, -0.008312485, -0.0018643624, -0.020057969, -0.007643886, -0.0015292594, -0.010343997, -0.031166997, -0.0023433107, 0.01253623, -0.0037126527, -0.041658856, -0.02176804, 0.049116306, 0.003545503, 0.028415455, 0.04633905, -0.014927757, -0.014079151, -0.0033333513, -0.021896616, 0.011109028, -0.037210103, 0.031604156, 0.008396059, -0.016162094, 0.01535206, 9.638231e-05, -0.025972497, 0.016663542, 0.0046673347, -0.0002151651, 0.03162987, -0.01684355, 0.023130951, 0.0051719984, 0.009296097, -0.02959836, -0.0071938676, 0.0059241722, -0.0001261659, -0.014400592, 0.0012745167, -0.0126262335, -0.017705014, 0.015364918, -0.0024477793, -0.01684355, -0.012439798, 0.018707912, -0.026409658, -0.01281267, -0.010614008, 0.0191065, 0.0013757709, 0.0006967251, 0.014220585, 0.0054387953, -0.021935187, 0.016393531, 0.0092382375, -0.022796651, -0.013770566, -0.0058566695, 0.0042430316, 0.022989517, -0.015943512, 0.025278183, -0.005361649, 0.015043476, -0.025946781, -0.021588031, 0.020945147, 0.022848083, -0.008100334, 0.010266851, 0.009694684, 0.003008695, 0.004191601, 0.004738052, -0.008106762, 0.0073095863, -0.017589295, 0.0066731316, 0.0009281632, -0.012021923, -0.010086844, -0.038984463, -0.004480899, 0.0024943883, -0.014722034, 0.010974023, -0.0127998125, -0.024943883, -0.030678404, 0.002169732, 0.022719506, -0.024712445, -0.0071938676, 0.0031276287, -0.027361125, -0.035924334, -0.0074767363, -0.16581254, 0.03026696, 0.017267853, -0.007759605, 0.0019350796, 0.013256259, 0.0101961335, -0.0062038265, -0.0066667027, -8.598568e-05, 0.02307952, 0.0005066726, -0.054927975, -0.0023593828, 0.013018393, 0.010710441, -0.010678296, 0.017679298, -0.001503544, 0.017087845, 0.015146337, -0.0063099023, -0.003600148, 0.014837753, -0.023812408, 0.006522054, -0.013886286, 0.028029725, -0.025136748, 0.012671236, -0.032221325, -0.011546189, 0.027772572, 0.017525006, 0.0054580816, -0.0027338625, 0.019247934, -0.0170107, 0.016637828, 0.006377405, 0.013487698, 0.02766971, 0.00039898962, 0.00944396, -0.018193606, 0.014696319, 0.0041273125, -0.015750648, 0.029521214, -0.0050080633, -0.018347898, -0.011880489, 0.022475211, 0.006583128, 0.0134105515, 0.026435373, 0.002676003, 0.0022645574, 0.016612113, -0.0057570226, 0.019646522, -0.03026696, 0.0012303184, -0.025856778, -0.002221163, -0.022436637, -0.012304792, 0.003423355, -0.008582496, 0.023015233, -0.000782309, -0.004821627, 0.026795387, -0.011301894, 0.0069560003, 0.013461983, -0.01530063, 0.023426678, 0.006554198, -0.0038122996, -0.016046375, 0.02098372, 0.00017739569, 0.015506352, 0.009630396, 0.022873798, 0.0132305445, -0.0012857672, -0.018566478, -0.0075024515, 0.04811341, -0.017717872, -0.010009698, 0.004384466, -0.002184197, 0.008511779, -0.015506352, 0.0058952426, -0.0062102554, -0.027772572, -0.0063870484, -0.015943512, -0.009726829, 0.008351058, 0.020996578, 0.008813934, 0.011829058, 0.0077017453, 0.029778369, -0.015043476, -0.0073803035, 0.0132305445, 0.009013228, 0.029906945, 0.003568004, 0.035692897, -0.014902041, -0.0030954846, -0.008415346, -0.00767603, 0.05533942, -0.013500555, -0.008601783, 0.0085503515, -0.01513348, -0.010144703, -0.058888137, -0.031141281, 0.0239667, 0.023259528, -0.008537494, 0.005397008, 0.0045355437, 0.015082049, -0.029418353, 0.016856408, -0.0056927344, -0.015827794, -0.012966962, -0.004468041, 0.038007278, -0.022873798, -0.009116089, 0.005233072, -0.013731994, 0.0239667, -0.025831062, -0.0012889816, 0.0011113851, -0.009681826, -0.0065477695, 0.0015903333, -0.04585046, 0.014734892, 0.0066538453, 0.010607579, -0.0043748226, 0.0013404123, 0.008293198, -0.021279447, -0.022449495, -0.010652581, -0.023825265, -0.006859568, 0.020585133, -0.030035522, 0.012156929, -0.00090244785, -0.0010896877, -0.021498026, -0.010646152, -0.005898457, -0.0038476584, 0.017306427, 0.00065453583, -0.031681303, -0.018913636, -0.024095276, -0.03155273, 0.023555255, 0.025561051, -0.021125155, 0.014477738, 0.021793753, 0.018836489, 0.005840597, 0.012555516, -0.00025313543, -0.023696689, 0.019633666, 0.013359121, 0.018990781, -0.026769673, 0.003452285, 0.012465512, 0.0035326453, -0.0028511886, 0.025329614, -0.016496394, 0.009861834, -0.010999738, -0.008537494, -0.008736788, -0.01236908, 0.018707912, -0.006512411, -0.00576988, -0.02700111, 0.002184197, 0.0014633638, 0.024095276, 0.01717785, 0.011546189, -0.018309325, -0.009598252, -0.016316386, -0.0052427156, 0.018759344, 0.00472198, -0.018849347, -0.014053435, -0.0061266804, 0.0035326453, -0.0066152723, 0.0032176324, 0.0042494605, 6.438881e-05, -0.018322183, -0.07154009, 0.021960903, -0.0071488656, -0.026923966, 0.015660644, -0.0101125585, 0.008813934, -0.026641095, 0.012876959, -0.011790485, -0.006885283, 0.016136378, -0.0010792408, 0.012221217, -0.004378037, -0.00635169, 0.035307165, -0.0033815678, 0.00850535, 0.010549719, 0.0059788176, 0.0037705123, 0.020289406, -0.014812038, 0.019312223, -0.0035776473, -0.012439798, 0.019800814, -0.033275656, 0.0011571904, -0.0046962644, -0.037827272, 1.2041511e-05, 0.023053806, -0.0024799234, -0.033661384, 0.012407653, 0.009855405, 0.013307691, 0.0065895566, -0.01694641, -0.033095647, 0.01888792, -0.029701222, -0.019852245, -0.0050466363, -0.017306427, 0.011835487, 0.022012334, -0.0020925861, 0.01690784, 0.027309695, -0.02302809, -0.00051189604, 0.019967964, -0.055905156, 0.028646892, 0.028955476, 0.0015509566, -7.850211e-05, 0.023696689, 0.010929021, 0.012613376, -0.017692156, -0.00037447968, 0.009983982, -0.011141173, -0.008801077, 0.0015887261, -0.03440713, -0.009386101, 0.0063806195, 0.002992623, 0.009701113, 0.0066859894, 0.0031019133, -0.0063034734, 0.008498921, -0.026242508, 0.023606686, 0.013513413, 0.0017454289, -0.008376773, -0.00201544, 0.048164837, 0.0074188765, -0.0010181669, -0.017190708, 0.008029616, 0.029572645, -0.025008172, -0.005091638, -0.024866737, 0.007271013, -0.002328846, 0.0062713292, -0.016894981, 7.393161e-05, 0.022732364, 0.012375509, 0.0014272016, 1.2298916e-05, -0.0191065, -0.016637828, -0.01452917, -0.012137642, -0.02307952, -0.0001341015, 0.004043738, 0.024545295, 0.014516312, -0.01766644, -0.020893717, 0.009263952, 0.008563209, 0.0018948994, -0.013500555, -0.0034780002, -0.015814936, 0.044204675, 0.008093905, 0.007367446, 0.011366182, 0.004853771, 0.0030826267, -0.0080231875, -0.006621701, -0.03985878, 0.007791749, -0.00018603443, -0.0026872533, -0.016419247, -0.008408917, -0.027489703, -0.024545295, 0.0034490705, -0.020456556, 0.010427572, 0.01578922, 0.04991348, -0.0014159511, -0.005191285, 0.021253731, 0.00052837, 0.03108985, 0.0034940722, 0.0030553043, 0.0004680996, -0.009630396, 0.0140148625, -0.031115565, -0.013976289, -0.007766034, -0.021742323, -0.0062552574, -0.017164992, 0.013513413, -0.025535336, -0.006444908, 0.027412556, 0.0075345957, 0.01264552, -0.0009112875, -0.029315492, -0.021215158, 0.028801184, -0.0032497765, -0.020687994, -0.03129557, 0.0037962275, -0.001365324, -0.02805544, -0.005638089, 0.02689825, -0.007695317, -0.0027724355, -0.00074895937, -0.0056798765, 0.0045580445, -0.008325342, -0.008858936, -0.0070717195, -0.020276548, 0.03600148, -0.0047123367, -0.016599255, 0.01573779, -0.028595462]}], "model": "text-embedding-ada-002-v2", "usage": {"prompt_tokens": 3, "total_tokens": 3}} \ No newline at end of file diff --git a/tests/data/rsp_cache.json b/tests/data/rsp_cache.json index db452f676..eb67021a5 100644 --- a/tests/data/rsp_cache.json +++ b/tests/data/rsp_cache.json @@ -141,5 +141,239 @@ "\nRole: You are a senior development and qa engineer, your role is summarize the code running result.\nIf the running result does not include an error, you should explicitly approve the result.\nOn the other hand, if the running result indicates some error, you should point out which part, the development code or the test code, produces the error,\nand give specific instructions on fixing the errors. Here is the code info:\n\n## Development Code File Name\na.txt\n## Development Code\n```python\nprint('Hello, World')\n```\n## Test File Name\n\n## Test Code\n```python\nNone\n```\n## Running Command\n\n## Running Output\nstandard output: \n```text\n\n```\nstandard errors: \n```text\n\n```\n\nNow you should begin your analysis\n---\n## instruction:\nPlease summarize the cause of the errors and give correction instruction\n## File To Rewrite:\nDetermine the ONE file to rewrite in order to fix the error, for example, xyz.py, or test_xyz.py\n## Status:\nDetermine if all of the code works fine, if so write PASS, else FAIL,\nWRITE ONLY ONE WORD, PASS OR FAIL, IN THIS SECTION\n## Send To:\nPlease write Engineer if the errors are due to problematic development codes, and QaEngineer to problematic test codes, and NoOne if there are no errors,\nWRITE ONLY ONE WORD, Engineer OR QaEngineer OR NoOne, IN THIS SECTION.\n---\nYou should fill in necessary instruction, status, send to, and finally return all content between the --- segment line.\n": "---\ninstruction: There are no errors in the provided code.\n\nFile To Rewrite: N/A\n\nStatus: PASS\n\nSend To: NoOne\n---", "\nRole: You are a senior development and qa engineer, your role is summarize the code running result.\nIf the running result does not include an error, you should explicitly approve the result.\nOn the other hand, if the running result indicates some error, you should point out which part, the development code or the test code, produces the error,\nand give specific instructions on fixing the errors. Here is the code info:\n\n## Development Code File Name\na.sh\n## Development Code\n```python\necho 'Hello World'\n```\n## Test File Name\n\n## Test Code\n```python\nNone\n```\n## Running Command\necho Hello World\n## Running Output\nstandard output: \n```text\nHello World\n\n```\nstandard errors: \n```text\n\n```\n\nNow you should begin your analysis\n---\n## instruction:\nPlease summarize the cause of the errors and give correction instruction\n## File To Rewrite:\nDetermine the ONE file to rewrite in order to fix the error, for example, xyz.py, or test_xyz.py\n## Status:\nDetermine if all of the code works fine, if so write PASS, else FAIL,\nWRITE ONLY ONE WORD, PASS OR FAIL, IN THIS SECTION\n## Send To:\nPlease write Engineer if the errors are due to problematic development codes, and QaEngineer to problematic test codes, and NoOne if there are no errors,\nWRITE ONLY ONE WORD, Engineer OR QaEngineer OR NoOne, IN THIS SECTION.\n---\nYou should fill in necessary instruction, status, send to, and finally return all content between the --- segment line.\n": "The running result indicates no error. The code works fine.\n\n## File To Rewrite:\nNo file needs to be rewritten.\n\n## Status:\nPASS\n\n## Send To:\nNoOne\n\n---", "\nRole: You are a senior development and qa engineer, your role is summarize the code running result.\nIf the running result does not include an error, you should explicitly approve the result.\nOn the other hand, if the running result indicates some error, you should point out which part, the development code or the test code, produces the error,\nand give specific instructions on fixing the errors. Here is the code info:\n\n## Development Code File Name\na.py\n## Development Code\n```python\npython -c \"print(1/0)\"\n```\n## Test File Name\n\n## Test Code\n```python\nNone\n```\n## Running Command\npython -c print(1/0)\n## Running Output\nstandard output: \n```text\n\n```\nstandard errors: \n```text\nTraceback (most recent call last):\n File \"\", line 1, in \nZeroDivisionError: division by zero\n\n```\n\nNow you should begin your analysis\n---\n## instruction:\nPlease summarize the cause of the errors and give correction instruction\n## File To Rewrite:\nDetermine the ONE file to rewrite in order to fix the error, for example, xyz.py, or test_xyz.py\n## Status:\nDetermine if all of the code works fine, if so write PASS, else FAIL,\nWRITE ONLY ONE WORD, PASS OR FAIL, IN THIS SECTION\n## Send To:\nPlease write Engineer if the errors are due to problematic development codes, and QaEngineer to problematic test codes, and NoOne if there are no errors,\nWRITE ONLY ONE WORD, Engineer OR QaEngineer OR NoOne, IN THIS SECTION.\n---\nYou should fill in necessary instruction, status, send to, and finally return all content between the --- segment line.\n": "---\nThe error is caused by the development code in file a.py, which attempts to divide by zero. To fix the error, the development code in a.py should be modified to handle the ZeroDivisionError, for example by using a try-except block.\n\nFile To Rewrite:\na.py\n\nStatus:\nFAIL\n\nSend To:\nEngineer\n---", - "\nNOTICE\n1. Role: You are a Development Engineer or QA engineer;\n2. Task: You received this message from another Development Engineer or QA engineer who ran or tested your code. \nBased on the message, first, figure out your own role, i.e. Engineer or QaEngineer,\nthen rewrite the development code or the test code based on your role, the error, and the summary, such that all bugs are fixed and the code performs well.\nAttention: Use '##' to split sections, not '#', and '## ' SHOULD WRITE BEFORE the test case or script and triple quotes.\nThe message is as follows:\n# Legacy Code\n```python\n\nfrom typing import List\nfrom deck import Deck\nfrom card import Card\n\nclass Player:\n \"\"\"\n A class representing a player in the Black Jack game.\n \"\"\"\n\n def __init__(self, name: str):\n \"\"\"\n Initialize a Player object.\n \n Args:\n name (str): The name of the player.\n \"\"\"\n self.name = name\n self.hand: List[Card] = []\n self.score = 0\n\n def draw(self, deck: Deck):\n \"\"\"\n Draw a card from the deck and add it to the player's hand.\n \n Args:\n deck (Deck): The deck of cards.\n \"\"\"\n card = deck.draw_card()\n self.hand.append(card)\n self.calculate_score()\n\n def calculate_score(self) -> int:\n \"\"\"\n Calculate the score of the player's hand.\n \n Returns:\n int: The score of the player's hand.\n \"\"\"\n self.score = sum(card.value for card in self.hand)\n # Handle the case where Ace is counted as 11 and causes the score to exceed 21\n if self.score > 21 and any(card.rank == 'A' for card in self.hand):\n self.score -= 10\n return self.score\n\n```\n---\n# Unit Test Code\n```python\n\nimport unittest\nfrom blackjack_game.player import Player\nfrom blackjack_game.deck import Deck\nfrom blackjack_game.card import Card\n\nclass TestPlayer(unittest.TestCase):\n ## Test the Player's initialization\n def test_player_initialization(self):\n player = Player(\"Test Player\")\n self.assertEqual(player.name, \"Test Player\")\n self.assertEqual(player.hand, [])\n self.assertEqual(player.score, 0)\n\n ## Test the Player's draw method\n def test_player_draw(self):\n deck = Deck()\n player = Player(\"Test Player\")\n player.draw(deck)\n self.assertEqual(len(player.hand), 1)\n self.assertEqual(player.score, player.hand[0].value)\n\n ## Test the Player's calculate_score method\n def test_player_calculate_score(self):\n deck = Deck()\n player = Player(\"Test Player\")\n player.draw(deck)\n player.draw(deck)\n self.assertEqual(player.score, sum(card.value for card in player.hand))\n\n ## Test the Player's calculate_score method with Ace card\n def test_player_calculate_score_with_ace(self):\n deck = Deck()\n player = Player(\"Test Player\")\n player.hand.append(Card('A', 'Hearts', 11))\n player.hand.append(Card('K', 'Hearts', 10))\n player.calculate_score()\n self.assertEqual(player.score, 21)\n\n ## Test the Player's calculate_score method with multiple Aces\n def test_player_calculate_score_with_multiple_aces(self):\n deck = Deck()\n player = Player(\"Test Player\")\n player.hand.append(Card('A', 'Hearts', 11))\n player.hand.append(Card('A', 'Diamonds', 11))\n player.calculate_score()\n self.assertEqual(player.score, 12)\n\nif __name__ == '__main__':\n unittest.main()\n\n\n```\n---\n# Console logs\n```text\n\n```\n---\nNow you should start rewriting the code:\n## file name of the code to rewrite: Write code with triple quote. Do your best to implement THIS IN ONLY ONE FILE.\n": "```python\n## Development Engineer\n\nfrom typing import List\nfrom deck import Deck\nfrom card import Card\n\nclass Player:\n \"\"\"\n A class representing a player in the Black Jack game.\n \"\"\"\n\n def __init__(self, name: str):\n \"\"\"\n Initialize a Player object.\n \n Args:\n name (str): The name of the player.\n \"\"\"\n self.name = name\n self.hand: List[Card] = []\n self.score = 0\n\n def draw(self, deck: Deck):\n \"\"\"\n Draw a card from the deck and add it to the player's hand.\n \n Args:\n deck (Deck): The deck of cards.\n \"\"\"\n card = deck.draw_card()\n self.hand.append(card)\n self.calculate_score()\n\n def calculate_score(self) -> int:\n \"\"\"\n Calculate the score of the player's hand.\n \n Returns:\n int: The score of the player's hand.\n \"\"\"\n self.score = sum(card.value for card in self.hand)\n # Handle the case where Ace is counted as 11 and causes the score to exceed 21\n ace_count = sum(1 for card in self.hand if card.rank == 'A')\n while self.score > 21 and ace_count > 0:\n self.score -= 10\n ace_count -= 1\n return self.score\n```\n\n```python\n## QaEngineer\n\nimport unittest\nfrom player import Player\nfrom deck import Deck\nfrom card import Card\n\nclass TestPlayer(unittest.TestCase):\n ## Test the Player's initialization\n def test_player_initialization(self):\n player = Player(\"Test Player\")\n self.assertEqual(player.name, \"Test Player\")\n self.assertEqual(player.hand, [])\n self.assertEqual(player.score, 0)\n\n ## Test the Player's draw method\n def test_player_draw(self):\n deck = Deck()\n player = Player(\"Test Player\")\n player.draw(deck)\n self.assertEqual(len(player.hand), 1)\n self.assertEqual(player.score, player.hand[0].value)\n\n ## Test the Player's calculate_score method\n def test_player_calculate_score(self):\n deck = Deck()\n player = Player(\"Test Player\")\n player.draw(deck)\n player.draw(deck)\n self.assertEqual(player.score, sum(card.value for card in player.hand))\n\n ## Test the Player's calculate_score method with Ace card\n def test_player_calculate_score_with_ace(self):\n player = Player(\"Test Player\")\n player.hand.append(Card('A', 'Hearts', 11))\n player.hand.append(Card('K', 'Hearts', 10))\n player.calculate_score()\n self.assertEqual(player.score, 21)\n\n ## Test the Player's calculate_score method with multiple Aces\n def test_player_calculate_score_with_multiple_aces(self):\n player = Player(\"Test Player\")\n player.hand.append(Card('A', 'Hearts', 11))\n player.hand.append(Card('A', 'Diamonds', 11))\n player.calculate_score()\n self.assertEqual(player.score, 12)\n\nif __name__ == '__main__':\n unittest.main()\n```" + "\nNOTICE\n1. Role: You are a Development Engineer or QA engineer;\n2. Task: You received this message from another Development Engineer or QA engineer who ran or tested your code. \nBased on the message, first, figure out your own role, i.e. Engineer or QaEngineer,\nthen rewrite the development code or the test code based on your role, the error, and the summary, such that all bugs are fixed and the code performs well.\nAttention: Use '##' to split sections, not '#', and '## ' SHOULD WRITE BEFORE the test case or script and triple quotes.\nThe message is as follows:\n# Legacy Code\n```python\n\nfrom typing import List\nfrom deck import Deck\nfrom card import Card\n\nclass Player:\n \"\"\"\n A class representing a player in the Black Jack game.\n \"\"\"\n\n def __init__(self, name: str):\n \"\"\"\n Initialize a Player object.\n \n Args:\n name (str): The name of the player.\n \"\"\"\n self.name = name\n self.hand: List[Card] = []\n self.score = 0\n\n def draw(self, deck: Deck):\n \"\"\"\n Draw a card from the deck and add it to the player's hand.\n \n Args:\n deck (Deck): The deck of cards.\n \"\"\"\n card = deck.draw_card()\n self.hand.append(card)\n self.calculate_score()\n\n def calculate_score(self) -> int:\n \"\"\"\n Calculate the score of the player's hand.\n \n Returns:\n int: The score of the player's hand.\n \"\"\"\n self.score = sum(card.value for card in self.hand)\n # Handle the case where Ace is counted as 11 and causes the score to exceed 21\n if self.score > 21 and any(card.rank == 'A' for card in self.hand):\n self.score -= 10\n return self.score\n\n```\n---\n# Unit Test Code\n```python\n\nimport unittest\nfrom blackjack_game.player import Player\nfrom blackjack_game.deck import Deck\nfrom blackjack_game.card import Card\n\nclass TestPlayer(unittest.TestCase):\n ## Test the Player's initialization\n def test_player_initialization(self):\n player = Player(\"Test Player\")\n self.assertEqual(player.name, \"Test Player\")\n self.assertEqual(player.hand, [])\n self.assertEqual(player.score, 0)\n\n ## Test the Player's draw method\n def test_player_draw(self):\n deck = Deck()\n player = Player(\"Test Player\")\n player.draw(deck)\n self.assertEqual(len(player.hand), 1)\n self.assertEqual(player.score, player.hand[0].value)\n\n ## Test the Player's calculate_score method\n def test_player_calculate_score(self):\n deck = Deck()\n player = Player(\"Test Player\")\n player.draw(deck)\n player.draw(deck)\n self.assertEqual(player.score, sum(card.value for card in player.hand))\n\n ## Test the Player's calculate_score method with Ace card\n def test_player_calculate_score_with_ace(self):\n deck = Deck()\n player = Player(\"Test Player\")\n player.hand.append(Card('A', 'Hearts', 11))\n player.hand.append(Card('K', 'Hearts', 10))\n player.calculate_score()\n self.assertEqual(player.score, 21)\n\n ## Test the Player's calculate_score method with multiple Aces\n def test_player_calculate_score_with_multiple_aces(self):\n deck = Deck()\n player = Player(\"Test Player\")\n player.hand.append(Card('A', 'Hearts', 11))\n player.hand.append(Card('A', 'Diamonds', 11))\n player.calculate_score()\n self.assertEqual(player.score, 12)\n\nif __name__ == '__main__':\n unittest.main()\n\n\n```\n---\n# Console logs\n```text\n\n```\n---\nNow you should start rewriting the code:\n## file name of the code to rewrite: Write code with triple quote. Do your best to implement THIS IN ONLY ONE FILE.\n": "```python\n## Development Engineer\n\nfrom typing import List\nfrom deck import Deck\nfrom card import Card\n\nclass Player:\n \"\"\"\n A class representing a player in the Black Jack game.\n \"\"\"\n\n def __init__(self, name: str):\n \"\"\"\n Initialize a Player object.\n \n Args:\n name (str): The name of the player.\n \"\"\"\n self.name = name\n self.hand: List[Card] = []\n self.score = 0\n\n def draw(self, deck: Deck):\n \"\"\"\n Draw a card from the deck and add it to the player's hand.\n \n Args:\n deck (Deck): The deck of cards.\n \"\"\"\n card = deck.draw_card()\n self.hand.append(card)\n self.calculate_score()\n\n def calculate_score(self) -> int:\n \"\"\"\n Calculate the score of the player's hand.\n \n Returns:\n int: The score of the player's hand.\n \"\"\"\n self.score = sum(card.value for card in self.hand)\n # Handle the case where Ace is counted as 11 and causes the score to exceed 21\n ace_count = sum(1 for card in self.hand if card.rank == 'A')\n while self.score > 21 and ace_count > 0:\n self.score -= 10\n ace_count -= 1\n return self.score\n```\n\n```python\n## QaEngineer\n\nimport unittest\nfrom player import Player\nfrom deck import Deck\nfrom card import Card\n\nclass TestPlayer(unittest.TestCase):\n ## Test the Player's initialization\n def test_player_initialization(self):\n player = Player(\"Test Player\")\n self.assertEqual(player.name, \"Test Player\")\n self.assertEqual(player.hand, [])\n self.assertEqual(player.score, 0)\n\n ## Test the Player's draw method\n def test_player_draw(self):\n deck = Deck()\n player = Player(\"Test Player\")\n player.draw(deck)\n self.assertEqual(len(player.hand), 1)\n self.assertEqual(player.score, player.hand[0].value)\n\n ## Test the Player's calculate_score method\n def test_player_calculate_score(self):\n deck = Deck()\n player = Player(\"Test Player\")\n player.draw(deck)\n player.draw(deck)\n self.assertEqual(player.score, sum(card.value for card in player.hand))\n\n ## Test the Player's calculate_score method with Ace card\n def test_player_calculate_score_with_ace(self):\n player = Player(\"Test Player\")\n player.hand.append(Card('A', 'Hearts', 11))\n player.hand.append(Card('K', 'Hearts', 10))\n player.calculate_score()\n self.assertEqual(player.score, 21)\n\n ## Test the Player's calculate_score method with multiple Aces\n def test_player_calculate_score_with_multiple_aces(self):\n player = Player(\"Test Player\")\n player.hand.append(Card('A', 'Hearts', 11))\n player.hand.append(Card('A', 'Diamonds', 11))\n player.calculate_score()\n self.assertEqual(player.score, 12)\n\nif __name__ == '__main__':\n unittest.main()\n```", + "\n # Context:\n user: \n## User Requirement\nThis is a titanic passenger survival dataset, your goal is to predict passenger survival outcome. The target column is Survived. Perform data analysis, data preprocessing, feature engineering, and modeling to predict the target. Report accuracy on the eval data. Train data path: 'tests/data/ml_datasets/titanic/split_train.csv', eval data path: 'tests/data/ml_datasets/titanic/split_eval.csv'.\n## Context\n\n## Current Plan\n[]\n## Current Task\n{}\n\n # Task:\n Based on the context, write a plan or modify an existing plan of what you should do to achieve the goal. A plan consists of one to 3 tasks.\n If you are modifying an existing plan, carefully follow the instruction, don't make unnecessary changes. Give the whole plan unless instructed to modify only one task of the plan.\n If you encounter errors on the current task, revise and output the current single task only.\n Output a list of jsons following the format:\n ```json\n [\n {\n \"task_id\": str = \"unique identifier for a task in plan, can be an ordinal\",\n \"dependent_task_ids\": list[str] = \"ids of tasks prerequisite to this task\",\n \"instruction\": \"what you should do in this task, one short phrase or sentence\",\n },\n ...\n ]\n ```\n ": "```json\n[\n {\n \"task_id\": \"1\",\n \"dependent_task_ids\": [],\n \"instruction\": \"Perform exploratory data analysis on the train dataset to understand the features and target variable.\"\n },\n {\n \"task_id\": \"2\",\n \"dependent_task_ids\": [\"1\"],\n \"instruction\": \"Preprocess the data by handling missing values, encoding categorical variables, and scaling features.\"\n },\n {\n \"task_id\": \"3\",\n \"dependent_task_ids\": [\"2\"],\n \"instruction\": \"Conduct feature engineering to create new features that may help improve model performance.\"\n },\n {\n \"task_id\": \"4\",\n \"dependent_task_ids\": [\"3\"],\n \"instruction\": \"Select and train a machine learning model using the processed train dataset.\"\n },\n {\n \"task_id\": \"5\",\n \"dependent_task_ids\": [\"4\"],\n \"instruction\": \"Evaluate the model's accuracy using the eval dataset and report the results.\"\n }\n]\n```", + "[{\"role\": \"user\", \"content\": \"\\nPlease assign a task type to each task in the list below from the given categories:\\nTask 1: Perform exploratory data analysis on the train dataset to understand the features and target variable.\\nTask 2: Preprocess the data by handling missing values, encoding categorical variables, and scaling features.\\nTask 3: Conduct feature engineering to create new features that may help improve model performance.\\nTask 4: Select and train a machine learning model using the processed train dataset.\\nTask 5: Evaluate the model's accuracy using the eval dataset and report the results.\\n\\n## All Task Type:\\n- **eda**: For performing exploratory data analysis\\n- **data_preprocess**: Only for changing value inplace.\\n- **feature_engineering**: Only for creating new columns for input data.\\n- **model_train**: Only for training model.\\n- **model_evaluate**: Only for evaluating model.\\n- **stable_diffusion**: Related to text2image, image2image using stable diffusion model.\\n- **image2webpage**: For converting image into webpage code.\\n- **web_scraping**: For scraping data from web pages.\\n- **other**: Any tools not in the defined categories\\n\"}]": { + "task_type": [ + "eda", + "data_preprocess", + "feature_engineering", + "model_train", + "model_evaluate" + ] + }, + "[{\"role\": \"user\", \"content\": \"\\n# Background\\nAs a data scientist, you need to help user to achieve their goal [This is a titanic passenger survival dataset, your goal is to predict passenger survival outcome. The target column is Survived. Perform data analysis, data preprocessing, feature engineering, and modeling to predict the target. Report accuracy on the eval data. Train data path: 'tests/data/ml_datasets/titanic/split_train.csv', eval data path: 'tests/data/ml_datasets/titanic/split_eval.csv'.] step-by-step in an continuous Jupyter notebook.\\n\\n## Done Tasks\\n```python\\n\\n```end\\n\\n## Current Task\\nPerform exploratory data analysis on the train dataset to understand the features and target variable.\\n\\n# Latest Data Info\\nLatest data info after previous tasks:\\n\\n\\n# Task\\nWrite complete code for 'Current Task'. And avoid duplicating code from 'Done Tasks', such as repeated import of packages, reading data, etc.\\nSpecifically, \\n\\n# Code Steps:\\nStrictly follow steps below when you writing code if it's convenient.\\n\\n\\n# Output Example:\\nwhen current task is \\\"train a lightgbm model on training data\\\", and their are two steps in 'Code Steps', the code be like:\\n```python\\n# Step 1: check data type and convert to numeric\\nojb_cols = train.select_dtypes(include='object').columns.tolist()\\n\\nfor col in obj_cols:\\n encoder = LabelEncoder()\\n train[col] = encoder.fit_transform(train[col].unique().tolist() + ['unknown'])\\n test[col] = test[col].apply(lambda x: x if x in encoder.classes_ else 'unknown')\\n test[col] = encoder.transform(test[col])\\n\\n# Step 2: train lightgbm model\\nmodel = LGBMClassifier()\\nmodel.fit(train, y_train)\\n```end\\n\\n# Constraints:\\n- Ensure the output new code is executable in the same Jupyter notebook with previous tasks code have been executed.\\n- The output code should contain all steps implemented in 'Code Steps'.\\n\"}]": { + "code": "# Step 1: Load the training dataset\nimport pandas as pd\ntrain_data = pd.read_csv('tests/data/ml_datasets/titanic/split_train.csv')\n\n# Step 2: Display the first few rows of the dataset\ndisplay(train_data.head())\n\n# Step 3: Get a summary of the dataset\ntrain_data.info()\n\n# Step 4: Describe the numerical features\ntrain_data.describe()\n\n# Step 5: Describe the categorical features\ntrain_data.describe(include=['O'])\n\n# Step 6: Check for missing values\nmissing_values = train_data.isnull().sum()\nmissing_values[missing_values > 0]\n\n# Step 7: Visualize the distribution of the target variable 'Survived'\nimport seaborn as sns\nimport matplotlib.pyplot as plt\nsns.countplot(x='Survived', data=train_data)\nplt.title('Distribution of Survival')\nplt.show()\n\n# Step 8: Explore the relationship between features and the target variable\n# Visualize the survival rate by sex\nsns.barplot(x='Sex', y='Survived', data=train_data)\nplt.title('Survival Rate by Sex')\nplt.show()\n\n# Visualize the survival rate by class\nsns.barplot(x='Pclass', y='Survived', data=train_data)\nplt.title('Survival Rate by Class')\nplt.show()\n\n# Visualize the survival rate by embarkation port\nsns.barplot(x='Embarked', y='Survived', data=train_data)\nplt.title('Survival Rate by Embarkation Port')\nplt.show()" + }, + "[{\"role\": \"user\", \"content\": \"\\n# Background\\nKeep dataset column information updated before model train.\\n## Done Tasks\\n```python\\nimport pandas as pd\\ntrain_data = pd.read_csv('tests/data/ml_datasets/titanic/split_train.csv')\\ndisplay(train_data.head())\\ntrain_data.info()\\ntrain_data.describe()\\ntrain_data.describe(include=['O'])\\nmissing_values = train_data.isnull().sum()\\nmissing_values[missing_values > 0]\\nimport seaborn as sns\\nimport matplotlib.pyplot as plt\\nsns.countplot(x='Survived', data=train_data)\\nplt.title('Distribution of Survival')\\nplt.show()\\nsns.barplot(x='Sex', y='Survived', data=train_data)\\nplt.title('Survival Rate by Sex')\\nplt.show()\\nsns.barplot(x='Pclass', y='Survived', data=train_data)\\nplt.title('Survival Rate by Class')\\nplt.show()\\nsns.barplot(x='Embarked', y='Survived', data=train_data)\\nplt.title('Survival Rate by Embarkation Port')\\nplt.show()\\n```end\\n\\n# Task\\nUpdate and print the dataset's column information only if the train or test data has changed. Use the following code:\\n```python\\nfrom metagpt.tools.libs.data_preprocess import get_column_info\\n\\ncolumn_info = get_column_info(df)\\nprint(\\\"column_info\\\")\\nprint(column_info)\\n```end\\n\\n# Constraints:\\n- Use the DataFrame variable from 'Done Tasks' in place of df.\\n- Import `get_column_info` only if it's not already imported.\\n\"}]": { + "code": "from metagpt.tools.libs.data_preprocess import get_column_info\n\ncolumn_info = get_column_info(train_data)\nprint(\"column_info\")\nprint(column_info)" + }, + "[{\"role\": \"user\", \"content\": \"\\n## User Requirement:\\nPreprocess the data by handling missing values, encoding categorical variables, and scaling features.\\n\\n## Task\\nRecommend up to five tools from 'Available Tools' that can help solve the 'User Requirement'. \\nThis is a detailed code steps for current task. You can refer to it when recommending tools.\\n\\n\\n## Available Tools:\\n{'FillMissingValue': 'Completing missing values with simple strategies'}\\n\\n## Tool Selection and Instructions:\\n- Select tools most relevant to completing the 'User Requirement'.\\n- If you believe that no tools are suitable, indicate with an empty list.\\n- Only list the names of the tools, not the full schema of each tool.\\n- Ensure selected tools are listed in 'Available Tools'.\\n\"}]": { + "recommend_tools": [ + "FillMissingValue" + ] + }, + "[{\"role\": \"user\", \"content\": \"\\n# Background\\nAs a data scientist, you need to help user to achieve their goal [This is a titanic passenger survival dataset, your goal is to predict passenger survival outcome. The target column is Survived. Perform data analysis, data preprocessing, feature engineering, and modeling to predict the target. Report accuracy on the eval data. Train data path: 'tests/data/ml_datasets/titanic/split_train.csv', eval data path: 'tests/data/ml_datasets/titanic/split_eval.csv'.] step-by-step in an continuous Jupyter notebook.\\n\\n## Done Tasks\\n```python\\nimport pandas as pd\\ntrain_data = pd.read_csv('tests/data/ml_datasets/titanic/split_train.csv')\\ndisplay(train_data.head())\\ntrain_data.info()\\ntrain_data.describe()\\ntrain_data.describe(include=['O'])\\nmissing_values = train_data.isnull().sum()\\nmissing_values[missing_values > 0]\\nimport seaborn as sns\\nimport matplotlib.pyplot as plt\\nsns.countplot(x='Survived', data=train_data)\\nplt.title('Distribution of Survival')\\nplt.show()\\nsns.barplot(x='Sex', y='Survived', data=train_data)\\nplt.title('Survival Rate by Sex')\\nplt.show()\\nsns.barplot(x='Pclass', y='Survived', data=train_data)\\nplt.title('Survival Rate by Class')\\nplt.show()\\nsns.barplot(x='Embarked', y='Survived', data=train_data)\\nplt.title('Survival Rate by Embarkation Port')\\nplt.show()\\n```end\\n\\n## Current Task\\nPreprocess the data by handling missing values, encoding categorical variables, and scaling features.\\n\\n# Latest Data Info\\nLatest data info after previous tasks:\\ncolumn_info\\n{'Category': ['Name', 'Sex', 'Ticket', 'Cabin', 'Embarked'], 'Numeric': ['PassengerId', 'Survived', 'Pclass', 'Age', 'SibSp', 'Parch', 'Fare'], 'Datetime': [], 'Others': []}\\n\\n\\n# Task\\nWrite complete code for 'Current Task'. And avoid duplicating code from 'Done Tasks', such as repeated import of packages, reading data, etc.\\nSpecifically, \\nThe current task is about data preprocessing, please note the following:\\n- Monitor data types per column, applying appropriate methods.\\n- Ensure operations are on existing dataset columns.\\n- Avoid writing processed data to files.\\n- Avoid any change to label column, such as standardization, etc.\\n- Prefer alternatives to one-hot encoding for categorical data.\\n- Only encode or scale necessary columns to allow for potential feature-specific engineering tasks (like time_extract, binning, extraction, etc.) later.\\n- Each step do data preprocessing to train, must do same for test separately at the same time.\\n\\n\\n# Code Steps:\\nStrictly follow steps below when you writing code if it's convenient.\\n\\n\\n# Capabilities\\n- You can utilize pre-defined tools in any code lines from 'Available Tools' in the form of Python Class.\\n- You can freely combine the use of any other public packages, like sklearn, numpy, pandas, etc..\\n\\n# Available Tools:\\nEach Class tool is described in JSON format. When you call a tool, import the tool from its path first.\\n{'FillMissingValue': {'type': 'class', 'description': 'Completing missing values with simple strategies', 'methods': {'__init__': {'description': 'Initialize self.', 'parameters': {'properties': {'features': {'type': 'list', 'description': 'columns to be processed'}, 'strategy': {'type': 'str', 'description': 'the imputation strategy, notice mean/median can only be used for numeric features', 'default': 'mean', 'enum': ['mean', 'median', 'most_frequent', 'constant']}, 'fill_value': {'type': 'int', 'description': 'fill_value is used to replace all occurrences of missing_values', 'default': None}}, 'required': ['features']}}, 'fit': {'description': 'Fit the FillMissingValue model.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}}, 'transform': {'description': 'Transform the input DataFrame with the fitted model.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': {'df': {'type': 'DataFrame', 'description': 'The transformed DataFrame.'}}}, 'fit_transform': {'description': 'Fit and transform the input DataFrame.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': {'df': {'type': 'DataFrame', 'description': 'The transformed DataFrame.'}}}}, 'tool_path': 'metagpt/tools/libs/data_preprocess.py'}}\\n\\n# Output Example:\\nwhen current task is \\\"do data preprocess, like fill missing value, handle outliers, etc.\\\", and their are two steps in 'Code Steps', the code be like:\\n```python\\n# Step 1: fill missing value\\n# Tools used: ['FillMissingValue']\\nfrom metagpt.tools.libs.data_preprocess import FillMissingValue\\n\\ntrain_processed = train.copy()\\ntest_processed = test.copy()\\nnum_cols = train_processed.select_dtypes(include='number').columns.tolist()\\nif 'label' in num_cols:\\n num_cols.remove('label')\\nfill_missing_value = FillMissingValue(features=num_cols, strategy='mean')\\nfill_missing_value.fit(train_processed)\\ntrain_processed = fill_missing_value.transform(train_processed)\\ntest_processed = fill_missing_value.transform(test_processed)\\n\\n# Step 2: handle outliers\\nfor col in num_cols:\\n low, high = train_processed[col].quantile([0.01, 0.99])\\n train_processed[col] = train_processed[col].clip(low, high)\\n test_processed[col] = test_processed[col].clip(low, high)\\n```end\\n\\n# Constraints:\\n- Ensure the output new code is executable in the same Jupyter notebook with previous tasks code have been executed.\\n- Always prioritize using pre-defined tools for the same functionality.\\n- Always copy the DataFrame before processing it and use the copy to process.\\n- The output code should contain all steps implemented correctly in 'Code Steps'.\\n\"}]": { + "code": "# Step 1: Handle missing values\nfrom metagpt.tools.libs.data_preprocess import FillMissingValue\n\n# Create copies of the datasets to avoid changing the original data\ntrain_processed = train_data.copy()\neval_data = pd.read_csv('tests/data/ml_datasets/titanic/split_eval.csv')\n\n# Fill missing values for 'Age' with median and 'Embarked' with most frequent value\nfill_missing_age = FillMissingValue(features=['Age'], strategy='median')\ntrain_processed = fill_missing_age.fit_transform(train_processed)\neval_processed = fill_missing_age.transform(eval_data)\n\nfill_missing_embarked = FillMissingValue(features=['Embarked'], strategy='most_frequent')\ntrain_processed = fill_missing_embarked.fit_transform(train_processed)\neval_processed = fill_missing_embarked.transform(eval_data)\n\n# Since 'Cabin' has a lot of missing values, we'll create a binary feature indicating whether cabin info is missing\ntrain_processed['Cabin_Ind'] = train_processed['Cabin'].isnull().astype(int)\neval_processed['Cabin_Ind'] = eval_processed['Cabin'].isnull().astype(int)\n\n# Step 2: Encode categorical variables\nfrom sklearn.preprocessing import LabelEncoder\n\nlabel_encoder = LabelEncoder()\n\n# Encode 'Sex'\ntrain_processed['Sex'] = label_encoder.fit_transform(train_processed['Sex'])\neval_processed['Sex'] = label_encoder.transform(eval_processed['Sex'])\n\n# Encode 'Embarked'\ntrain_processed['Embarked'] = label_encoder.fit_transform(train_processed['Embarked'])\neval_processed['Embarked'] = label_encoder.transform(eval_processed['Embarked'])\n\n# Step 3: Scale features\nfrom sklearn.preprocessing import StandardScaler\n\nscaler = StandardScaler()\n\n# Scale 'Age' and 'Fare'\nfor feature in ['Age', 'Fare']:\n train_processed[[feature]] = scaler.fit_transform(train_processed[[feature]])\n eval_processed[[feature]] = scaler.transform(eval_processed[[feature]])\n\n# Display the first few rows of the processed train dataset\ndisplay(train_processed.head())" + }, + "[{\"role\": \"user\", \"content\": \"\\n# Background\\nKeep dataset column information updated before model train.\\n## Done Tasks\\n```python\\nimport pandas as pd\\ntrain_data = pd.read_csv('tests/data/ml_datasets/titanic/split_train.csv')\\ndisplay(train_data.head())\\ntrain_data.info()\\ntrain_data.describe()\\ntrain_data.describe(include=['O'])\\nmissing_values = train_data.isnull().sum()\\nmissing_values[missing_values > 0]\\nimport seaborn as sns\\nimport matplotlib.pyplot as plt\\nsns.countplot(x='Survived', data=train_data)\\nplt.title('Distribution of Survival')\\nplt.show()\\nsns.barplot(x='Sex', y='Survived', data=train_data)\\nplt.title('Survival Rate by Sex')\\nplt.show()\\nsns.barplot(x='Pclass', y='Survived', data=train_data)\\nplt.title('Survival Rate by Class')\\nplt.show()\\nsns.barplot(x='Embarked', y='Survived', data=train_data)\\nplt.title('Survival Rate by Embarkation Port')\\nplt.show()\\n\\nfrom metagpt.tools.libs.data_preprocess import FillMissingValue\\ntrain_processed = train_data.copy()\\neval_data = pd.read_csv('tests/data/ml_datasets/titanic/split_eval.csv')\\nfill_missing_age = FillMissingValue(features=['Age'], strategy='median')\\ntrain_processed = fill_missing_age.fit_transform(train_processed)\\neval_processed = fill_missing_age.transform(eval_data)\\nfill_missing_embarked = FillMissingValue(features=['Embarked'], strategy='most_frequent')\\ntrain_processed = fill_missing_embarked.fit_transform(train_processed)\\neval_processed = fill_missing_embarked.transform(eval_data)\\ntrain_processed['Cabin_Ind'] = train_processed['Cabin'].isnull().astype(int)\\neval_processed['Cabin_Ind'] = eval_processed['Cabin'].isnull().astype(int)\\nfrom sklearn.preprocessing import LabelEncoder\\nlabel_encoder = LabelEncoder()\\ntrain_processed['Sex'] = label_encoder.fit_transform(train_processed['Sex'])\\neval_processed['Sex'] = label_encoder.transform(eval_processed['Sex'])\\ntrain_processed['Embarked'] = label_encoder.fit_transform(train_processed['Embarked'])\\neval_processed['Embarked'] = label_encoder.transform(eval_processed['Embarked'])\\nfrom sklearn.preprocessing import StandardScaler\\nscaler = StandardScaler()\\nfor feature in ['Age', 'Fare']:\\n train_processed[[feature]] = scaler.fit_transform(train_processed[[feature]])\\n eval_processed[[feature]] = scaler.transform(eval_processed[[feature]])\\ndisplay(train_processed.head())\\n```end\\n\\n# Task\\nUpdate and print the dataset's column information only if the train or test data has changed. Use the following code:\\n```python\\nfrom metagpt.tools.libs.data_preprocess import get_column_info\\n\\ncolumn_info = get_column_info(df)\\nprint(\\\"column_info\\\")\\nprint(column_info)\\n```end\\n\\n# Constraints:\\n- Use the DataFrame variable from 'Done Tasks' in place of df.\\n- Import `get_column_info` only if it's not already imported.\\n\"}]": { + "code": "from metagpt.tools.libs.data_preprocess import get_column_info\ncolumn_info = get_column_info(train_processed)\nprint(\"column_info\")\nprint(column_info)" + }, + "[{\"role\": \"user\", \"content\": \"\\n## User Requirement:\\nConduct feature engineering to create new features that may help improve model performance.\\n\\n## Task\\nRecommend up to five tools from 'Available Tools' that can help solve the 'User Requirement'. \\nThis is a detailed code steps for current task. You can refer to it when recommending tools.\\n\\n\\n## Available Tools:\\n{'CatCross': 'Add pairwise crossed features and convert them to numerical features.'}\\n\\n## Tool Selection and Instructions:\\n- Select tools most relevant to completing the 'User Requirement'.\\n- If you believe that no tools are suitable, indicate with an empty list.\\n- Only list the names of the tools, not the full schema of each tool.\\n- Ensure selected tools are listed in 'Available Tools'.\\n\"}]": { + "recommend_tools": [ + "CatCross" + ] + }, + "[{\"role\": \"user\", \"content\": \"\\n# Background\\nAs a data scientist, you need to help user to achieve their goal [This is a titanic passenger survival dataset, your goal is to predict passenger survival outcome. The target column is Survived. Perform data analysis, data preprocessing, feature engineering, and modeling to predict the target. Report accuracy on the eval data. Train data path: 'tests/data/ml_datasets/titanic/split_train.csv', eval data path: 'tests/data/ml_datasets/titanic/split_eval.csv'.] step-by-step in an continuous Jupyter notebook.\\n\\n## Done Tasks\\n```python\\nimport pandas as pd\\ntrain_data = pd.read_csv('tests/data/ml_datasets/titanic/split_train.csv')\\ndisplay(train_data.head())\\ntrain_data.info()\\ntrain_data.describe()\\ntrain_data.describe(include=['O'])\\nmissing_values = train_data.isnull().sum()\\nmissing_values[missing_values > 0]\\nimport seaborn as sns\\nimport matplotlib.pyplot as plt\\nsns.countplot(x='Survived', data=train_data)\\nplt.title('Distribution of Survival')\\nplt.show()\\nsns.barplot(x='Sex', y='Survived', data=train_data)\\nplt.title('Survival Rate by Sex')\\nplt.show()\\nsns.barplot(x='Pclass', y='Survived', data=train_data)\\nplt.title('Survival Rate by Class')\\nplt.show()\\nsns.barplot(x='Embarked', y='Survived', data=train_data)\\nplt.title('Survival Rate by Embarkation Port')\\nplt.show()\\n\\nfrom metagpt.tools.libs.data_preprocess import FillMissingValue\\ntrain_processed = train_data.copy()\\neval_data = pd.read_csv('tests/data/ml_datasets/titanic/split_eval.csv')\\nfill_missing_age = FillMissingValue(features=['Age'], strategy='median')\\ntrain_processed = fill_missing_age.fit_transform(train_processed)\\neval_processed = fill_missing_age.transform(eval_data)\\nfill_missing_embarked = FillMissingValue(features=['Embarked'], strategy='most_frequent')\\ntrain_processed = fill_missing_embarked.fit_transform(train_processed)\\neval_processed = fill_missing_embarked.transform(eval_data)\\ntrain_processed['Cabin_Ind'] = train_processed['Cabin'].isnull().astype(int)\\neval_processed['Cabin_Ind'] = eval_processed['Cabin'].isnull().astype(int)\\nfrom sklearn.preprocessing import LabelEncoder\\nlabel_encoder = LabelEncoder()\\ntrain_processed['Sex'] = label_encoder.fit_transform(train_processed['Sex'])\\neval_processed['Sex'] = label_encoder.transform(eval_processed['Sex'])\\ntrain_processed['Embarked'] = label_encoder.fit_transform(train_processed['Embarked'])\\neval_processed['Embarked'] = label_encoder.transform(eval_processed['Embarked'])\\nfrom sklearn.preprocessing import StandardScaler\\nscaler = StandardScaler()\\nfor feature in ['Age', 'Fare']:\\n train_processed[[feature]] = scaler.fit_transform(train_processed[[feature]])\\n eval_processed[[feature]] = scaler.transform(eval_processed[[feature]])\\ndisplay(train_processed.head())\\n```end\\n\\n## Current Task\\nConduct feature engineering to create new features that may help improve model performance.\\n\\n# Latest Data Info\\nLatest data info after previous tasks:\\ncolumn_info\\n{'Category': ['Name', 'Ticket', 'Cabin'], 'Numeric': ['PassengerId', 'Survived', 'Pclass', 'Sex', 'Age', 'SibSp', 'Parch', 'Fare', 'Embarked', 'Cabin_Ind'], 'Datetime': [], 'Others': []}\\n\\n\\n# Task\\nWrite complete code for 'Current Task'. And avoid duplicating code from 'Done Tasks', such as repeated import of packages, reading data, etc.\\nSpecifically, \\nThe current task is about feature engineering. when performing it, please adhere to the following principles:\\n- Generate as diverse features as possible to improve the model's performance step-by-step. \\n- If potential impactful features are not included in 'Code Steps', add new steps to generate them.\\n- Avoid creating redundant or excessively numerous features in one step.\\n- Exclude ID columns from feature generation and remove them.\\n- Each step do feature engineering to train, must do same for test separately at the same time.\\n- Avoid using the label column to create features, except for cat encoding.\\n- Use the data from previous task result if exist, do not mock or reload data yourself.\\n\\n\\n# Code Steps:\\nStrictly follow steps below when you writing code if it's convenient.\\n\\n\\n# Capabilities\\n- You can utilize pre-defined tools in any code lines from 'Available Tools' in the form of Python Class.\\n- You can freely combine the use of any other public packages, like sklearn, numpy, pandas, etc..\\n\\n# Available Tools:\\nEach Class tool is described in JSON format. When you call a tool, import the tool from its path first.\\n{'CatCross': {'type': 'class', 'description': 'Add pairwise crossed features and convert them to numerical features.', 'methods': {'__init__': {'description': 'Initialize self.', 'parameters': {'properties': {'cols': {'type': 'list', 'description': 'Columns to be pairwise crossed, at least 2 columns.'}, 'max_cat_num': {'type': 'int', 'description': 'Maximum unique categories per crossed feature.', 'default': 100}}}, 'required': ['cols']}, 'fit': {'description': 'Fit the CatCross model.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}}, 'transform': {'description': 'Transform the input DataFrame with the fitted model.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': {'df': {'type': 'DataFrame', 'description': 'The transformed DataFrame.'}}}, 'fit_transform': {'description': 'Fit and transform the input DataFrame.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': {'df': {'type': 'DataFrame', 'description': 'The transformed DataFrame.'}}}}, 'tool_path': 'metagpt/tools/libs/feature_engineering.py'}}\\n\\n# Output Example:\\nwhen current task is \\\"do data preprocess, like fill missing value, handle outliers, etc.\\\", and their are two steps in 'Code Steps', the code be like:\\n```python\\n# Step 1: fill missing value\\n# Tools used: ['FillMissingValue']\\nfrom metagpt.tools.libs.data_preprocess import FillMissingValue\\n\\ntrain_processed = train.copy()\\ntest_processed = test.copy()\\nnum_cols = train_processed.select_dtypes(include='number').columns.tolist()\\nif 'label' in num_cols:\\n num_cols.remove('label')\\nfill_missing_value = FillMissingValue(features=num_cols, strategy='mean')\\nfill_missing_value.fit(train_processed)\\ntrain_processed = fill_missing_value.transform(train_processed)\\ntest_processed = fill_missing_value.transform(test_processed)\\n\\n# Step 2: handle outliers\\nfor col in num_cols:\\n low, high = train_processed[col].quantile([0.01, 0.99])\\n train_processed[col] = train_processed[col].clip(low, high)\\n test_processed[col] = test_processed[col].clip(low, high)\\n```end\\n\\n# Constraints:\\n- Ensure the output new code is executable in the same Jupyter notebook with previous tasks code have been executed.\\n- Always prioritize using pre-defined tools for the same functionality.\\n- Always copy the DataFrame before processing it and use the copy to process.\\n- The output code should contain all steps implemented correctly in 'Code Steps'.\\n\"}]": { + "code": "# Step 1: Create a feature for the number of family members on board\ntrain_processed['Family_Size'] = train_processed['SibSp'] + train_processed['Parch']\neval_processed['Family_Size'] = eval_processed['SibSp'] + eval_processed['Parch']\n\n# Step 2: Create a feature to indicate whether a passenger was alone or not\ntrain_processed['Is_Alone'] = (train_processed['Family_Size'] == 0).astype(int)\neval_processed['Is_Alone'] = (eval_processed['Family_Size'] == 0).astype(int)\n\n# Step 3: Create a feature for the title extracted from the passenger's name\ntrain_processed['Title'] = train_processed['Name'].str.extract(' ([A-Za-z]+)\\.', expand=False)\neval_processed['Title'] = eval_processed['Name'].str.extract(' ([A-Za-z]+)\\.', expand=False)\n\n# Step 4: Create a feature for the deck extracted from the cabin number\ntrain_processed['Deck'] = train_processed['Cabin'].str[0]\neval_processed['Deck'] = eval_processed['Cabin'].str[0]\n\n# Step 5: Create a feature for the ticket class and sex cross\n# Tools used: ['CatCross']\nfrom metagpt.tools.libs.feature_engineering import CatCross\n\n# Initialize CatCross with the desired columns\ncat_cross = CatCross(cols=['Pclass', 'Sex'])\n\n# Fit and transform the train data\ntrain_processed = cat_cross.fit_transform(train_processed)\n\n# Transform the eval data\neval_processed = cat_cross.transform(eval_processed)\n\n# Remove unnecessary columns\nfor df in [train_processed, eval_processed]:\n df.drop(['PassengerId', 'Name', 'Ticket', 'Cabin'], axis=1, inplace=True)" + }, + "[{\"role\": \"user\", \"content\": \"\\n# Background\\nKeep dataset column information updated before model train.\\n## Done Tasks\\n```python\\nimport pandas as pd\\ntrain_data = pd.read_csv('tests/data/ml_datasets/titanic/split_train.csv')\\ndisplay(train_data.head())\\ntrain_data.info()\\ntrain_data.describe()\\ntrain_data.describe(include=['O'])\\nmissing_values = train_data.isnull().sum()\\nmissing_values[missing_values > 0]\\nimport seaborn as sns\\nimport matplotlib.pyplot as plt\\nsns.countplot(x='Survived', data=train_data)\\nplt.title('Distribution of Survival')\\nplt.show()\\nsns.barplot(x='Sex', y='Survived', data=train_data)\\nplt.title('Survival Rate by Sex')\\nplt.show()\\nsns.barplot(x='Pclass', y='Survived', data=train_data)\\nplt.title('Survival Rate by Class')\\nplt.show()\\nsns.barplot(x='Embarked', y='Survived', data=train_data)\\nplt.title('Survival Rate by Embarkation Port')\\nplt.show()\\n\\nfrom metagpt.tools.libs.data_preprocess import FillMissingValue\\ntrain_processed = train_data.copy()\\neval_data = pd.read_csv('tests/data/ml_datasets/titanic/split_eval.csv')\\nfill_missing_age = FillMissingValue(features=['Age'], strategy='median')\\ntrain_processed = fill_missing_age.fit_transform(train_processed)\\neval_processed = fill_missing_age.transform(eval_data)\\nfill_missing_embarked = FillMissingValue(features=['Embarked'], strategy='most_frequent')\\ntrain_processed = fill_missing_embarked.fit_transform(train_processed)\\neval_processed = fill_missing_embarked.transform(eval_data)\\ntrain_processed['Cabin_Ind'] = train_processed['Cabin'].isnull().astype(int)\\neval_processed['Cabin_Ind'] = eval_processed['Cabin'].isnull().astype(int)\\nfrom sklearn.preprocessing import LabelEncoder\\nlabel_encoder = LabelEncoder()\\ntrain_processed['Sex'] = label_encoder.fit_transform(train_processed['Sex'])\\neval_processed['Sex'] = label_encoder.transform(eval_processed['Sex'])\\ntrain_processed['Embarked'] = label_encoder.fit_transform(train_processed['Embarked'])\\neval_processed['Embarked'] = label_encoder.transform(eval_processed['Embarked'])\\nfrom sklearn.preprocessing import StandardScaler\\nscaler = StandardScaler()\\nfor feature in ['Age', 'Fare']:\\n train_processed[[feature]] = scaler.fit_transform(train_processed[[feature]])\\n eval_processed[[feature]] = scaler.transform(eval_processed[[feature]])\\ndisplay(train_processed.head())\\n\\ntrain_processed['Family_Size'] = train_processed['SibSp'] + train_processed['Parch']\\neval_processed['Family_Size'] = eval_processed['SibSp'] + eval_processed['Parch']\\ntrain_processed['Is_Alone'] = (train_processed['Family_Size'] == 0).astype(int)\\neval_processed['Is_Alone'] = (eval_processed['Family_Size'] == 0).astype(int)\\ntrain_processed['Title'] = train_processed['Name'].str.extract(' ([A-Za-z]+)\\\\.', expand=False)\\neval_processed['Title'] = eval_processed['Name'].str.extract(' ([A-Za-z]+)\\\\.', expand=False)\\ntrain_processed['Deck'] = train_processed['Cabin'].str[0]\\neval_processed['Deck'] = eval_processed['Cabin'].str[0]\\nfrom metagpt.tools.libs.feature_engineering import CatCross\\ncat_cross = CatCross(cols=['Pclass', 'Sex'])\\ntrain_processed = cat_cross.fit_transform(train_processed)\\neval_processed = cat_cross.transform(eval_processed)\\nfor df in [train_processed, eval_processed]:\\n df.drop(['PassengerId', 'Name', 'Ticket', 'Cabin'], axis=1, inplace=True)\\n```end\\n\\n# Task\\nUpdate and print the dataset's column information only if the train or test data has changed. Use the following code:\\n```python\\nfrom metagpt.tools.libs.data_preprocess import get_column_info\\n\\ncolumn_info = get_column_info(df)\\nprint(\\\"column_info\\\")\\nprint(column_info)\\n```end\\n\\n# Constraints:\\n- Use the DataFrame variable from 'Done Tasks' in place of df.\\n- Import `get_column_info` only if it's not already imported.\\n\"}]": { + "code": "from metagpt.tools.libs.data_preprocess import get_column_info\ncolumn_info = get_column_info(train_processed)\nprint(\"column_info\")\nprint(column_info)" + }, + "[{\"role\": \"user\", \"content\": \"\\n# Background\\nAs a data scientist, you need to help user to achieve their goal [This is a titanic passenger survival dataset, your goal is to predict passenger survival outcome. The target column is Survived. Perform data analysis, data preprocessing, feature engineering, and modeling to predict the target. Report accuracy on the eval data. Train data path: 'tests/data/ml_datasets/titanic/split_train.csv', eval data path: 'tests/data/ml_datasets/titanic/split_eval.csv'.] step-by-step in an continuous Jupyter notebook.\\n\\n## Done Tasks\\n```python\\nimport pandas as pd\\ntrain_data = pd.read_csv('tests/data/ml_datasets/titanic/split_train.csv')\\ndisplay(train_data.head())\\ntrain_data.info()\\ntrain_data.describe()\\ntrain_data.describe(include=['O'])\\nmissing_values = train_data.isnull().sum()\\nmissing_values[missing_values > 0]\\nimport seaborn as sns\\nimport matplotlib.pyplot as plt\\nsns.countplot(x='Survived', data=train_data)\\nplt.title('Distribution of Survival')\\nplt.show()\\nsns.barplot(x='Sex', y='Survived', data=train_data)\\nplt.title('Survival Rate by Sex')\\nplt.show()\\nsns.barplot(x='Pclass', y='Survived', data=train_data)\\nplt.title('Survival Rate by Class')\\nplt.show()\\nsns.barplot(x='Embarked', y='Survived', data=train_data)\\nplt.title('Survival Rate by Embarkation Port')\\nplt.show()\\n\\nfrom metagpt.tools.libs.data_preprocess import FillMissingValue\\ntrain_processed = train_data.copy()\\neval_data = pd.read_csv('tests/data/ml_datasets/titanic/split_eval.csv')\\nfill_missing_age = FillMissingValue(features=['Age'], strategy='median')\\ntrain_processed = fill_missing_age.fit_transform(train_processed)\\neval_processed = fill_missing_age.transform(eval_data)\\nfill_missing_embarked = FillMissingValue(features=['Embarked'], strategy='most_frequent')\\ntrain_processed = fill_missing_embarked.fit_transform(train_processed)\\neval_processed = fill_missing_embarked.transform(eval_data)\\ntrain_processed['Cabin_Ind'] = train_processed['Cabin'].isnull().astype(int)\\neval_processed['Cabin_Ind'] = eval_processed['Cabin'].isnull().astype(int)\\nfrom sklearn.preprocessing import LabelEncoder\\nlabel_encoder = LabelEncoder()\\ntrain_processed['Sex'] = label_encoder.fit_transform(train_processed['Sex'])\\neval_processed['Sex'] = label_encoder.transform(eval_processed['Sex'])\\ntrain_processed['Embarked'] = label_encoder.fit_transform(train_processed['Embarked'])\\neval_processed['Embarked'] = label_encoder.transform(eval_processed['Embarked'])\\nfrom sklearn.preprocessing import StandardScaler\\nscaler = StandardScaler()\\nfor feature in ['Age', 'Fare']:\\n train_processed[[feature]] = scaler.fit_transform(train_processed[[feature]])\\n eval_processed[[feature]] = scaler.transform(eval_processed[[feature]])\\ndisplay(train_processed.head())\\n\\ntrain_processed['Family_Size'] = train_processed['SibSp'] + train_processed['Parch']\\neval_processed['Family_Size'] = eval_processed['SibSp'] + eval_processed['Parch']\\ntrain_processed['Is_Alone'] = (train_processed['Family_Size'] == 0).astype(int)\\neval_processed['Is_Alone'] = (eval_processed['Family_Size'] == 0).astype(int)\\ntrain_processed['Title'] = train_processed['Name'].str.extract(' ([A-Za-z]+)\\\\.', expand=False)\\neval_processed['Title'] = eval_processed['Name'].str.extract(' ([A-Za-z]+)\\\\.', expand=False)\\ntrain_processed['Deck'] = train_processed['Cabin'].str[0]\\neval_processed['Deck'] = eval_processed['Cabin'].str[0]\\nfrom metagpt.tools.libs.feature_engineering import CatCross\\ncat_cross = CatCross(cols=['Pclass', 'Sex'])\\ntrain_processed = cat_cross.fit_transform(train_processed)\\neval_processed = cat_cross.transform(eval_processed)\\nfor df in [train_processed, eval_processed]:\\n df.drop(['PassengerId', 'Name', 'Ticket', 'Cabin'], axis=1, inplace=True)\\n```end\\n\\n## Current Task\\nSelect and train a machine learning model using the processed train dataset.\\n\\n# Latest Data Info\\nLatest data info after previous tasks:\\ncolumn_info\\n{'Category': ['Title', 'Deck'], 'Numeric': ['Survived', 'Pclass', 'Sex', 'Age', 'SibSp', 'Parch', 'Fare', 'Embarked', 'Cabin_Ind', 'Family_Size', 'Is_Alone', 'Pclass_Sex'], 'Datetime': [], 'Others': []}\\n\\n\\n# Task\\nWrite complete code for 'Current Task'. And avoid duplicating code from 'Done Tasks', such as repeated import of packages, reading data, etc.\\nSpecifically, \\nThe current task is about training a model, please ensure high performance:\\n- Keep in mind that your user prioritizes results and is highly focused on model performance. So, when needed, feel free to use models of any complexity to improve effectiveness, such as XGBoost, CatBoost, etc.\\n- If non-numeric columns exist, perform label encode together with all steps.\\n- Use the data from previous task result directly, do not mock or reload data yourself.\\n- Set suitable hyperparameters for the model, make metrics as high as possible.\\n\\n\\n# Code Steps:\\nStrictly follow steps below when you writing code if it's convenient.\\n\\n\\n# Output Example:\\nwhen current task is \\\"train a lightgbm model on training data\\\", and their are two steps in 'Code Steps', the code be like:\\n```python\\n# Step 1: check data type and convert to numeric\\nojb_cols = train.select_dtypes(include='object').columns.tolist()\\n\\nfor col in obj_cols:\\n encoder = LabelEncoder()\\n train[col] = encoder.fit_transform(train[col].unique().tolist() + ['unknown'])\\n test[col] = test[col].apply(lambda x: x if x in encoder.classes_ else 'unknown')\\n test[col] = encoder.transform(test[col])\\n\\n# Step 2: train lightgbm model\\nmodel = LGBMClassifier()\\nmodel.fit(train, y_train)\\n```end\\n\\n# Constraints:\\n- Ensure the output new code is executable in the same Jupyter notebook with previous tasks code have been executed.\\n- The output code should contain all steps implemented in 'Code Steps'.\\n\"}]": { + "code": "from sklearn.model_selection import train_test_split\nfrom sklearn.metrics import accuracy_score\nfrom xgboost import XGBClassifier\n\n# Step 1: Prepare the data\n# Separate features and target variable\nX = train_processed.drop('Survived', axis=1)\ny = train_processed['Survived']\n\n# Step 2: Split the data into training and validation sets\nX_train, X_val, y_train, y_val = train_test_split(X, y, test_size=0.2, random_state=42)\n\n# Step 3: Initialize and train the XGBoost model\n# Note: Hyperparameters are set to values that generally perform well, but further tuning could be done.\nmodel = XGBClassifier(n_estimators=100, learning_rate=0.05, max_depth=4, subsample=0.8, colsample_bytree=0.8, random_state=42)\nmodel.fit(X_train, y_train)\n\n# Step 4: Evaluate the model\ny_pred = model.predict(X_val)\naccuracy = accuracy_score(y_val, y_pred)\nprint(f'Validation Accuracy: {accuracy:.4f}')" + }, + "[{\"role\": \"system\", \"content\": \"You are an AI Python assistant. You will be given your previous implementation code of a task, runtime error results, and a hint to change the implementation appropriately. Write your full implementation \"}, {\"role\": \"user\", \"content\": \"\\nHere is an example for you.\\n\\nExample 1:\\n[previous impl]:\\n```python\\ndef add(a: int, b: int) -> int:\\n \\\"\\\"\\\"\\n Given integers a and b, return the total value of a and b.\\n \\\"\\\"\\\"\\n return a - b\\n```\\n\\n[runtime Error]:\\nTested passed:\\n\\nTests failed:\\nassert add(1, 2) == 3 # output: -1\\nassert add(1, 2) == 4 # output: -1\\n\\n[reflection on previous impl]:\\nThe implementation failed the test cases where the input integers are 1 and 2. The issue arises because the code does not add the two integers together, but instead subtracts the second integer from the first. To fix this issue, we should change the operator from `-` to `+` in the return statement. This will ensure that the function returns the correct output for the given input.\\n\\n[improved impl]:\\n```python\\ndef add(a: int, b: int) -> int:\\n \\\"\\\"\\\"\\n Given integers a and b, return the total value of a and b.\\n \\\"\\\"\\\"\\n return a + b\\n```\\n\\n[context]\\n[user: \\n# Background\\nAs a data scientist, you need to help user to achieve their goal [This is a titanic passenger survival dataset, your goal is to predict passenger survival outcome. The target column is Survived. Perform data analysis, data preprocessing, feature engineering, and modeling to predict the target. Report accuracy on the eval data. Train data path: 'tests/data/ml_datasets/titanic/split_train.csv', eval data path: 'tests/data/ml_datasets/titanic/split_eval.csv'.] step-by-step in an continuous Jupyter notebook.\\n\\n## Done Tasks\\n```python\\nimport pandas as pd\\ntrain_data = pd.read_csv('tests/data/ml_datasets/titanic/split_train.csv')\\ndisplay(train_data.head())\\ntrain_data.info()\\ntrain_data.describe()\\ntrain_data.describe(include=['O'])\\nmissing_values = train_data.isnull().sum()\\nmissing_values[missing_values > 0]\\nimport seaborn as sns\\nimport matplotlib.pyplot as plt\\nsns.countplot(x='Survived', data=train_data)\\nplt.title('Distribution of Survival')\\nplt.show()\\nsns.barplot(x='Sex', y='Survived', data=train_data)\\nplt.title('Survival Rate by Sex')\\nplt.show()\\nsns.barplot(x='Pclass', y='Survived', data=train_data)\\nplt.title('Survival Rate by Class')\\nplt.show()\\nsns.barplot(x='Embarked', y='Survived', data=train_data)\\nplt.title('Survival Rate by Embarkation Port')\\nplt.show()\\n\\nfrom metagpt.tools.libs.data_preprocess import FillMissingValue\\ntrain_processed = train_data.copy()\\neval_data = pd.read_csv('tests/data/ml_datasets/titanic/split_eval.csv')\\nfill_missing_age = FillMissingValue(features=['Age'], strategy='median')\\ntrain_processed = fill_missing_age.fit_transform(train_processed)\\neval_processed = fill_missing_age.transform(eval_data)\\nfill_missing_embarked = FillMissingValue(features=['Embarked'], strategy='most_frequent')\\ntrain_processed = fill_missing_embarked.fit_transform(train_processed)\\neval_processed = fill_missing_embarked.transform(eval_data)\\ntrain_processed['Cabin_Ind'] = train_processed['Cabin'].isnull().astype(int)\\neval_processed['Cabin_Ind'] = eval_processed['Cabin'].isnull().astype(int)\\nfrom sklearn.preprocessing import LabelEncoder\\nlabel_encoder = LabelEncoder()\\ntrain_processed['Sex'] = label_encoder.fit_transform(train_processed['Sex'])\\neval_processed['Sex'] = label_encoder.transform(eval_processed['Sex'])\\ntrain_processed['Embarked'] = label_encoder.fit_transform(train_processed['Embarked'])\\neval_processed['Embarked'] = label_encoder.transform(eval_processed['Embarked'])\\nfrom sklearn.preprocessing import StandardScaler\\nscaler = StandardScaler()\\nfor feature in ['Age', 'Fare']:\\n train_processed[[feature]] = scaler.fit_transform(train_processed[[feature]])\\n eval_processed[[feature]] = scaler.transform(eval_processed[[feature]])\\ndisplay(train_processed.head())\\n\\ntrain_processed['Family_Size'] = train_processed['SibSp'] + train_processed['Parch']\\neval_processed['Family_Size'] = eval_processed['SibSp'] + eval_processed['Parch']\\ntrain_processed['Is_Alone'] = (train_processed['Family_Size'] == 0).astype(int)\\neval_processed['Is_Alone'] = (eval_processed['Family_Size'] == 0).astype(int)\\ntrain_processed['Title'] = train_processed['Name'].str.extract(' ([A-Za-z]+)\\\\.', expand=False)\\neval_processed['Title'] = eval_processed['Name'].str.extract(' ([A-Za-z]+)\\\\.', expand=False)\\ntrain_processed['Deck'] = train_processed['Cabin'].str[0]\\neval_processed['Deck'] = eval_processed['Cabin'].str[0]\\nfrom metagpt.tools.libs.feature_engineering import CatCross\\ncat_cross = CatCross(cols=['Pclass', 'Sex'])\\ntrain_processed = cat_cross.fit_transform(train_processed)\\neval_processed = cat_cross.transform(eval_processed)\\nfor df in [train_processed, eval_processed]:\\n df.drop(['PassengerId', 'Name', 'Ticket', 'Cabin'], axis=1, inplace=True)\\n```end\\n\\n## Current Task\\nSelect and train a machine learning model using the processed train dataset.\\n\\n# Latest Data Info\\nLatest data info after previous tasks:\\ncolumn_info\\n{'Category': ['Title', 'Deck'], 'Numeric': ['Survived', 'Pclass', 'Sex', 'Age', 'SibSp', 'Parch', 'Fare', 'Embarked', 'Cabin_Ind', 'Family_Size', 'Is_Alone', 'Pclass_Sex'], 'Datetime': [], 'Others': []}\\n\\n\\n# Task\\nWrite complete code for 'Current Task'. And avoid duplicating code from 'Done Tasks', such as repeated import of packages, reading data, etc.\\nSpecifically, \\nThe current task is about training a model, please ensure high performance:\\n- Keep in mind that your user prioritizes results and is highly focused on model performance. So, when needed, feel free to use models of any complexity to improve effectiveness, such as XGBoost, CatBoost, etc.\\n- If non-numeric columns exist, perform label encode together with all steps.\\n- Use the data from previous task result directly, do not mock or reload data yourself.\\n- Set suitable hyperparameters for the model, make metrics as high as possible.\\n\\n\\n# Code Steps:\\nStrictly follow steps below when you writing code if it's convenient.\\n\\n\\n# Output Example:\\nwhen current task is \\\"train a lightgbm model on training data\\\", and their are two steps in 'Code Steps', the code be like:\\n```python\\n# Step 1: check data type and convert to numeric\\nojb_cols = train.select_dtypes(include='object').columns.tolist()\\n\\nfor col in obj_cols:\\n encoder = LabelEncoder()\\n train[col] = encoder.fit_transform(train[col].unique().tolist() + ['unknown'])\\n test[col] = test[col].apply(lambda x: x if x in encoder.classes_ else 'unknown')\\n test[col] = encoder.transform(test[col])\\n\\n# Step 2: train lightgbm model\\nmodel = LGBMClassifier()\\nmodel.fit(train, y_train)\\n```end\\n\\n# Constraints:\\n- Ensure the output new code is executable in the same Jupyter notebook with previous tasks code have been executed.\\n- The output code should contain all steps implemented in 'Code Steps'.\\n]\\n\\n[previous impl]\\nfrom sklearn.model_selection import train_test_split\\nfrom sklearn.metrics import accuracy_score\\nfrom xgboost import XGBClassifier\\n\\n# Step 1: Prepare the data\\n# Separate features and target variable\\nX = train_processed.drop('Survived', axis=1)\\ny = train_processed['Survived']\\n\\n# Step 2: Split the data into training and validation sets\\nX_train, X_val, y_train, y_val = train_test_split(X, y, test_size=0.2, random_state=42)\\n\\n# Step 3: Initialize and train the XGBoost model\\n# Note: Hyperparameters are set to values that generally perform well, but further tuning could be done.\\nmodel = XGBClassifier(n_estimators=100, learning_rate=0.05, max_depth=4, subsample=0.8, colsample_bytree=0.8, random_state=42)\\nmodel.fit(X_train, y_train)\\n\\n# Step 4: Evaluate the model\\ny_pred = model.predict(X_val)\\naccuracy = accuracy_score(y_val, y_pred)\\nprint(f'Validation Accuracy: {accuracy:.4f}')\\n[runtime Error]\\n[assistant: from sklearn.model_selection import train_test_split\\nfrom sklearn.metrics import accuracy_score\\nfrom xgboost import XGBClassifier\\n\\n# Step 1: Prepare the data\\n# Separate features and target variable\\nX = train_processed.drop('Survived', axis=1)\\ny = train_processed['Survived']\\n\\n# Step 2: Split the data into training and validation sets\\nX_train, X_val, y_train, y_val = train_test_split(X, y, test_size=0.2, random_state=42)\\n\\n# Step 3: Initialize and train the XGBoost model\\n# Note: Hyperparameters are set to values that generally perform well, but further tuning could be done.\\nmodel = XGBClassifier(n_estimators=100, learning_rate=0.05, max_depth=4, subsample=0.8, colsample_bytree=0.8, random_state=42)\\nmodel.fit(X_train, y_train)\\n\\n# Step 4: Evaluate the model\\ny_pred = model.predict(X_val)\\naccuracy = accuracy_score(y_val, y_pred)\\nprint(f'Validation Accuracy: {accuracy:.4f}'), user: Executed code failed, please reflect the cause of bug and then debug. Truncated to show only last 2000 characters\\n= self._temporary_data\\n 622 else:\\n--> 623 new, cat_codes, feature_names, feature_types = _proxy_transform(\\n 624 data,\\n 625 feature_names,\\n 626 feature_types,\\n 627 self._enable_categorical,\\n 628 )\\n 629 # Stage the data, meta info are copied inside C++ MetaInfo.\\n 630 self._temporary_data = (new, cat_codes, feature_names, feature_types)\\n\\nFile ~/miniconda3/envs/mg_temp/lib/python3.9/site-packages/xgboost/data.py:1315, in _proxy_transform(data, feature_names, feature_types, enable_categorical)\\n 1313 data = pd.DataFrame(data)\\n 1314 if _is_pandas_df(data):\\n-> 1315 arr, feature_names, feature_types = _transform_pandas_df(\\n 1316 data, enable_categorical, feature_names, feature_types\\n 1317 )\\n 1318 arr, _ = _ensure_np_dtype(arr, arr.dtype)\\n 1319 return arr, None, feature_names, feature_types\\n\\nFile ~/miniconda3/envs/mg_temp/lib/python3.9/site-packages/xgboost/data.py:490, in _transform_pandas_df(data, enable_categorical, feature_names, feature_types, meta, meta_type)\\n 483 for dtype in data.dtypes:\\n 484 if not (\\n 485 (dtype.name in _pandas_dtype_mapper)\\n 486 or is_pd_sparse_dtype(dtype)\\n 487 or (is_pd_cat_dtype(dtype) and enable_categorical)\\n 488 or is_pa_ext_dtype(dtype)\\n 489 ):\\n--> 490 _invalid_dataframe_dtype(data)\\n 491 if is_pa_ext_dtype(dtype):\\n 492 pyarrow_extension = True\\n\\nFile ~/miniconda3/envs/mg_temp/lib/python3.9/site-packages/xgboost/data.py:308, in _invalid_dataframe_dtype(data)\\n 306 type_err = \\\"DataFrame.dtypes for data must be int, float, bool or category.\\\"\\n 307 msg = f\\\"\\\"\\\"{type_err} {_ENABLE_CAT_ERR} {err}\\\"\\\"\\\"\\n--> 308 raise ValueError(msg)\\n\\nValueError: DataFrame.dtypes for data must be int, float, bool or category. When categorical type is supplied, The experimental DMatrix parameter`enable_categorical` must be set to `True`. Invalid columns:Title: object, Deck: object\\n\\n]\\n\\nAnalysis the error step by step, provide me improve method and code. Remember to follow [context] rerquirement. Don't forget write code for steps behind the error step.\\n[reflection on previous impl]:\\nxxx\\n\"}]": { + "reflection": "The error message indicates that the XGBoost model cannot handle columns with 'object' data types directly. The columns 'Title' and 'Deck' are of 'object' type and need to be encoded as numeric values before they can be used for training the XGBoost model. The error also suggests that if we want to use categorical data, we must set the `enable_categorical` parameter to `True` when creating the DMatrix. However, a more common practice is to convert these object types to numeric types using techniques such as label encoding or one-hot encoding. To resolve this issue, we should encode the 'Title' and 'Deck' columns using label encoding before training the model. Additionally, we should ensure that any other non-numeric columns are also converted to numeric types if they exist.", + "improved_impl": "from sklearn.preprocessing import LabelEncoder\n\n# Step 1: Encode non-numeric columns\nencoder = LabelEncoder()\ntrain_processed['Title'] = encoder.fit_transform(train_processed['Title'])\neval_processed['Title'] = encoder.transform(eval_processed['Title'])\ntrain_processed['Deck'] = encoder.fit_transform(train_processed['Deck'].astype(str))\neval_processed['Deck'] = encoder.transform(eval_processed['Deck'].astype(str))\n\n# Step 2: Prepare the data\n# Separate features and target variable\nX = train_processed.drop('Survived', axis=1)\ny = train_processed['Survived']\n\n# Step 3: Train the XGBoost model\nmodel = XGBClassifier(n_estimators=100, learning_rate=0.05, max_depth=4, subsample=0.8, colsample_bytree=0.8, random_state=42)\nmodel.fit(X, y)\n\n# Step 4: Evaluate the model on evaluation data\nX_eval = eval_processed.drop('Survived', axis=1)\ny_eval = eval_processed['Survived']\ny_pred_eval = model.predict(X_eval)\naccuracy_eval = accuracy_score(y_eval, y_pred_eval)\nprint(f'Evaluation Accuracy: {accuracy_eval:.4f}')" + }, + "[{\"role\": \"user\", \"content\": \"\\n# Background\\nAs a data scientist, you need to help user to achieve their goal [This is a titanic passenger survival dataset, your goal is to predict passenger survival outcome. The target column is Survived. Perform data analysis, data preprocessing, feature engineering, and modeling to predict the target. Report accuracy on the eval data. Train data path: 'tests/data/ml_datasets/titanic/split_train.csv', eval data path: 'tests/data/ml_datasets/titanic/split_eval.csv'.] step-by-step in an continuous Jupyter notebook.\\n\\n## Done Tasks\\n```python\\nimport pandas as pd\\ntrain_data = pd.read_csv('tests/data/ml_datasets/titanic/split_train.csv')\\ndisplay(train_data.head())\\ntrain_data.info()\\ntrain_data.describe()\\ntrain_data.describe(include=['O'])\\nmissing_values = train_data.isnull().sum()\\nmissing_values[missing_values > 0]\\nimport seaborn as sns\\nimport matplotlib.pyplot as plt\\nsns.countplot(x='Survived', data=train_data)\\nplt.title('Distribution of Survival')\\nplt.show()\\nsns.barplot(x='Sex', y='Survived', data=train_data)\\nplt.title('Survival Rate by Sex')\\nplt.show()\\nsns.barplot(x='Pclass', y='Survived', data=train_data)\\nplt.title('Survival Rate by Class')\\nplt.show()\\nsns.barplot(x='Embarked', y='Survived', data=train_data)\\nplt.title('Survival Rate by Embarkation Port')\\nplt.show()\\n\\nfrom metagpt.tools.libs.data_preprocess import FillMissingValue\\ntrain_processed = train_data.copy()\\neval_data = pd.read_csv('tests/data/ml_datasets/titanic/split_eval.csv')\\nfill_missing_age = FillMissingValue(features=['Age'], strategy='median')\\ntrain_processed = fill_missing_age.fit_transform(train_processed)\\neval_processed = fill_missing_age.transform(eval_data)\\nfill_missing_embarked = FillMissingValue(features=['Embarked'], strategy='most_frequent')\\ntrain_processed = fill_missing_embarked.fit_transform(train_processed)\\neval_processed = fill_missing_embarked.transform(eval_data)\\ntrain_processed['Cabin_Ind'] = train_processed['Cabin'].isnull().astype(int)\\neval_processed['Cabin_Ind'] = eval_processed['Cabin'].isnull().astype(int)\\nfrom sklearn.preprocessing import LabelEncoder\\nlabel_encoder = LabelEncoder()\\ntrain_processed['Sex'] = label_encoder.fit_transform(train_processed['Sex'])\\neval_processed['Sex'] = label_encoder.transform(eval_processed['Sex'])\\ntrain_processed['Embarked'] = label_encoder.fit_transform(train_processed['Embarked'])\\neval_processed['Embarked'] = label_encoder.transform(eval_processed['Embarked'])\\nfrom sklearn.preprocessing import StandardScaler\\nscaler = StandardScaler()\\nfor feature in ['Age', 'Fare']:\\n train_processed[[feature]] = scaler.fit_transform(train_processed[[feature]])\\n eval_processed[[feature]] = scaler.transform(eval_processed[[feature]])\\ndisplay(train_processed.head())\\n\\ntrain_processed['Family_Size'] = train_processed['SibSp'] + train_processed['Parch']\\neval_processed['Family_Size'] = eval_processed['SibSp'] + eval_processed['Parch']\\ntrain_processed['Is_Alone'] = (train_processed['Family_Size'] == 0).astype(int)\\neval_processed['Is_Alone'] = (eval_processed['Family_Size'] == 0).astype(int)\\ntrain_processed['Title'] = train_processed['Name'].str.extract(' ([A-Za-z]+)\\\\.', expand=False)\\neval_processed['Title'] = eval_processed['Name'].str.extract(' ([A-Za-z]+)\\\\.', expand=False)\\ntrain_processed['Deck'] = train_processed['Cabin'].str[0]\\neval_processed['Deck'] = eval_processed['Cabin'].str[0]\\nfrom metagpt.tools.libs.feature_engineering import CatCross\\ncat_cross = CatCross(cols=['Pclass', 'Sex'])\\ntrain_processed = cat_cross.fit_transform(train_processed)\\neval_processed = cat_cross.transform(eval_processed)\\nfor df in [train_processed, eval_processed]:\\n df.drop(['PassengerId', 'Name', 'Ticket', 'Cabin'], axis=1, inplace=True)\\n\\nfrom sklearn.preprocessing import LabelEncoder\\nencoder = LabelEncoder()\\ntrain_processed['Title'] = encoder.fit_transform(train_processed['Title'])\\neval_processed['Title'] = encoder.transform(eval_processed['Title'])\\ntrain_processed['Deck'] = encoder.fit_transform(train_processed['Deck'].astype(str))\\neval_processed['Deck'] = encoder.transform(eval_processed['Deck'].astype(str))\\nX = train_processed.drop('Survived', axis=1)\\ny = train_processed['Survived']\\nmodel = XGBClassifier(n_estimators=100, learning_rate=0.05, max_depth=4, subsample=0.8, colsample_bytree=0.8, random_state=42)\\nmodel.fit(X, y)\\nX_eval = eval_processed.drop('Survived', axis=1)\\ny_eval = eval_processed['Survived']\\ny_pred_eval = model.predict(X_eval)\\naccuracy_eval = accuracy_score(y_eval, y_pred_eval)\\nprint(f'Evaluation Accuracy: {accuracy_eval:.4f}')\\n```end\\n\\n## Current Task\\nEvaluate the model's accuracy using the eval dataset and report the results.\\n\\n# Latest Data Info\\nLatest data info after previous tasks:\\n\\n\\n# Task\\nWrite complete code for 'Current Task'. And avoid duplicating code from 'Done Tasks', such as repeated import of packages, reading data, etc.\\nSpecifically, \\nThe current task is about evaluating a model, please note the following:\\n- Ensure that the evaluated data is same processed as the training data. If not, remember use object in 'Done Tasks' to transform the data.\\n- Use trained model from previous task result directly, do not mock or reload model yourself.\\n\\n\\n# Code Steps:\\nStrictly follow steps below when you writing code if it's convenient.\\n\\n\\n# Output Example:\\nwhen current task is \\\"train a lightgbm model on training data\\\", and their are two steps in 'Code Steps', the code be like:\\n```python\\n# Step 1: check data type and convert to numeric\\nojb_cols = train.select_dtypes(include='object').columns.tolist()\\n\\nfor col in obj_cols:\\n encoder = LabelEncoder()\\n train[col] = encoder.fit_transform(train[col].unique().tolist() + ['unknown'])\\n test[col] = test[col].apply(lambda x: x if x in encoder.classes_ else 'unknown')\\n test[col] = encoder.transform(test[col])\\n\\n# Step 2: train lightgbm model\\nmodel = LGBMClassifier()\\nmodel.fit(train, y_train)\\n```end\\n\\n# Constraints:\\n- Ensure the output new code is executable in the same Jupyter notebook with previous tasks code have been executed.\\n- The output code should contain all steps implemented in 'Code Steps'.\\n\"}]": { + "code": "# Evaluate the model's accuracy using the eval dataset\nfrom sklearn.metrics import accuracy_score\n\n# Predict the target for the evaluation set\ny_pred_eval = model.predict(X_eval)\n\n# Calculate the accuracy for the evaluation set\naccuracy_eval = accuracy_score(y_eval, y_pred_eval)\n\n# Report the accuracy\nprint(f'Evaluation Accuracy: {accuracy_eval:.4f}')" + }, + "\n # Context:\n user: \n## User Requirement\nRun data analysis on sklearn Iris dataset, include a plot\n## Context\n\n## Current Plan\n[]\n## Current Task\n{}\n\n # Task:\n Based on the context, write a plan or modify an existing plan of what you should do to achieve the goal. A plan consists of one to 3 tasks.\n If you are modifying an existing plan, carefully follow the instruction, don't make unnecessary changes. Give the whole plan unless instructed to modify only one task of the plan.\n If you encounter errors on the current task, revise and output the current single task only.\n Output a list of jsons following the format:\n ```json\n [\n {\n \"task_id\": str = \"unique identifier for a task in plan, can be an ordinal\",\n \"dependent_task_ids\": list[str] = \"ids of tasks prerequisite to this task\",\n \"instruction\": \"what you should do in this task, one short phrase or sentence\",\n },\n ...\n ]\n ```\n ": "```json\n[\n {\n \"task_id\": \"1\",\n \"dependent_task_ids\": [],\n \"instruction\": \"Load the sklearn Iris dataset.\"\n },\n {\n \"task_id\": \"2\",\n \"dependent_task_ids\": [\"1\"],\n \"instruction\": \"Perform exploratory data analysis on the Iris dataset.\"\n },\n {\n \"task_id\": \"3\",\n \"dependent_task_ids\": [\"2\"],\n \"instruction\": \"Create a plot visualizing the Iris dataset features.\"\n }\n]\n```", + "[{\"role\": \"user\", \"content\": \"\\nPlease assign a task type to each task in the list below from the given categories:\\nTask 1: Load the sklearn Iris dataset.\\nTask 2: Perform exploratory data analysis on the Iris dataset.\\nTask 3: Create a plot visualizing the Iris dataset features.\\n\\n## All Task Type:\\n- **eda**: For performing exploratory data analysis\\n- **data_preprocess**: Only for changing value inplace.\\n- **feature_engineering**: Only for creating new columns for input data.\\n- **model_train**: Only for training model.\\n- **model_evaluate**: Only for evaluating model.\\n- **stable_diffusion**: Related to text2image, image2image using stable diffusion model.\\n- **image2webpage**: For converting image into webpage code.\\n- **web_scraping**: For scraping data from web pages.\\n- **other**: Any tools not in the defined categories\\n\"}]": { + "task_type": [ + "other", + "eda", + "other" + ] + }, + "[{\"role\": \"system\", \"content\": \"You are Code Interpreter, a world-class programmer that can complete any goal by executing code. Strictly follow the plan and generate code step by step. Each step of the code will be executed on the user's machine, and the user will provide the code execution results to you.**Notice: The code for the next step depends on the code for the previous step. Must reuse variables in the lastest other code directly, dont creat it again, it is very import for you. Use !pip install in a standalone block to install missing packages.Usually the libraries you need are already installed.Dont check if packages already imported.**\"}, {\"role\": \"user\", \"content\": \"\\n## User Requirement\\nRun data analysis on sklearn Iris dataset, include a plot\\n## Context\\n\\n## Current Plan\\n[\\n {\\n \\\"task_id\\\": \\\"1\\\",\\n \\\"dependent_task_ids\\\": [],\\n \\\"instruction\\\": \\\"Load the sklearn Iris dataset.\\\",\\n \\\"task_type\\\": \\\"other\\\",\\n \\\"code\\\": \\\"\\\",\\n \\\"result\\\": \\\"\\\",\\n \\\"is_success\\\": false,\\n \\\"is_finished\\\": false\\n },\\n {\\n \\\"task_id\\\": \\\"2\\\",\\n \\\"dependent_task_ids\\\": [\\n \\\"1\\\"\\n ],\\n \\\"instruction\\\": \\\"Perform exploratory data analysis on the Iris dataset.\\\",\\n \\\"task_type\\\": \\\"eda\\\",\\n \\\"code\\\": \\\"\\\",\\n \\\"result\\\": \\\"\\\",\\n \\\"is_success\\\": false,\\n \\\"is_finished\\\": false\\n },\\n {\\n \\\"task_id\\\": \\\"3\\\",\\n \\\"dependent_task_ids\\\": [\\n \\\"2\\\"\\n ],\\n \\\"instruction\\\": \\\"Create a plot visualizing the Iris dataset features.\\\",\\n \\\"task_type\\\": \\\"other\\\",\\n \\\"code\\\": \\\"\\\",\\n \\\"result\\\": \\\"\\\",\\n \\\"is_success\\\": false,\\n \\\"is_finished\\\": false\\n }\\n]\\n## Current Task\\n{\\\"task_id\\\":\\\"1\\\",\\\"dependent_task_ids\\\":[],\\\"instruction\\\":\\\"Load the sklearn Iris dataset.\\\",\\\"task_type\\\":\\\"other\\\",\\\"code_steps\\\":\\\"\\\",\\\"code\\\":\\\"\\\",\\\"result\\\":\\\"\\\",\\\"is_success\\\":false,\\\"is_finished\\\":false}\\n\"}, {\"role\": \"user\", \"content\": \"\\n# Instruction\\nWrite complete code for 'Current Task'. And avoid duplicating code from finished tasks, such as repeated import of packages, reading data, etc.\\nSpecifically, \\n\\n# Capabilities\\n- You can utilize pre-defined tools in any code lines from 'Available Tools' in the form of Python Class.\\n- You can freely combine the use of any other public packages, like sklearn, numpy, pandas, etc..\\n\\n# Available Tools (can be empty):\\nEach Class tool is described in JSON format. When you call a tool, import the tool first.\\n{}\\n\\n# Constraints:\\n- Ensure the output new code is executable in the same Jupyter notebook with previous tasks code have been executed.\\n- Always prioritize using pre-defined tools for the same functionality.\\n\"}]": { + "code": "from sklearn.datasets import load_iris\niris_data = load_iris()" + }, + "[{\"role\": \"system\", \"content\": \"You are Code Interpreter, a world-class programmer that can complete any goal by executing code. Strictly follow the plan and generate code step by step. Each step of the code will be executed on the user's machine, and the user will provide the code execution results to you.**Notice: The code for the next step depends on the code for the previous step. Must reuse variables in the lastest other code directly, dont creat it again, it is very import for you. Use !pip install in a standalone block to install missing packages.Usually the libraries you need are already installed.Dont check if packages already imported.**\"}, {\"role\": \"user\", \"content\": \"\\n## User Requirement\\nRun data analysis on sklearn Iris dataset, include a plot\\n## Context\\n\\n## Current Plan\\n[\\n {\\n \\\"task_id\\\": \\\"1\\\",\\n \\\"dependent_task_ids\\\": [],\\n \\\"instruction\\\": \\\"Load the sklearn Iris dataset.\\\",\\n \\\"task_type\\\": \\\"other\\\",\\n \\\"code\\\": \\\"from sklearn.datasets import load_iris\\\\niris_data = load_iris()\\\",\\n \\\"result\\\": \\\"\\\",\\n \\\"is_success\\\": true,\\n \\\"is_finished\\\": true\\n },\\n {\\n \\\"task_id\\\": \\\"2\\\",\\n \\\"dependent_task_ids\\\": [\\n \\\"1\\\"\\n ],\\n \\\"instruction\\\": \\\"Perform exploratory data analysis on the Iris dataset.\\\",\\n \\\"task_type\\\": \\\"eda\\\",\\n \\\"code\\\": \\\"\\\",\\n \\\"result\\\": \\\"\\\",\\n \\\"is_success\\\": false,\\n \\\"is_finished\\\": false\\n },\\n {\\n \\\"task_id\\\": \\\"3\\\",\\n \\\"dependent_task_ids\\\": [\\n \\\"2\\\"\\n ],\\n \\\"instruction\\\": \\\"Create a plot visualizing the Iris dataset features.\\\",\\n \\\"task_type\\\": \\\"other\\\",\\n \\\"code\\\": \\\"\\\",\\n \\\"result\\\": \\\"\\\",\\n \\\"is_success\\\": false,\\n \\\"is_finished\\\": false\\n }\\n]\\n## Current Task\\n{\\\"task_id\\\":\\\"2\\\",\\\"dependent_task_ids\\\":[\\\"1\\\"],\\\"instruction\\\":\\\"Perform exploratory data analysis on the Iris dataset.\\\",\\\"task_type\\\":\\\"eda\\\",\\\"code_steps\\\":\\\"\\\",\\\"code\\\":\\\"\\\",\\\"result\\\":\\\"\\\",\\\"is_success\\\":false,\\\"is_finished\\\":false}\\n\"}, {\"role\": \"user\", \"content\": \"\\n# Instruction\\nWrite complete code for 'Current Task'. And avoid duplicating code from finished tasks, such as repeated import of packages, reading data, etc.\\nSpecifically, \\n\\n# Capabilities\\n- You can utilize pre-defined tools in any code lines from 'Available Tools' in the form of Python Class.\\n- You can freely combine the use of any other public packages, like sklearn, numpy, pandas, etc..\\n\\n# Available Tools (can be empty):\\nEach Class tool is described in JSON format. When you call a tool, import the tool first.\\n{}\\n\\n# Constraints:\\n- Ensure the output new code is executable in the same Jupyter notebook with previous tasks code have been executed.\\n- Always prioritize using pre-defined tools for the same functionality.\\n\"}]": { + "code": "import pandas as pd\n\n# Convert the loaded Iris dataset into a DataFrame for easier manipulation\niris_df = pd.DataFrame(data=iris_data.data, columns=iris_data.feature_names)\niris_df['target'] = iris_data.target\n\n# Display basic statistical details like percentile, mean, std etc. of a data frame\niris_stats = iris_df.describe()\n\n# Display the first few rows of the DataFrame\niris_head = iris_df.head()\n\n# Display the class distribution\niris_target_counts = iris_df['target'].value_counts()\n\n# Output the results\nprint('Basic Statistical Details:\\n', iris_stats)\nprint('\\nFirst Five Rows:\\n', iris_head)\nprint('\\nClass Distribution:\\n', iris_target_counts)" + }, + "[{\"role\": \"system\", \"content\": \"You are Code Interpreter, a world-class programmer that can complete any goal by executing code. Strictly follow the plan and generate code step by step. Each step of the code will be executed on the user's machine, and the user will provide the code execution results to you.**Notice: The code for the next step depends on the code for the previous step. Must reuse variables in the lastest other code directly, dont creat it again, it is very import for you. Use !pip install in a standalone block to install missing packages.Usually the libraries you need are already installed.Dont check if packages already imported.**\"}, {\"role\": \"user\", \"content\": \"\\n## User Requirement\\nRun data analysis on sklearn Iris dataset, include a plot\\n## Context\\n\\n## Current Plan\\n[\\n {\\n \\\"task_id\\\": \\\"1\\\",\\n \\\"dependent_task_ids\\\": [],\\n \\\"instruction\\\": \\\"Load the sklearn Iris dataset.\\\",\\n \\\"task_type\\\": \\\"other\\\",\\n \\\"code\\\": \\\"from sklearn.datasets import load_iris\\\\niris_data = load_iris()\\\",\\n \\\"result\\\": \\\"\\\",\\n \\\"is_success\\\": true,\\n \\\"is_finished\\\": true\\n },\\n {\\n \\\"task_id\\\": \\\"2\\\",\\n \\\"dependent_task_ids\\\": [\\n \\\"1\\\"\\n ],\\n \\\"instruction\\\": \\\"Perform exploratory data analysis on the Iris dataset.\\\",\\n \\\"task_type\\\": \\\"eda\\\",\\n \\\"code\\\": \\\"import pandas as pd\\\\n\\\\n# Convert the loaded Iris dataset into a DataFrame for easier manipulation\\\\niris_df = pd.DataFrame(data=iris_data.data, columns=iris_data.feature_names)\\\\niris_df['target'] = iris_data.target\\\\n\\\\n# Display basic statistical details like percentile, mean, std etc. of a data frame\\\\niris_stats = iris_df.describe()\\\\n\\\\n# Display the first few rows of the DataFrame\\\\niris_head = iris_df.head()\\\\n\\\\n# Display the class distribution\\\\niris_target_counts = iris_df['target'].value_counts()\\\\n\\\\n# Output the results\\\\nprint('Basic Statistical Details:\\\\\\\\n', iris_stats)\\\\nprint('\\\\\\\\nFirst Five Rows:\\\\\\\\n', iris_head)\\\\nprint('\\\\\\\\nClass Distribution:\\\\\\\\n', iris_target_counts)\\\",\\n \\\"result\\\": \\\"Basic Statistical Details:\\\\n sepal length (cm) sepal width (cm) petal length (cm) \\\\\\\\\\\\ncount 150.000000 150.000000 150.000000 \\\\nmean 5.843333 3.057333 3.758000 \\\\nstd 0.828066 0.435866 1.765298 \\\\nmin 4.300000 2.000000 1.000000 \\\\n25% 5.100000 2.800000 1.600000 \\\\n50% 5.800000 3.000000 4.350000 \\\\n75% 6.400000 3.300000 5.100000 \\\\nmax 7.900000 4.400000 6.900000 \\\\n\\\\n petal width (cm) target \\\\ncount 150.000000 150.000000 \\\\nmean 1.199333 1.000000 \\\\nstd 0.762238 0.819232 \\\\nmin 0.100000 0.000000 \\\\n25% 0.300000 0.000000 \\\\n50% 1.300000 1.000000 \\\\n75% 1.800000 2.000000 \\\\nmax 2.500000 2.000000 \\\\n\\\\nFirst Five Rows:\\\\n sepal length (cm) sepal width (cm) petal length (cm) petal width (cm) \\\\\\\\\\\\n0 5.1 3.5 1.4 0.2 \\\\n1 4.9 3.0 1.4 0.2 \\\\n2 4.7 3.2 1.3 0.2 \\\\n3 4.6 3.1 1.5 0.2 \\\\n4 5.0 3.6 1.4 0.2 \\\\n\\\\n target \\\\n0 0 \\\\n1 0 \\\\n2 0 \\\\n3 0 \\\\n4 0 \\\\n\\\\nClass Distribution:\\\\n target\\\\n0 50\\\\n1 50\\\\n2 50\\\\nName: count, dtype: int64\\\\n\\\",\\n \\\"is_success\\\": true,\\n \\\"is_finished\\\": true\\n },\\n {\\n \\\"task_id\\\": \\\"3\\\",\\n \\\"dependent_task_ids\\\": [\\n \\\"2\\\"\\n ],\\n \\\"instruction\\\": \\\"Create a plot visualizing the Iris dataset features.\\\",\\n \\\"task_type\\\": \\\"other\\\",\\n \\\"code\\\": \\\"\\\",\\n \\\"result\\\": \\\"\\\",\\n \\\"is_success\\\": false,\\n \\\"is_finished\\\": false\\n }\\n]\\n## Current Task\\n{\\\"task_id\\\":\\\"3\\\",\\\"dependent_task_ids\\\":[\\\"2\\\"],\\\"instruction\\\":\\\"Create a plot visualizing the Iris dataset features.\\\",\\\"task_type\\\":\\\"other\\\",\\\"code_steps\\\":\\\"\\\",\\\"code\\\":\\\"\\\",\\\"result\\\":\\\"\\\",\\\"is_success\\\":false,\\\"is_finished\\\":false}\\n\"}, {\"role\": \"user\", \"content\": \"\\n# Instruction\\nWrite complete code for 'Current Task'. And avoid duplicating code from finished tasks, such as repeated import of packages, reading data, etc.\\nSpecifically, \\n\\n# Capabilities\\n- You can utilize pre-defined tools in any code lines from 'Available Tools' in the form of Python Class.\\n- You can freely combine the use of any other public packages, like sklearn, numpy, pandas, etc..\\n\\n# Available Tools (can be empty):\\nEach Class tool is described in JSON format. When you call a tool, import the tool first.\\n{}\\n\\n# Constraints:\\n- Ensure the output new code is executable in the same Jupyter notebook with previous tasks code have been executed.\\n- Always prioritize using pre-defined tools for the same functionality.\\n\"}]": { + "code": "import matplotlib.pyplot as plt\nimport seaborn as sns\n\n# Set the style of seaborn\nsns.set(style=\"whitegrid\")\n\n# Create a pairplot to visualize the relationships between features\npair_plot = sns.pairplot(iris_df, hue='target', markers=['o', 's', 'D'], palette='husl')\n\n# Show the plot\nplt.show()" + }, + "\n # Context:\n user: run analysis on sklearn iris dataset\n # Task:\n Based on the context, write a plan or modify an existing plan of what you should do to achieve the goal. A plan consists of one to 5 tasks.\n If you are modifying an existing plan, carefully follow the instruction, don't make unnecessary changes. Give the whole plan unless instructed to modify only one task of the plan.\n If you encounter errors on the current task, revise and output the current single task only.\n Output a list of jsons following the format:\n ```json\n [\n {\n \"task_id\": str = \"unique identifier for a task in plan, can be an ordinal\",\n \"dependent_task_ids\": list[str] = \"ids of tasks prerequisite to this task\",\n \"instruction\": \"what you should do in this task, one short phrase or sentence\",\n },\n ...\n ]\n ```\n ": "```json\n[\n {\n \"task_id\": \"1\",\n \"dependent_task_ids\": [],\n \"instruction\": \"Import the Iris dataset from scikit-learn.\"\n },\n {\n \"task_id\": \"2\",\n \"dependent_task_ids\": [\"1\"],\n \"instruction\": \"Perform exploratory data analysis to understand the dataset.\"\n },\n {\n \"task_id\": \"3\",\n \"dependent_task_ids\": [\"2\"],\n \"instruction\": \"Preprocess the data if necessary (e.g., scaling, encoding).\"\n },\n {\n \"task_id\": \"4\",\n \"dependent_task_ids\": [\"3\"],\n \"instruction\": \"Split the dataset into training and testing sets.\"\n },\n {\n \"task_id\": \"5\",\n \"dependent_task_ids\": [\"4\"],\n \"instruction\": \"Choose a suitable model and train it on the dataset.\"\n },\n {\n \"task_id\": \"6\",\n \"dependent_task_ids\": [\"5\"],\n \"instruction\": \"Evaluate the model's performance on the test set.\"\n },\n {\n \"task_id\": \"7\",\n \"dependent_task_ids\": [\"6\"],\n \"instruction\": \"Report the results of the analysis.\"\n }\n]\n```", + "[{\"role\": \"user\", \"content\": \"\\n## User Requirement:\\n对数据集进行数据清洗\\n\\n## Task\\nRecommend up to five tools from 'Available Tools' that can help solve the 'User Requirement'. \\nThis is a detailed code steps for current task. You can refer to it when recommending tools.\\n\\n\\n## Available Tools:\\n{'FillMissingValue': 'Completing missing values with simple strategies', 'MinMaxScale': 'Transform features by scaling each feature to a range, witch is (0, 1)', 'StandardScale': 'Standardize features by removing the mean and scaling to unit variance', 'MaxAbsScale': 'cale each feature by its maximum absolute value', 'RobustScale': 'Apply the RobustScaler to scale features using statistics that are robust to outliers.', 'OrdinalEncode': 'Encode categorical features as ordinal integers.', 'OneHotEncode': 'Apply one-hot encoding to specified categorical columns, the original columns will be dropped.', 'LabelEncode': 'Apply label encoding to specified categorical columns in-place.'}\\n\\n## Tool Selection and Instructions:\\n- Select tools most relevant to completing the 'User Requirement'.\\n- If you believe that no tools are suitable, indicate with an empty list.\\n- Only list the names of the tools, not the full schema of each tool.\\n- Ensure selected tools are listed in 'Available Tools'.\\n\"}]": { + "recommend_tools": [ + "FillMissingValue", + "MinMaxScale", + "StandardScale", + "RobustScale", + "OneHotEncode" + ] + }, + "[{\"role\": \"user\", \"content\": \"\\n# Background\\nAs a data scientist, you need to help user to achieve their goal [构造数据集并进行数据清洗] step-by-step in an continuous Jupyter notebook.\\n\\n## Done Tasks\\n```python\\n import pandas as pd\\n df = pd.DataFrame({\\n 'a': [1, 2, 3, 4, 5],\\n 'b': [1.1, 2.2, 3.3, 4.4, np.nan],\\n 'c': ['aa', 'bb', 'cc', 'dd', 'ee'],\\n 'd': [1, 2, 3, 4, 5]\\n })\\n```end\\n\\n## Current Task\\n对数据集进行数据清洗\\n\\n# Latest Data Info\\nLatest data info after previous tasks:\\n\\n\\n# Task\\nWrite complete code for 'Current Task'. And avoid duplicating code from 'Done Tasks', such as repeated import of packages, reading data, etc.\\nSpecifically, \\nThe current task is about data preprocessing, please note the following:\\n- Monitor data types per column, applying appropriate methods.\\n- Ensure operations are on existing dataset columns.\\n- Avoid writing processed data to files.\\n- Avoid any change to label column, such as standardization, etc.\\n- Prefer alternatives to one-hot encoding for categorical data.\\n- Only encode or scale necessary columns to allow for potential feature-specific engineering tasks (like time_extract, binning, extraction, etc.) later.\\n- Each step do data preprocessing to train, must do same for test separately at the same time.\\n\\n\\n# Code Steps:\\nStrictly follow steps below when you writing code if it's convenient.\\n\\n\\n# Capabilities\\n- You can utilize pre-defined tools in any code lines from 'Available Tools' in the form of Python Class.\\n- You can freely combine the use of any other public packages, like sklearn, numpy, pandas, etc..\\n\\n# Available Tools:\\nEach Class tool is described in JSON format. When you call a tool, import the tool from its path first.\\n{'FillMissingValue': {'type': 'class', 'description': 'Completing missing values with simple strategies', 'methods': {'__init__': {'description': 'Initialize self.', 'parameters': {'properties': {'features': {'type': 'list', 'description': 'columns to be processed'}, 'strategy': {'type': 'str', 'description': 'the imputation strategy, notice mean/median can only be used for numeric features', 'default': 'mean', 'enum': ['mean', 'median', 'most_frequent', 'constant']}, 'fill_value': {'type': 'int', 'description': 'fill_value is used to replace all occurrences of missing_values', 'default': None}}, 'required': ['features']}}, 'fit': {'description': 'Fit the FillMissingValue model.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}}, 'transform': {'description': 'Transform the input DataFrame with the fitted model.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': {'df': {'type': 'DataFrame', 'description': 'The transformed DataFrame.'}}}, 'fit_transform': {'description': 'Fit and transform the input DataFrame.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': {'df': {'type': 'DataFrame', 'description': 'The transformed DataFrame.'}}}}, 'tool_path': 'metagpt/tools/libs/data_preprocess.py'}, 'MinMaxScale': {'type': 'class', 'description': 'Transform features by scaling each feature to a range, witch is (0, 1)', 'methods': {'__init__': {'description': 'Initialize self.', 'parameters': {'properties': {'features': {'type': 'list', 'description': 'columns to be processed'}}, 'required': ['features']}}, 'fit': {'description': 'Fit the MinMaxScale model.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}}, 'transform': {'description': 'Transform the input DataFrame with the fitted model.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': {'df': {'type': 'DataFrame', 'description': 'The transformed DataFrame.'}}}, 'fit_transform': {'description': 'Fit and transform the input DataFrame.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': {'df': {'type': 'DataFrame', 'description': 'The transformed DataFrame.'}}}}, 'tool_path': 'metagpt/tools/libs/data_preprocess.py'}, 'StandardScale': {'type': 'class', 'description': 'Standardize features by removing the mean and scaling to unit variance', 'methods': {'__init__': {'description': 'Initialize self.', 'parameters': {'properties': {'features': {'type': 'list', 'description': 'columns to be processed'}}, 'required': ['features']}}, 'fit': {'description': 'Fit the StandardScale model.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}}, 'transform': {'description': 'Transform the input DataFrame with the fitted model.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': {'df': {'type': 'DataFrame', 'description': 'The transformed DataFrame.'}}}, 'fit_transform': {'description': 'Fit and transform the input DataFrame.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': {'df': {'type': 'DataFrame', 'description': 'The transformed DataFrame.'}}}}, 'tool_path': 'metagpt/tools/libs/data_preprocess.py'}, 'RobustScale': {'type': 'class', 'description': 'Apply the RobustScaler to scale features using statistics that are robust to outliers.', 'methods': {'__init__': {'description': 'Initialize the RobustScale instance with feature names. ', 'parameters': {'properties': {'features': {'type': 'list', 'description': 'List of feature names to be scaled.'}}, 'required': ['features']}}, 'fit': {'description': 'Compute the median and IQR for scaling. ', 'parameters': {'properties': {'df': {'type': 'pd.DataFrame', 'description': 'Dataframe containing the features.'}}, 'required': ['df']}}, 'fit_transform': {'description': 'Fit and transform the input DataFrame. ', 'parameters': {'properties': {'df': {'type': 'pd.DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': [{'type': 'pd.DataFrame', 'description': 'The transformed DataFrame.'}]}, 'transform': {'description': 'Scale features using the previously computed median and IQR. ', 'parameters': {'properties': {'df': {'type': 'pd.DataFrame', 'description': 'Dataframe containing the features to be scaled.'}}, 'required': ['df']}, 'returns': [{'type': 'pd.DataFrame', 'description': 'A new dataframe with scaled features.'}]}}, 'tool_path': 'metagpt/tools/libs/data_preprocess.py'}, 'OneHotEncode': {'type': 'class', 'description': 'Apply one-hot encoding to specified categorical columns, the original columns will be dropped.', 'methods': {'__init__': {'description': 'Initialize self.', 'parameters': {'properties': {'features': {'type': 'list', 'description': 'Categorical columns to be one-hot encoded and dropped'}}, 'required': ['features']}}, 'fit': {'description': 'Fit the OneHotEncoding model.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}}, 'transform': {'description': 'Transform the input DataFrame with the fitted model.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': {'df': {'type': 'DataFrame', 'description': 'The transformed DataFrame.'}}}, 'fit_transform': {'description': 'Fit and transform the input DataFrame.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': {'df': {'type': 'DataFrame', 'description': 'The transformed DataFrame.'}}}}, 'tool_path': 'metagpt/tools/libs/data_preprocess.py'}}\\n\\n# Output Example:\\nwhen current task is \\\"do data preprocess, like fill missing value, handle outliers, etc.\\\", and their are two steps in 'Code Steps', the code be like:\\n```python\\n# Step 1: fill missing value\\n# Tools used: ['FillMissingValue']\\nfrom metagpt.tools.libs.data_preprocess import FillMissingValue\\n\\ntrain_processed = train.copy()\\ntest_processed = test.copy()\\nnum_cols = train_processed.select_dtypes(include='number').columns.tolist()\\nif 'label' in num_cols:\\n num_cols.remove('label')\\nfill_missing_value = FillMissingValue(features=num_cols, strategy='mean')\\nfill_missing_value.fit(train_processed)\\ntrain_processed = fill_missing_value.transform(train_processed)\\ntest_processed = fill_missing_value.transform(test_processed)\\n\\n# Step 2: handle outliers\\nfor col in num_cols:\\n low, high = train_processed[col].quantile([0.01, 0.99])\\n train_processed[col] = train_processed[col].clip(low, high)\\n test_processed[col] = test_processed[col].clip(low, high)\\n```end\\n\\n# Constraints:\\n- Ensure the output new code is executable in the same Jupyter notebook with previous tasks code have been executed.\\n- Always prioritize using pre-defined tools for the same functionality.\\n- Always copy the DataFrame before processing it and use the copy to process.\\n- The output code should contain all steps implemented correctly in 'Code Steps'.\\n\"}]": { + "code": "# Step 1: Fill missing values\nfrom metagpt.tools.libs.data_preprocess import FillMissingValue\n\ndf_processed = df.copy()\n\n# Identify numeric columns with missing values\nnumeric_cols_with_na = df_processed.select_dtypes(include=['number']).columns[df_processed.isna().any()].tolist()\n\n# Fill missing values with the mean for numeric columns\nfill_missing_value = FillMissingValue(features=numeric_cols_with_na, strategy='mean')\ndf_processed = fill_missing_value.fit_transform(df_processed)\n\n# Step 2: Scale numeric columns\nfrom metagpt.tools.libs.data_preprocess import MinMaxScale\n\n# Identify numeric columns\nnumeric_cols = df_processed.select_dtypes(include=['number']).columns.tolist()\n\n# Scale numeric columns\nmin_max_scaler = MinMaxScale(features=numeric_cols)\ndf_processed = min_max_scaler.fit_transform(df_processed)\n\n# Note: Since there is no separate test set provided, the scaling is only applied to the existing dataset." + }, + "[{\"role\": \"user\", \"content\": \"\\n## User Requirement:\\nclean and preprocess the data\\n\\n## Task\\nRecommend up to five tools from 'Available Tools' that can help solve the 'User Requirement'. \\nThis is a detailed code steps for current task. You can refer to it when recommending tools.\\n\\n\\n## Available Tools:\\n{'FillMissingValue': 'Filling missing values', 'SplitBins': 'Bin continuous data into intervals and return the bin identifier encoded as an integer value'}\\n\\n## Tool Selection and Instructions:\\n- Select tools most relevant to completing the 'User Requirement'.\\n- If you believe that no tools are suitable, indicate with an empty list.\\n- Only list the names of the tools, not the full schema of each tool.\\n- Ensure selected tools are listed in 'Available Tools'.\\n\"}]": { + "recommend_tools": [ + "FillMissingValue" + ] + }, + "[{\"role\": \"system\", \"content\": \"You are Code Interpreter, a world-class programmer that can complete any goal by executing code. Strictly follow the plan and generate code step by step. Each step of the code will be executed on the user's machine, and the user will provide the code execution results to you.**Notice: The code for the next step depends on the code for the previous step. Must reuse variables in the lastest other code directly, dont creat it again, it is very import for you. Use !pip install in a standalone block to install missing packages.Usually the libraries you need are already installed.Dont check if packages already imported.**\"}, {\"role\": \"user\", \"content\": \"\\n## User Requirement\\n构造数据集并进行数据清洗\\n## Context\\n\\n## Current Plan\\n[Task(task_id='1', dependent_task_ids=[], instruction='随机生成一个pandas DataFrame数据集', task_type='other', code_steps='', code=\\\"\\\\n import pandas as pd\\\\n df = pd.DataFrame({\\\\n 'a': [1, 2, 3, 4, 5],\\\\n 'b': [1.1, 2.2, 3.3, 4.4, np.nan],\\\\n 'c': ['aa', 'bb', 'cc', 'dd', 'ee'],\\\\n 'd': [1, 2, 3, 4, 5]\\\\n })\\\\n \\\", result='', is_success=False, is_finished=True), Task(task_id='2', dependent_task_ids=['1'], instruction='对数据集进行数据清洗', task_type='data_preprocess', code_steps='', code='', result='', is_success=False, is_finished=False)]\\n## Current Task\\n{\\\"task_id\\\":\\\"2\\\",\\\"dependent_task_ids\\\":[\\\"1\\\"],\\\"instruction\\\":\\\"对数据集进行数据清洗\\\",\\\"task_type\\\":\\\"data_preprocess\\\",\\\"code_steps\\\":\\\"\\\",\\\"code\\\":\\\"\\\",\\\"result\\\":\\\"\\\",\\\"is_success\\\":false,\\\"is_finished\\\":false}\\n\"}, {\"role\": \"user\", \"content\": \"\\n# Instruction\\nWrite complete code for 'Current Task'. And avoid duplicating code from finished tasks, such as repeated import of packages, reading data, etc.\\nSpecifically, \\nThe current task is about data preprocessing, please note the following:\\n- Monitor data types per column, applying appropriate methods.\\n- Ensure operations are on existing dataset columns.\\n- Avoid writing processed data to files.\\n- Avoid any change to label column, such as standardization, etc.\\n- Prefer alternatives to one-hot encoding for categorical data.\\n- Only encode or scale necessary columns to allow for potential feature-specific engineering tasks (like time_extract, binning, extraction, etc.) later.\\n- Each step do data preprocessing to train, must do same for test separately at the same time.\\n\\n\\n# Capabilities\\n- You can utilize pre-defined tools in any code lines from 'Available Tools' in the form of Python Class.\\n- You can freely combine the use of any other public packages, like sklearn, numpy, pandas, etc..\\n\\n# Available Tools (can be empty):\\nEach Class tool is described in JSON format. When you call a tool, import the tool first.\\n{'FillMissingValue': {'type': 'class', 'description': 'Completing missing values with simple strategies', 'methods': {'__init__': {'description': 'Initialize self.', 'parameters': {'properties': {'features': {'type': 'list', 'description': 'columns to be processed'}, 'strategy': {'type': 'str', 'description': 'the imputation strategy, notice mean/median can only be used for numeric features', 'default': 'mean', 'enum': ['mean', 'median', 'most_frequent', 'constant']}, 'fill_value': {'type': 'int', 'description': 'fill_value is used to replace all occurrences of missing_values', 'default': None}}, 'required': ['features']}}, 'fit': {'description': 'Fit the FillMissingValue model.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}}, 'transform': {'description': 'Transform the input DataFrame with the fitted model.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': {'df': {'type': 'DataFrame', 'description': 'The transformed DataFrame.'}}}, 'fit_transform': {'description': 'Fit and transform the input DataFrame.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': {'df': {'type': 'DataFrame', 'description': 'The transformed DataFrame.'}}}}, 'tool_path': 'metagpt/tools/libs/data_preprocess.py'}, 'MinMaxScale': {'type': 'class', 'description': 'Transform features by scaling each feature to a range, witch is (0, 1)', 'methods': {'__init__': {'description': 'Initialize self.', 'parameters': {'properties': {'features': {'type': 'list', 'description': 'columns to be processed'}}, 'required': ['features']}}, 'fit': {'description': 'Fit the MinMaxScale model.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}}, 'transform': {'description': 'Transform the input DataFrame with the fitted model.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': {'df': {'type': 'DataFrame', 'description': 'The transformed DataFrame.'}}}, 'fit_transform': {'description': 'Fit and transform the input DataFrame.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': {'df': {'type': 'DataFrame', 'description': 'The transformed DataFrame.'}}}}, 'tool_path': 'metagpt/tools/libs/data_preprocess.py'}, 'StandardScale': {'type': 'class', 'description': 'Standardize features by removing the mean and scaling to unit variance', 'methods': {'__init__': {'description': 'Initialize self.', 'parameters': {'properties': {'features': {'type': 'list', 'description': 'columns to be processed'}}, 'required': ['features']}}, 'fit': {'description': 'Fit the StandardScale model.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}}, 'transform': {'description': 'Transform the input DataFrame with the fitted model.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': {'df': {'type': 'DataFrame', 'description': 'The transformed DataFrame.'}}}, 'fit_transform': {'description': 'Fit and transform the input DataFrame.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': {'df': {'type': 'DataFrame', 'description': 'The transformed DataFrame.'}}}}, 'tool_path': 'metagpt/tools/libs/data_preprocess.py'}, 'RobustScale': {'type': 'class', 'description': 'Apply the RobustScaler to scale features using statistics that are robust to outliers.', 'methods': {'__init__': {'description': 'Initialize the RobustScale instance with feature names. ', 'parameters': {'properties': {'features': {'type': 'list', 'description': 'List of feature names to be scaled.'}}, 'required': ['features']}}, 'fit': {'description': 'Compute the median and IQR for scaling. ', 'parameters': {'properties': {'df': {'type': 'pd.DataFrame', 'description': 'Dataframe containing the features.'}}, 'required': ['df']}}, 'fit_transform': {'description': 'Fit and transform the input DataFrame. ', 'parameters': {'properties': {'df': {'type': 'pd.DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': [{'type': 'pd.DataFrame', 'description': 'The transformed DataFrame.'}]}, 'transform': {'description': 'Scale features using the previously computed median and IQR. ', 'parameters': {'properties': {'df': {'type': 'pd.DataFrame', 'description': 'Dataframe containing the features to be scaled.'}}, 'required': ['df']}, 'returns': [{'type': 'pd.DataFrame', 'description': 'A new dataframe with scaled features.'}]}}, 'tool_path': 'metagpt/tools/libs/data_preprocess.py'}, 'OneHotEncode': {'type': 'class', 'description': 'Apply one-hot encoding to specified categorical columns, the original columns will be dropped.', 'methods': {'__init__': {'description': 'Initialize self.', 'parameters': {'properties': {'features': {'type': 'list', 'description': 'Categorical columns to be one-hot encoded and dropped'}}, 'required': ['features']}}, 'fit': {'description': 'Fit the OneHotEncoding model.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}}, 'transform': {'description': 'Transform the input DataFrame with the fitted model.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': {'df': {'type': 'DataFrame', 'description': 'The transformed DataFrame.'}}}, 'fit_transform': {'description': 'Fit and transform the input DataFrame.', 'parameters': {'properties': {'df': {'type': 'DataFrame', 'description': 'The input DataFrame.'}}, 'required': ['df']}, 'returns': {'df': {'type': 'DataFrame', 'description': 'The transformed DataFrame.'}}}}, 'tool_path': 'metagpt/tools/libs/data_preprocess.py'}}\\n\\n# Constraints:\\n- Ensure the output new code is executable in the same Jupyter notebook with previous tasks code have been executed.\\n- Always prioritize using pre-defined tools for the same functionality.\\n\"}]": { + "code": "from metagpt.tools.libs.data_preprocess import FillMissingValue\n\n# Instantiate the FillMissingValue class for numeric columns with missing values\ncolumns_with_missing_values = ['b']\nfill_missing_value = FillMissingValue(features=columns_with_missing_values, strategy='mean')\n\n# Apply the fit_transform method to the DataFrame to fill missing values\ndf = fill_missing_value.fit_transform(df)\n" + }, + "[{\"role\": \"system\", \"content\": \"You are Code Interpreter, a world-class programmer that can complete any goal by executing code. Strictly follow the plan and generate code step by step. Each step of the code will be executed on the user's machine, and the user will provide the code execution results to you.**Notice: The code for the next step depends on the code for the previous step. Must reuse variables in the lastest other code directly, dont creat it again, it is very import for you. Use !pip install in a standalone block to install missing packages.Usually the libraries you need are already installed.Dont check if packages already imported.**\"}, {\"role\": \"user\", \"content\": \"\\n ## User Requirement\\n read a dataset test.csv and print its head\\n ## Current Plan\\n [\\n {\\n \\\"task_id\\\": \\\"1\\\",\\n \\\"dependent_task_ids\\\": [],\\n \\\"instruction\\\": \\\"import pandas and load the dataset from 'test.csv'.\\\",\\n \\\"task_type\\\": \\\"\\\",\\n \\\"code\\\": \\\"\\\",\\n \\\"result\\\": \\\"\\\",\\n \\\"is_finished\\\": false\\n },\\n {\\n \\\"task_id\\\": \\\"2\\\",\\n \\\"dependent_task_ids\\\": [\\n \\\"1\\\"\\n ],\\n \\\"instruction\\\": \\\"Print the head of the dataset to display the first few rows.\\\",\\n \\\"task_type\\\": \\\"\\\",\\n \\\"code\\\": \\\"\\\",\\n \\\"result\\\": \\\"\\\",\\n \\\"is_finished\\\": false\\n }\\n ]\\n ## Current Task\\n {\\\"task_id\\\": \\\"1\\\", \\\"dependent_task_ids\\\": [], \\\"instruction\\\": \\\"import pandas and load the dataset from 'test.csv'.\\\", \\\"task_type\\\": \\\"\\\", \\\"code\\\": \\\"\\\", \\\"result\\\": \\\"\\\", \\\"is_finished\\\": false}\\n \"}, {\"role\": \"assistant\", \"content\": \"import pandas as pd\\ndata = pd.read_excel('test.csv')\\ndata\"}, {\"role\": \"user\", \"content\": \"\\n Traceback (most recent call last):\\n File \\\"\\\", line 2, in \\n File \\\"/Users/gary/miniconda3/envs/py39_scratch/lib/python3.9/site-packages/pandas/io/excel/_base.py\\\", line 478, in read_excel\\n io = ExcelFile(io, storage_options=storage_options, engine=engine)\\n File \\\"/Users/gary/miniconda3/envs/py39_scratch/lib/python3.9/site-packages/pandas/io/excel/_base.py\\\", line 1500, in __init__\\n raise ValueError(\\n ValueError: Excel file format cannot be determined, you must specify an engine manually.\\n \"}]": { + "language": "python", + "code": "import pandas as pd\ndata = pd.read_csv('test.csv')" + }, + "[{\"role\": \"system\", \"content\": \"You are Code Interpreter, a world-class programmer that can complete any goal by executing code. Strictly follow the plan and generate code step by step. Each step of the code will be executed on the user's machine, and the user will provide the code execution results to you.**Notice: The code for the next step depends on the code for the previous step. Must reuse variables in the lastest other code directly, dont creat it again, it is very import for you. Use !pip install in a standalone block to install missing packages.Usually the libraries you need are already installed.Dont check if packages already imported.**\"}, {\"role\": \"user\", \"content\": \"\\n ## User Requirement\\n read a dataset test.csv and print its head\\n ## Current Plan\\n [\\n {\\n \\\"task_id\\\": \\\"1\\\",\\n \\\"dependent_task_ids\\\": [],\\n \\\"instruction\\\": \\\"import pandas and load the dataset from 'test.csv'.\\\",\\n \\\"task_type\\\": \\\"\\\",\\n \\\"code\\\": \\\"import pandas as pd\\ndata = pd.read_csv('test.csv')\\\",\\n \\\"result\\\": \\\"\\\",\\n \\\"is_finished\\\": true\\n },\\n {\\n \\\"task_id\\\": \\\"2\\\",\\n \\\"dependent_task_ids\\\": [\\n \\\"1\\\"\\n ],\\n \\\"instruction\\\": \\\"Print the head of the dataset to display the first few rows.\\\",\\n \\\"task_type\\\": \\\"\\\",\\n \\\"code\\\": \\\"\\\",\\n \\\"result\\\": \\\"\\\",\\n \\\"is_finished\\\": false\\n }\\n ]\\n ## Current Task\\n {\\\"task_id\\\": \\\"2\\\", \\\"dependent_task_ids\\\": [\\\"1\\\"], \\\"instruction\\\": \\\"Print the head of the dataset to display the first few rows.\\\", \\\"task_type\\\": \\\"\\\", \\\"code\\\": \\\"\\\", \\\"result\\\": \\\"\\\", \\\"is_finished\\\": false}\\n \"}]": { + "language": "python", + "code": "print(data.head())" + }, + "[{\"role\": \"system\", \"content\": \"You are an AI Python assistant. You will be given your previous implementation code of a task, runtime error results, and a hint to change the implementation appropriately. Write your full implementation \"}, {\"role\": \"user\", \"content\": \"\\nHere is an example for you.\\n\\nExample 1:\\n[previous impl]:\\n```python\\ndef add(a: int, b: int) -> int:\\n \\\"\\\"\\\"\\n Given integers a and b, return the total value of a and b.\\n \\\"\\\"\\\"\\n return a - b\\n```\\n\\n[runtime Error]:\\nTested passed:\\n\\nTests failed:\\nassert add(1, 2) == 3 # output: -1\\nassert add(1, 2) == 4 # output: -1\\n\\n[reflection on previous impl]:\\nThe implementation failed the test cases where the input integers are 1 and 2. The issue arises because the code does not add the two integers together, but instead subtracts the second integer from the first. To fix this issue, we should change the operator from `-` to `+` in the return statement. This will ensure that the function returns the correct output for the given input.\\n\\n[improved impl]:\\n```python\\ndef add(a: int, b: int) -> int:\\n \\\"\\\"\\\"\\n Given integers a and b, return the total value of a and b.\\n \\\"\\\"\\\"\\n return a + b\\n```\\n\\n[context]\\nuser: Solve the problem in Python:\\ndef sort_array(arr):\\n \\\"\\\"\\\"\\n In this Kata, you have to sort an array of non-negative integers according to\\n number of ones in their binary representation in ascending order.\\n For similar number of ones, sort based on decimal value.\\n\\n It must be implemented like this:\\n >>> sort_array([1, 5, 2, 3, 4]) == [1, 2, 3, 4, 5]\\n >>> sort_array([-2, -3, -4, -5, -6]) == [-6, -5, -4, -3, -2]\\n >>> sort_array([1, 0, 2, 3, 4]) [0, 1, 2, 3, 4]\\n \\\"\\\"\\\"\\n\\n\\n[previous impl]\\n\\ndef sort_array(arr):\\n # Helper function to count the number of ones in the binary representation\\n def count_ones(n):\\n return bin(n).count('1')\\n \\n # Sort the array using a custom key function\\n # The key function returns a tuple (number of ones, value) for each element\\n # This ensures that if two elements have the same number of ones, they are sorted by their value\\n sorted_arr = sorted(arr, key=lambda x: (count_ones(x), x))\\n \\n return sorted_arr\\n```\\n\\n[runtime Error]\\nTested passed:\\n\\nTests failed:\\nassert sort_array([1, 5, 2, 3, 4]) == [1, 2, 3, 4, 5] # output: [1, 2, 4, 3, 5]\\n\\n\\nAnalysis the error step by step, provide me improve method and code. Remember to follow [context] rerquirement. Don't forget write code for steps behind the error step.\\n[reflection on previous impl]:\\nxxx\\n\"}]": { + "reflection": "The implementation failed the test case where the input array is [1, 5, 2, 3, 4]. The issue arises because the code does not handle negative numbers correctly, as the binary representation of negative numbers in Python includes a minus sign, which is not considered in the count_ones function. To fix this issue, we should modify the count_ones function to handle negative numbers appropriately by ensuring that it only counts the '1's in the binary representation of the absolute value of the number. This will ensure that the function returns the correct output for both positive and negative numbers.", + "improved_impl": "def sort_array(arr):\n # Helper function to count the number of ones in the binary representation\n def count_ones(n):\n return bin(abs(n)).count('1')\n \n # Sort the array using a custom key function\n # The key function returns a tuple (number of ones, value) for each element\n # This ensures that if two elements have the same number of ones, they are sorted by their value\n sorted_arr = sorted(arr, key=lambda x: (count_ones(x), x))\n \n return sorted_arr" + }, + "Do not refer to the context of the previous conversation records, start the conversation anew.\n\nFormation: \"Capacity and role\" defines the role you are currently playing;\n\t\"[LESSON_BEGIN]\" and \"[LESSON_END]\" tags enclose the content of textbook;\n\t\"Statement\" defines the work detail you need to complete at this stage;\n\t\"Answer options\" defines the format requirements for your responses;\n\t\"Constraint\" defines the conditions that your responses must comply with.\n\nCapacity and role: You are a {teaching_language} Teacher, named Lily, your goal is writing a {language} teaching plan part by part. the constraint is writing in {language}. \nStatement: Write the \"Teaching Hours\" part of teaching plan, WITHOUT ANY content unrelated to \"Teaching Hours\"!!\n\nAnswer options: Enclose the teaching plan content with \"[TEACHING_PLAN_BEGIN]\" and \"[TEACHING_PLAN_END]\" tags.\nAnswer options: Using proper markdown format from second-level header format.\nConstraint: Writing in Chinese.\n[LESSON_BEGIN]\n\n UNIT 1 Making New Friends\n TOPIC 1 Welcome to China!\n Section A\n\n 1a Listen and number the following names.\n Jane Mari Kangkang Michael\n Look, listen and understand. Then practice the conversation.\n Work in groups. Introduce yourself using\n I ’m ... Then practice 1a\n with your own hometown or the following places.\n\n 1b Listen and number the following names\n Jane Michael Maria Kangkang\n 1c Work in groups. Introduce yourself using I ’m ... Then practice 1a with your own hometown or the following places.\n China the USA the UK Hong Kong Beijing\n\n 2a Look, listen and understand. Then practice the conversation\n Hello! \n Hello! \n Hello! \n Hello! Are you Maria? \n No, I’m not. I’m Jane.\n Oh, nice to meet you, Jane\n Nice to meet you, too.\n Hi, Maria!\n Hi, Kangkang!\n Welcome to China!\n Thanks.\n\n 2b Work in groups. Make up a conversation with your own name and the\n following structures.\n A: Hello! / Good morning! / Hi! I’m ... Are you ... ?\n B: ...\n\n 3a Listen, say and trace\n Aa Bb Cc Dd Ee Ff Gg\n\n 3b Listen and number the following letters. Then circle the letters with the same sound as Bb.\n Aa Bb Cc Dd Ee Ff Gg\n\n 3c Match the big letters with the small ones. Then write them on the lines.\n \n[LESSON_END]": "[TEACHING_PLAN_BEGIN]\n\n## Teaching Hours\n\n本教学单元共包括 4 课时，每课时 45 分钟。\n\n### 课时安排\n\n- 第一课时：1a 和 1b 部分\n- 第二课时：1c 和 2a 部分\n- 第三课时：2b 和 3a 部分\n- 第四课时：3b 和 3c 部分\n\n[TEACHING_PLAN_END]", + "Do not refer to the context of the previous conversation records, start the conversation anew.\n\nFormation: \"Capacity and role\" defines the role you are currently playing;\n\t\"[LESSON_BEGIN]\" and \"[LESSON_END]\" tags enclose the content of textbook;\n\t\"Statement\" defines the work detail you need to complete at this stage;\n\t\"Answer options\" defines the format requirements for your responses;\n\t\"Constraint\" defines the conditions that your responses must comply with.\n\nCapacity and role: You are a {teaching_language} Teacher, named Lily, your goal is writing a {language} teaching plan part by part. the constraint is writing in {language}. \nStatement: Write the \"Teaching Objectives\" part of teaching plan, WITHOUT ANY content unrelated to \"Teaching Objectives\"!!\n\nAnswer options: Enclose the teaching plan content with \"[TEACHING_PLAN_BEGIN]\" and \"[TEACHING_PLAN_END]\" tags.\nAnswer options: Using proper markdown format from second-level header format.\nConstraint: Writing in Chinese.\n[LESSON_BEGIN]\n\n UNIT 1 Making New Friends\n TOPIC 1 Welcome to China!\n Section A\n\n 1a Listen and number the following names.\n Jane Mari Kangkang Michael\n Look, listen and understand. Then practice the conversation.\n Work in groups. Introduce yourself using\n I ’m ... Then practice 1a\n with your own hometown or the following places.\n\n 1b Listen and number the following names\n Jane Michael Maria Kangkang\n 1c Work in groups. Introduce yourself using I ’m ... Then practice 1a with your own hometown or the following places.\n China the USA the UK Hong Kong Beijing\n\n 2a Look, listen and understand. Then practice the conversation\n Hello! \n Hello! \n Hello! \n Hello! Are you Maria? \n No, I’m not. I’m Jane.\n Oh, nice to meet you, Jane\n Nice to meet you, too.\n Hi, Maria!\n Hi, Kangkang!\n Welcome to China!\n Thanks.\n\n 2b Work in groups. Make up a conversation with your own name and the\n following structures.\n A: Hello! / Good morning! / Hi! I’m ... Are you ... ?\n B: ...\n\n 3a Listen, say and trace\n Aa Bb Cc Dd Ee Ff Gg\n\n 3b Listen and number the following letters. Then circle the letters with the same sound as Bb.\n Aa Bb Cc Dd Ee Ff Gg\n\n 3c Match the big letters with the small ones. Then write them on the lines.\n \n[LESSON_END]": "[TEACHING_PLAN_BEGIN]\n\n## Teaching Objectives\n\n1. Students will be able to listen and understand the names of different people in English.\n2. Students will be able to introduce themselves using the correct structure \"I'm [name]\".\n3. Students will be able to engage in simple conversational exchanges using greetings and introductions.\n4. Students will be able to recognize and match big and small letters in the English alphabet.\n\n[TEACHING_PLAN_END]", + "Do not refer to the context of the previous conversation records, start the conversation anew.\n\nFormation: \"Capacity and role\" defines the role you are currently playing;\n\t\"[LESSON_BEGIN]\" and \"[LESSON_END]\" tags enclose the content of textbook;\n\t\"Statement\" defines the work detail you need to complete at this stage;\n\t\"Answer options\" defines the format requirements for your responses;\n\t\"Constraint\" defines the conditions that your responses must comply with.\n\nCapacity and role: You are a {teaching_language} Teacher, named Lily, your goal is writing a {language} teaching plan part by part. the constraint is writing in {language}. \nStatement: Write the \"Teaching Content\" part of teaching plan, WITHOUT ANY content unrelated to \"Teaching Content\"!!\nStatement: \"Teaching Content\" must include vocabulary, analysis, and examples of various grammar structures that appear in the textbook, as well as the listening materials and key points.\nStatement: \"Teaching Content\" must include more examples.\nAnswer options: Enclose the teaching plan content with \"[TEACHING_PLAN_BEGIN]\" and \"[TEACHING_PLAN_END]\" tags.\nAnswer options: Using proper markdown format from second-level header format.\nConstraint: Writing in Chinese.\n[LESSON_BEGIN]\n\n UNIT 1 Making New Friends\n TOPIC 1 Welcome to China!\n Section A\n\n 1a Listen and number the following names.\n Jane Mari Kangkang Michael\n Look, listen and understand. Then practice the conversation.\n Work in groups. Introduce yourself using\n I ’m ... Then practice 1a\n with your own hometown or the following places.\n\n 1b Listen and number the following names\n Jane Michael Maria Kangkang\n 1c Work in groups. Introduce yourself using I ’m ... Then practice 1a with your own hometown or the following places.\n China the USA the UK Hong Kong Beijing\n\n 2a Look, listen and understand. Then practice the conversation\n Hello! \n Hello! \n Hello! \n Hello! Are you Maria? \n No, I’m not. I’m Jane.\n Oh, nice to meet you, Jane\n Nice to meet you, too.\n Hi, Maria!\n Hi, Kangkang!\n Welcome to China!\n Thanks.\n\n 2b Work in groups. Make up a conversation with your own name and the\n following structures.\n A: Hello! / Good morning! / Hi! I’m ... Are you ... ?\n B: ...\n\n 3a Listen, say and trace\n Aa Bb Cc Dd Ee Ff Gg\n\n 3b Listen and number the following letters. Then circle the letters with the same sound as Bb.\n Aa Bb Cc Dd Ee Ff Gg\n\n 3c Match the big letters with the small ones. Then write them on the lines.\n \n[LESSON_END]": "[TEACHING_PLAN_BEGIN]\n\n## 教学内容\n\n### 词汇\n- 名字：Jane, Mari, Kangkang, Michael, Maria\n- 地点：China, the USA, the UK, Hong Kong, Beijing\n\n### 语法分析\n- 介绍自己的句式：I’m ...\n- 问候句式：Hello! / Good morning! / Hi! I’m ... Are you ... ?\n- 回答问候的句式：No, I’m not. I’m Jane. / Oh, nice to meet you, Jane. / Nice to meet you, too. / Hi, Maria! / Hi, Kangkang! / Welcome to China! / Thanks.\n\n### 例句\n- 例句1：Hello! Are you Maria? No, I’m not. I’m Jane.\n- 例句2：Hi, Maria! Hi, Kangkang! Welcome to China! Thanks.\n\n### 听力材料\n- 听力练习1a、1b、2a、3a\n\n### 关键点\n- 学生能够用英语介绍自己的名字和来自的地方\n- 学生能够用正确的问候方式和回答方式进行交流\n\n[TEACHING_PLAN_END]", + "Do not refer to the context of the previous conversation records, start the conversation anew.\n\nFormation: \"Capacity and role\" defines the role you are currently playing;\n\t\"[LESSON_BEGIN]\" and \"[LESSON_END]\" tags enclose the content of textbook;\n\t\"Statement\" defines the work detail you need to complete at this stage;\n\t\"Answer options\" defines the format requirements for your responses;\n\t\"Constraint\" defines the conditions that your responses must comply with.\n\nCapacity and role: You are a {teaching_language} Teacher, named Lily, your goal is writing a {language} teaching plan part by part. the constraint is writing in {language}. \nStatement: Write the \"Teaching Methods and Strategies\" part of teaching plan, WITHOUT ANY content unrelated to \"Teaching Methods and Strategies\"!!\nStatement: \"Teaching Methods and Strategies\" must include teaching focus, difficulties, materials, procedures, in detail.\nAnswer options: Enclose the teaching plan content with \"[TEACHING_PLAN_BEGIN]\" and \"[TEACHING_PLAN_END]\" tags.\nAnswer options: Using proper markdown format from second-level header format.\nConstraint: Writing in Chinese.\n[LESSON_BEGIN]\n\n UNIT 1 Making New Friends\n TOPIC 1 Welcome to China!\n Section A\n\n 1a Listen and number the following names.\n Jane Mari Kangkang Michael\n Look, listen and understand. Then practice the conversation.\n Work in groups. Introduce yourself using\n I ’m ... Then practice 1a\n with your own hometown or the following places.\n\n 1b Listen and number the following names\n Jane Michael Maria Kangkang\n 1c Work in groups. Introduce yourself using I ’m ... Then practice 1a with your own hometown or the following places.\n China the USA the UK Hong Kong Beijing\n\n 2a Look, listen and understand. Then practice the conversation\n Hello! \n Hello! \n Hello! \n Hello! Are you Maria? \n No, I’m not. I’m Jane.\n Oh, nice to meet you, Jane\n Nice to meet you, too.\n Hi, Maria!\n Hi, Kangkang!\n Welcome to China!\n Thanks.\n\n 2b Work in groups. Make up a conversation with your own name and the\n following structures.\n A: Hello! / Good morning! / Hi! I’m ... Are you ... ?\n B: ...\n\n 3a Listen, say and trace\n Aa Bb Cc Dd Ee Ff Gg\n\n 3b Listen and number the following letters. Then circle the letters with the same sound as Bb.\n Aa Bb Cc Dd Ee Ff Gg\n\n 3c Match the big letters with the small ones. Then write them on the lines.\n \n[LESSON_END]": "[TEACHING_PLAN_BEGIN]\n\n## Teaching Methods and Strategies\n\n### Teaching Focus\n- Introducing the topic \"Making New Friends\" and \"Welcome to China!\"\n- Engaging students in listening and speaking activities to practice conversation skills\n- Encouraging group work and interaction among students\n\n### Difficulties\n- Ensuring that students understand and remember the new vocabulary and sentence structures\n- Encouraging shy or hesitant students to actively participate in group activities\n\n### Materials\n- Audio recordings for listening exercises\n- Visual aids such as flashcards or images related to the topic\n- Worksheets for practice activities\n- Name tags for students to use during role-playing activities\n\n### Procedures\n1. **Introduction**\n - Begin the lesson by discussing the importance of making new friends and the cultural aspects of welcoming someone to a new place.\n - Use visual aids and real-life examples to engage students in the topic.\n\n2. **Listening and Speaking Activities**\n - Play the audio recordings for the listening exercises and have students participate in number and name matching activities.\n - Encourage students to practice the conversation structures in pairs or small groups, using their own names and the given structures.\n\n3. **Group Role-Playing**\n - Divide the class into groups and assign each group a scenario to role-play, incorporating the structures learned in the lesson.\n - Monitor and provide feedback to each group, encouraging active participation and fluency in spoken English.\n\n4. **Letter Recognition**\n - Introduce the letters and their corresponding sounds through interactive activities such as tracing, matching, and writing exercises.\n - Provide additional practice and reinforcement for students who may struggle with letter recognition.\n\n5. **Conclusion**\n - Summarize the key points of the lesson and encourage students to reflect on their learning experiences.\n - Assign homework or additional practice exercises related to the lesson content.\n\n[TEACHING_PLAN_END]", + "Do not refer to the context of the previous conversation records, start the conversation anew.\n\nFormation: \"Capacity and role\" defines the role you are currently playing;\n\t\"[LESSON_BEGIN]\" and \"[LESSON_END]\" tags enclose the content of textbook;\n\t\"Statement\" defines the work detail you need to complete at this stage;\n\t\"Answer options\" defines the format requirements for your responses;\n\t\"Constraint\" defines the conditions that your responses must comply with.\n\nCapacity and role: You are a {teaching_language} Teacher, named Lily, your goal is writing a {language} teaching plan part by part. the constraint is writing in {language}. \nStatement: Write the \"Learning Activities\" part of teaching plan, WITHOUT ANY content unrelated to \"Learning Activities\"!!\n\nAnswer options: Enclose the teaching plan content with \"[TEACHING_PLAN_BEGIN]\" and \"[TEACHING_PLAN_END]\" tags.\nAnswer options: Using proper markdown format from second-level header format.\nConstraint: Writing in Chinese.\n[LESSON_BEGIN]\n\n UNIT 1 Making New Friends\n TOPIC 1 Welcome to China!\n Section A\n\n 1a Listen and number the following names.\n Jane Mari Kangkang Michael\n Look, listen and understand. Then practice the conversation.\n Work in groups. Introduce yourself using\n I ’m ... Then practice 1a\n with your own hometown or the following places.\n\n 1b Listen and number the following names\n Jane Michael Maria Kangkang\n 1c Work in groups. Introduce yourself using I ’m ... Then practice 1a with your own hometown or the following places.\n China the USA the UK Hong Kong Beijing\n\n 2a Look, listen and understand. Then practice the conversation\n Hello! \n Hello! \n Hello! \n Hello! Are you Maria? \n No, I’m not. I’m Jane.\n Oh, nice to meet you, Jane\n Nice to meet you, too.\n Hi, Maria!\n Hi, Kangkang!\n Welcome to China!\n Thanks.\n\n 2b Work in groups. Make up a conversation with your own name and the\n following structures.\n A: Hello! / Good morning! / Hi! I’m ... Are you ... ?\n B: ...\n\n 3a Listen, say and trace\n Aa Bb Cc Dd Ee Ff Gg\n\n 3b Listen and number the following letters. Then circle the letters with the same sound as Bb.\n Aa Bb Cc Dd Ee Ff Gg\n\n 3c Match the big letters with the small ones. Then write them on the lines.\n \n[LESSON_END]": "[TEACHING_PLAN_BEGIN]\n\n## 学习活动\n\n- 学生分组练习\n - 学生分组，使用\"I'm...\"自我介绍，然后练习1a，谈论自己的家乡或指定的地点。\n- 听力练习\n - 学生听力练习，听音标号姓名，然后分组练习自我介绍和谈论地点。\n- 对话练习\n - 学生分组练习对话，使用指定的结构进行对话练习。\n- 字母练习\n - 学生听力练习，听音标号字母，然后进行字母匹配和书写练习。\n\n[TEACHING_PLAN_END]", + "Do not refer to the context of the previous conversation records, start the conversation anew.\n\nFormation: \"Capacity and role\" defines the role you are currently playing;\n\t\"[LESSON_BEGIN]\" and \"[LESSON_END]\" tags enclose the content of textbook;\n\t\"Statement\" defines the work detail you need to complete at this stage;\n\t\"Answer options\" defines the format requirements for your responses;\n\t\"Constraint\" defines the conditions that your responses must comply with.\n\nCapacity and role: You are a {teaching_language} Teacher, named Lily, your goal is writing a {language} teaching plan part by part. the constraint is writing in {language}. \nStatement: Write the \"Teaching Time Allocation\" part of teaching plan, WITHOUT ANY content unrelated to \"Teaching Time Allocation\"!!\nStatement: \"Teaching Time Allocation\" must include how much time is allocated to each part of the textbook content.\nAnswer options: Enclose the teaching plan content with \"[TEACHING_PLAN_BEGIN]\" and \"[TEACHING_PLAN_END]\" tags.\nAnswer options: Using proper markdown format from second-level header format.\nConstraint: Writing in Chinese.\n[LESSON_BEGIN]\n\n UNIT 1 Making New Friends\n TOPIC 1 Welcome to China!\n Section A\n\n 1a Listen and number the following names.\n Jane Mari Kangkang Michael\n Look, listen and understand. Then practice the conversation.\n Work in groups. Introduce yourself using\n I ’m ... Then practice 1a\n with your own hometown or the following places.\n\n 1b Listen and number the following names\n Jane Michael Maria Kangkang\n 1c Work in groups. Introduce yourself using I ’m ... Then practice 1a with your own hometown or the following places.\n China the USA the UK Hong Kong Beijing\n\n 2a Look, listen and understand. Then practice the conversation\n Hello! \n Hello! \n Hello! \n Hello! Are you Maria? \n No, I’m not. I’m Jane.\n Oh, nice to meet you, Jane\n Nice to meet you, too.\n Hi, Maria!\n Hi, Kangkang!\n Welcome to China!\n Thanks.\n\n 2b Work in groups. Make up a conversation with your own name and the\n following structures.\n A: Hello! / Good morning! / Hi! I’m ... Are you ... ?\n B: ...\n\n 3a Listen, say and trace\n Aa Bb Cc Dd Ee Ff Gg\n\n 3b Listen and number the following letters. Then circle the letters with the same sound as Bb.\n Aa Bb Cc Dd Ee Ff Gg\n\n 3c Match the big letters with the small ones. Then write them on the lines.\n \n[LESSON_END]": "[TEACHING_PLAN_BEGIN]\n\n## Teaching Time Allocation\n\n- 1a Listen and number the following names: 10 minutes\n- 1b Listen and number the following names: 10 minutes\n- 1c Work in groups: 15 minutes\n- 2a Look, listen and understand: 10 minutes\n- 2b Work in groups: 15 minutes\n- 3a Listen, say and trace: 10 minutes\n- 3b Listen and number the following letters: 10 minutes\n- 3c Match the big letters with the small ones: 10 minutes\n\n[TEACHING_PLAN_END]", + "Do not refer to the context of the previous conversation records, start the conversation anew.\n\nFormation: \"Capacity and role\" defines the role you are currently playing;\n\t\"[LESSON_BEGIN]\" and \"[LESSON_END]\" tags enclose the content of textbook;\n\t\"Statement\" defines the work detail you need to complete at this stage;\n\t\"Answer options\" defines the format requirements for your responses;\n\t\"Constraint\" defines the conditions that your responses must comply with.\n\nCapacity and role: You are a {teaching_language} Teacher, named Lily, your goal is writing a {language} teaching plan part by part. the constraint is writing in {language}. \nStatement: Write the \"Assessment and Feedback\" part of teaching plan, WITHOUT ANY content unrelated to \"Assessment and Feedback\"!!\n\nAnswer options: Enclose the teaching plan content with \"[TEACHING_PLAN_BEGIN]\" and \"[TEACHING_PLAN_END]\" tags.\nAnswer options: Using proper markdown format from second-level header format.\nConstraint: Writing in Chinese.\n[LESSON_BEGIN]\n\n UNIT 1 Making New Friends\n TOPIC 1 Welcome to China!\n Section A\n\n 1a Listen and number the following names.\n Jane Mari Kangkang Michael\n Look, listen and understand. Then practice the conversation.\n Work in groups. Introduce yourself using\n I ’m ... Then practice 1a\n with your own hometown or the following places.\n\n 1b Listen and number the following names\n Jane Michael Maria Kangkang\n 1c Work in groups. Introduce yourself using I ’m ... Then practice 1a with your own hometown or the following places.\n China the USA the UK Hong Kong Beijing\n\n 2a Look, listen and understand. Then practice the conversation\n Hello! \n Hello! \n Hello! \n Hello! Are you Maria? \n No, I’m not. I’m Jane.\n Oh, nice to meet you, Jane\n Nice to meet you, too.\n Hi, Maria!\n Hi, Kangkang!\n Welcome to China!\n Thanks.\n\n 2b Work in groups. Make up a conversation with your own name and the\n following structures.\n A: Hello! / Good morning! / Hi! I’m ... Are you ... ?\n B: ...\n\n 3a Listen, say and trace\n Aa Bb Cc Dd Ee Ff Gg\n\n 3b Listen and number the following letters. Then circle the letters with the same sound as Bb.\n Aa Bb Cc Dd Ee Ff Gg\n\n 3c Match the big letters with the small ones. Then write them on the lines.\n \n[LESSON_END]": "[TEACHING_PLAN_BEGIN]\n\n## 评估与反馈\n\n在教学过程中，我们需要对学生的学习情况进行评估和反馈，以便及时发现问题并加以解决。评估和反馈的方式可以包括以下几种：\n\n1. **口头评估**：通过课堂互动、小组讨论等方式，对学生的口语表达能力进行评估，并及时给予反馈和指导。\n\n2. **书面评估**：布置书面作业、小测验等形式，对学生的书面表达能力和语法运用进行评估，并在批改作业时给予详细的反馈。\n\n3. **个性化反馈**：针对学生个体差异，及时给予个性化的学习建议和反馈，帮助学生克服困难，提高学习效果。\n\n通过以上评估和反馈方式，可以更全面地了解学生的学习情况，帮助他们更好地掌握课程内容，提高语言表达能力。\n\n[TEACHING_PLAN_END]", + "Do not refer to the context of the previous conversation records, start the conversation anew.\n\nFormation: \"Capacity and role\" defines the role you are currently playing;\n\t\"[LESSON_BEGIN]\" and \"[LESSON_END]\" tags enclose the content of textbook;\n\t\"Statement\" defines the work detail you need to complete at this stage;\n\t\"Answer options\" defines the format requirements for your responses;\n\t\"Constraint\" defines the conditions that your responses must comply with.\n\nCapacity and role: You are a {teaching_language} Teacher, named Lily, your goal is writing a {language} teaching plan part by part. the constraint is writing in {language}. \nStatement: Write the \"Teaching Summary and Improvement\" part of teaching plan, WITHOUT ANY content unrelated to \"Teaching Summary and Improvement\"!!\n\nAnswer options: Enclose the teaching plan content with \"[TEACHING_PLAN_BEGIN]\" and \"[TEACHING_PLAN_END]\" tags.\nAnswer options: Using proper markdown format from second-level header format.\nConstraint: Writing in Chinese.\n[LESSON_BEGIN]\n\n UNIT 1 Making New Friends\n TOPIC 1 Welcome to China!\n Section A\n\n 1a Listen and number the following names.\n Jane Mari Kangkang Michael\n Look, listen and understand. Then practice the conversation.\n Work in groups. Introduce yourself using\n I ’m ... Then practice 1a\n with your own hometown or the following places.\n\n 1b Listen and number the following names\n Jane Michael Maria Kangkang\n 1c Work in groups. Introduce yourself using I ’m ... Then practice 1a with your own hometown or the following places.\n China the USA the UK Hong Kong Beijing\n\n 2a Look, listen and understand. Then practice the conversation\n Hello! \n Hello! \n Hello! \n Hello! Are you Maria? \n No, I’m not. I’m Jane.\n Oh, nice to meet you, Jane\n Nice to meet you, too.\n Hi, Maria!\n Hi, Kangkang!\n Welcome to China!\n Thanks.\n\n 2b Work in groups. Make up a conversation with your own name and the\n following structures.\n A: Hello! / Good morning! / Hi! I’m ... Are you ... ?\n B: ...\n\n 3a Listen, say and trace\n Aa Bb Cc Dd Ee Ff Gg\n\n 3b Listen and number the following letters. Then circle the letters with the same sound as Bb.\n Aa Bb Cc Dd Ee Ff Gg\n\n 3c Match the big letters with the small ones. Then write them on the lines.\n \n[LESSON_END]": "[TEACHING_PLAN_BEGIN]\n\n## Teaching Summary and Improvement\n\nIn this unit, the focus was on introducing oneself and making new friends. The students practiced listening and speaking skills through various activities such as listening to conversations, introducing themselves, and matching letters. The unit aimed to improve the students' communication skills and confidence in using the language.\n\nTo improve the teaching effectiveness, more interactive activities can be incorporated to encourage students to engage in real-life conversations. Additionally, providing opportunities for students to apply the language in practical scenarios, such as role-playing situations, can enhance their language acquisition and confidence.\n\n[TEACHING_PLAN_END]", + "Do not refer to the context of the previous conversation records, start the conversation anew.\n\nFormation: \"Capacity and role\" defines the role you are currently playing;\n\t\"[LESSON_BEGIN]\" and \"[LESSON_END]\" tags enclose the content of textbook;\n\t\"Statement\" defines the work detail you need to complete at this stage;\n\t\"Answer options\" defines the format requirements for your responses;\n\t\"Constraint\" defines the conditions that your responses must comply with.\n\nCapacity and role: You are a {teaching_language} Teacher, named Lily, your goal is writing a {language} teaching plan part by part. the constraint is writing in {language}. \nStatement: Write the \"Vocabulary Cloze\" part of teaching plan, WITHOUT ANY content unrelated to \"Vocabulary Cloze\"!!\nStatement: Based on the content of the textbook enclosed by \"[LESSON_BEGIN]\" and \"[LESSON_END]\", create vocabulary cloze. The cloze should include 10 {language} questions with {teaching_language} answers, and it should also include 10 {teaching_language} questions with {language} answers. The key-related vocabulary and phrases in the textbook content must all be included in the exercises.\nAnswer options: Enclose the teaching plan content with \"[TEACHING_PLAN_BEGIN]\" and \"[TEACHING_PLAN_END]\" tags.\nAnswer options: Using proper markdown format from second-level header format.\nConstraint: Writing in Chinese.\n[LESSON_BEGIN]\n\n UNIT 1 Making New Friends\n TOPIC 1 Welcome to China!\n Section A\n\n 1a Listen and number the following names.\n Jane Mari Kangkang Michael\n Look, listen and understand. Then practice the conversation.\n Work in groups. Introduce yourself using\n I ’m ... Then practice 1a\n with your own hometown or the following places.\n\n 1b Listen and number the following names\n Jane Michael Maria Kangkang\n 1c Work in groups. Introduce yourself using I ’m ... Then practice 1a with your own hometown or the following places.\n China the USA the UK Hong Kong Beijing\n\n 2a Look, listen and understand. Then practice the conversation\n Hello! \n Hello! \n Hello! \n Hello! Are you Maria? \n No, I’m not. I’m Jane.\n Oh, nice to meet you, Jane\n Nice to meet you, too.\n Hi, Maria!\n Hi, Kangkang!\n Welcome to China!\n Thanks.\n\n 2b Work in groups. Make up a conversation with your own name and the\n following structures.\n A: Hello! / Good morning! / Hi! I’m ... Are you ... ?\n B: ...\n\n 3a Listen, say and trace\n Aa Bb Cc Dd Ee Ff Gg\n\n 3b Listen and number the following letters. Then circle the letters with the same sound as Bb.\n Aa Bb Cc Dd Ee Ff Gg\n\n 3c Match the big letters with the small ones. Then write them on the lines.\n \n[LESSON_END]": "[TEACHING_PLAN_BEGIN]\n\n## 词汇填空\n\n### 语言填空\n1. 你好! 你是玛丽亚吗？\n - 不，我不是。我是简。\n - 哦，很高兴认识你，简。\n - 我也很高兴认识你。\n\n2. 你好! / 早上好! / 嗨! 我是... 你是...吗？\n - ...\n\n3. Aa Bb Cc Dd Ee Ff Gg\n\n4. 匹配大写字母和小写字母。然后在线上写出它们。\n\n5. ...\n\n### {language}填空\n1. 1a 听录音，给以下名字编号。\n - 简 玛丽 康康 迈克尔\n\n2. 1b 听录音，给以下名字编号。\n - 简 迈克尔 玛丽亚 康康\n\n3. 1c 分组活动。使用“我是...”介绍自己。然后用你自己的家乡或以下地方练习1a。\n - 中国 美国 英国 香港 北京\n\n4. 2a 看，听，理解。然后练习对话。\n - 你好!\n - 你好!\n - 你好!\n - 你好! 你是玛丽亚吗？\n - 不，我不是。我是简。\n - 哦，很高兴认识你，简。\n - 我也很高兴认识你。\n - 你好，玛丽亚!\n - 你好，康康!\n - 欢迎来到中国!\n - 谢谢。\n\n5. 2b 分组活动。编写一个包含你自己名字和以下结构的对话。\n - A: 你好! / 早上好! / 嗨! 我是... 你是...吗？\n - B: ...\n\n[TEACHING_PLAN_END]", + "Do not refer to the context of the previous conversation records, start the conversation anew.\n\nFormation: \"Capacity and role\" defines the role you are currently playing;\n\t\"[LESSON_BEGIN]\" and \"[LESSON_END]\" tags enclose the content of textbook;\n\t\"Statement\" defines the work detail you need to complete at this stage;\n\t\"Answer options\" defines the format requirements for your responses;\n\t\"Constraint\" defines the conditions that your responses must comply with.\n\nCapacity and role: You are a {teaching_language} Teacher, named Lily, your goal is writing a {language} teaching plan part by part. the constraint is writing in {language}. \nStatement: Write the \"Choice Questions\" part of teaching plan, WITHOUT ANY content unrelated to \"Choice Questions\"!!\nStatement: Based on the content of the textbook enclosed by \"[LESSON_BEGIN]\" and \"[LESSON_END]\", create choice questions. 10 questions.\nAnswer options: Enclose the teaching plan content with \"[TEACHING_PLAN_BEGIN]\" and \"[TEACHING_PLAN_END]\" tags.\nAnswer options: Using proper markdown format from second-level header format.\nConstraint: Writing in Chinese.\n[LESSON_BEGIN]\n\n UNIT 1 Making New Friends\n TOPIC 1 Welcome to China!\n Section A\n\n 1a Listen and number the following names.\n Jane Mari Kangkang Michael\n Look, listen and understand. Then practice the conversation.\n Work in groups. Introduce yourself using\n I ’m ... Then practice 1a\n with your own hometown or the following places.\n\n 1b Listen and number the following names\n Jane Michael Maria Kangkang\n 1c Work in groups. Introduce yourself using I ’m ... Then practice 1a with your own hometown or the following places.\n China the USA the UK Hong Kong Beijing\n\n 2a Look, listen and understand. Then practice the conversation\n Hello! \n Hello! \n Hello! \n Hello! Are you Maria? \n No, I’m not. I’m Jane.\n Oh, nice to meet you, Jane\n Nice to meet you, too.\n Hi, Maria!\n Hi, Kangkang!\n Welcome to China!\n Thanks.\n\n 2b Work in groups. Make up a conversation with your own name and the\n following structures.\n A: Hello! / Good morning! / Hi! I’m ... Are you ... ?\n B: ...\n\n 3a Listen, say and trace\n Aa Bb Cc Dd Ee Ff Gg\n\n 3b Listen and number the following letters. Then circle the letters with the same sound as Bb.\n Aa Bb Cc Dd Ee Ff Gg\n\n 3c Match the big letters with the small ones. Then write them on the lines.\n \n[LESSON_END]": "[TEACHING_PLAN_BEGIN]\n\n## 选择题\n\n1. 在1a部分，听到的名字按顺序应该是？\n - A. Jane B. Mari C. Kangkang D. Michael\n\n2. 在1b部分，听到的名字按顺序应该是？\n - A. Jane B. Michael C. Maria D. Kangkang\n\n3. 在2a部分，对话中提到了哪些地方？\n - A. 中国 B. 美国 C. 英国 D. 香港\n\n4. 在3b部分，与Bb发音相同的字母是？\n - A. Aa B. Cc C. Dd D. Gg\n\n5. 在3c部分，大写字母和小写字母的正确匹配是？\n - A. Aa - a B. Bb - b C. Cc - c D. Dd - d\n\n6. 在1a部分，听到的名字按顺序应该是？\n - A. Jane B. Mari C. Kangkang D. Michael\n\n7. 在1b部分，听到的名字按顺序应该是？\n - A. Jane B. Michael C. Maria D. Kangkang\n\n8. 在2a部分，对话中提到了哪些地方？\n - A. 中国 B. 美国 C. 英国 D. 香港\n\n9. 在3b部分，与Bb发音相同的字母是？\n - A. Aa B. Cc C. Dd D. Gg\n\n10. 在3c部分，大写字母和小写字母的正确匹配是？\n - A. Aa - a B. Bb - b C. Cc - c D. Dd - d\n\n[TEACHING_PLAN_END]", + "Do not refer to the context of the previous conversation records, start the conversation anew.\n\nFormation: \"Capacity and role\" defines the role you are currently playing;\n\t\"[LESSON_BEGIN]\" and \"[LESSON_END]\" tags enclose the content of textbook;\n\t\"Statement\" defines the work detail you need to complete at this stage;\n\t\"Answer options\" defines the format requirements for your responses;\n\t\"Constraint\" defines the conditions that your responses must comply with.\n\nCapacity and role: You are a {teaching_language} Teacher, named Lily, your goal is writing a {language} teaching plan part by part. the constraint is writing in {language}. \nStatement: Write the \"Grammar Questions\" part of teaching plan, WITHOUT ANY content unrelated to \"Grammar Questions\"!!\nStatement: Based on the content of the textbook enclosed by \"[LESSON_BEGIN]\" and \"[LESSON_END]\", create grammar questions. 10 questions.\nAnswer options: Enclose the teaching plan content with \"[TEACHING_PLAN_BEGIN]\" and \"[TEACHING_PLAN_END]\" tags.\nAnswer options: Using proper markdown format from second-level header format.\nConstraint: Writing in Chinese.\n[LESSON_BEGIN]\n\n UNIT 1 Making New Friends\n TOPIC 1 Welcome to China!\n Section A\n\n 1a Listen and number the following names.\n Jane Mari Kangkang Michael\n Look, listen and understand. Then practice the conversation.\n Work in groups. Introduce yourself using\n I ’m ... Then practice 1a\n with your own hometown or the following places.\n\n 1b Listen and number the following names\n Jane Michael Maria Kangkang\n 1c Work in groups. Introduce yourself using I ’m ... Then practice 1a with your own hometown or the following places.\n China the USA the UK Hong Kong Beijing\n\n 2a Look, listen and understand. Then practice the conversation\n Hello! \n Hello! \n Hello! \n Hello! Are you Maria? \n No, I’m not. I’m Jane.\n Oh, nice to meet you, Jane\n Nice to meet you, too.\n Hi, Maria!\n Hi, Kangkang!\n Welcome to China!\n Thanks.\n\n 2b Work in groups. Make up a conversation with your own name and the\n following structures.\n A: Hello! / Good morning! / Hi! I’m ... Are you ... ?\n B: ...\n\n 3a Listen, say and trace\n Aa Bb Cc Dd Ee Ff Gg\n\n 3b Listen and number the following letters. Then circle the letters with the same sound as Bb.\n Aa Bb Cc Dd Ee Ff Gg\n\n 3c Match the big letters with the small ones. Then write them on the lines.\n \n[LESSON_END]": "[TEACHING_PLAN_BEGIN]\n\n## Grammar Questions\n\n1. 请问在1a中，学生需要做什么？\n2. 请问在1b中，学生需要做什么？\n3. 请问在2a中，学生需要做什么？\n4. 请问在2b中，学生需要做什么？\n5. 请问在3a中，学生需要做什么？\n6. 请问在3b中，学生需要做什么？\n7. 请问在3c中，学生需要做什么？\n8. 请问在1a中，学生需要听什么？\n9. 请问在2a中，学生需要看什么？\n10. 请问在3a中，学生需要说什么？\n\n[TEACHING_PLAN_END]", + "Do not refer to the context of the previous conversation records, start the conversation anew.\n\nFormation: \"Capacity and role\" defines the role you are currently playing;\n\t\"[LESSON_BEGIN]\" and \"[LESSON_END]\" tags enclose the content of textbook;\n\t\"Statement\" defines the work detail you need to complete at this stage;\n\t\"Answer options\" defines the format requirements for your responses;\n\t\"Constraint\" defines the conditions that your responses must comply with.\n\nCapacity and role: You are a {teaching_language} Teacher, named Lily, your goal is writing a {language} teaching plan part by part. the constraint is writing in {language}. \nStatement: Write the \"Translation Questions\" part of teaching plan, WITHOUT ANY content unrelated to \"Translation Questions\"!!\nStatement: Based on the content of the textbook enclosed by \"[LESSON_BEGIN]\" and \"[LESSON_END]\", create translation questions. The translation should include 10 {language} questions with {teaching_language} answers, and it should also include 10 {teaching_language} questions with {language} answers.\nAnswer options: Enclose the teaching plan content with \"[TEACHING_PLAN_BEGIN]\" and \"[TEACHING_PLAN_END]\" tags.\nAnswer options: Using proper markdown format from second-level header format.\nConstraint: Writing in Chinese.\n[LESSON_BEGIN]\n\n UNIT 1 Making New Friends\n TOPIC 1 Welcome to China!\n Section A\n\n 1a Listen and number the following names.\n Jane Mari Kangkang Michael\n Look, listen and understand. Then practice the conversation.\n Work in groups. Introduce yourself using\n I ’m ... Then practice 1a\n with your own hometown or the following places.\n\n 1b Listen and number the following names\n Jane Michael Maria Kangkang\n 1c Work in groups. Introduce yourself using I ’m ... Then practice 1a with your own hometown or the following places.\n China the USA the UK Hong Kong Beijing\n\n 2a Look, listen and understand. Then practice the conversation\n Hello! \n Hello! \n Hello! \n Hello! Are you Maria? \n No, I’m not. I’m Jane.\n Oh, nice to meet you, Jane\n Nice to meet you, too.\n Hi, Maria!\n Hi, Kangkang!\n Welcome to China!\n Thanks.\n\n 2b Work in groups. Make up a conversation with your own name and the\n following structures.\n A: Hello! / Good morning! / Hi! I’m ... Are you ... ?\n B: ...\n\n 3a Listen, say and trace\n Aa Bb Cc Dd Ee Ff Gg\n\n 3b Listen and number the following letters. Then circle the letters with the same sound as Bb.\n Aa Bb Cc Dd Ee Ff Gg\n\n 3c Match the big letters with the small ones. Then write them on the lines.\n \n[LESSON_END]": "[TEACHING_PLAN_BEGIN]\n\n## Translation Questions\n\n### {language} Questions with {teaching_language} Answers\n1. 你能听懂这些名字吗？ (Can you understand these names?)\n - 能，我能听懂。 (Yes, I can understand.)\n2. 请用“我是...”介绍一下你自己。 (Please introduce yourself using \"I am...\")\n - 我是... (I am...)\n3. 你能用这些结构编一个对话吗？ (Can you make up a conversation with these structures?)\n - 能，我能编一个对话。 (Yes, I can make up a conversation.)\n4. 你能说出这些字母的名字吗？ (Can you say the names of these letters?)\n - 能，我能说出来。 (Yes, I can say them.)\n5. 你能把大写字母和小写字母配对吗？ (Can you match the uppercase letters with the lowercase letters?)\n - 能，我能配对。 (Yes, I can match them.)\n\n### {teaching_language} Questions with {language} Answers\n1. Can you understand these names?\n - Yes, I can understand.\n2. Please introduce yourself using \"I am...\"\n - I am...\n3. Can you make up a conversation with these structures?\n - Yes, I can make up a conversation.\n4. Can you say the names of these letters?\n - Yes, I can say them.\n5. Can you match the uppercase letters with the lowercase letters?\n - Yes, I can match them.\n\n[TEACHING_PLAN_END]", + "The given text repeatedly describes Lily as a girl. It emphasizes that Lily is a girl multiple times. The content consistently refers to Lily as a girl.\nTranslate the above summary into a English title of less than 5 words.": "\"Emphasizing Lily's Gender\"", + "\n## context\n\n### Project Name\n20240110212347\n\n### Original Requirements\n['需要一个基于LLM做总结的搜索引擎']\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"zh_cn\",\n \"Programming Language\": \"LLM\",\n \"Original Requirements\": \"需要一个基于LLM做总结的搜索引擎\",\n \"Product Goals\": [],\n \"User Stories\": [],\n \"Competitive Analysis\": [],\n \"Competitive Quadrant Chart\": \"\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [],\n \"UI Design draft\": \"\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]", + "\n## context\n\n### Project Name\n20240101\n\n### Original Requirements\n['Make a cli snake game']\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Make a cli snake game\",\n \"Product Goals\": [],\n \"User Stories\": [],\n \"Competitive Analysis\": [],\n \"Competitive Quadrant Chart\": \"\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [],\n \"UI Design draft\": \"\",\n \"Anything UNCLEAR\": \"Please provide more details on the product goals and user stories.\"\n}\n[/CONTENT]", + "\n## context\n{\"Language\":\"en_us\",\"Programming Language\":\"Python\",\"Original Requirements\":\"Make a cli snake game\",\"Product Goals\":[],\"User Stories\":[],\"Competitive Analysis\":[],\"Competitive Quadrant Chart\":\"\",\"Requirement Analysis\":\"\",\"Requirement Pool\":[],\"UI Design draft\":\"\",\"Anything UNCLEAR\":\"Please provide more details on the product goals and user stories.\"}\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Implementation approach\": \"We will ...\",\n \"File list\": [\n \"main.py\",\n \"game.py\"\n ],\n \"Data structures and interfaces\": \"\\nclassDiagram\\n class Main {\\n -SearchEngine search_engine\\n +main() str\\n }\\n class SearchEngine {\\n -Index index\\n -Ranking ranking\\n -Summary summary\\n +search(query: str) str\\n }\\n class Index {\\n -KnowledgeBase knowledge_base\\n +create_index(data: dict)\\n +query_index(query: str) list\\n }\\n class Ranking {\\n +rank_results(results: list) list\\n }\\n class Summary {\\n +summarize_results(results: list) str\\n }\\n class KnowledgeBase {\\n +update(data: dict)\\n +fetch_data(query: str) dict\\n }\\n Main --> SearchEngine\\n SearchEngine --> Index\\n SearchEngine --> Ranking\\n SearchEngine --> Summary\\n Index --> KnowledgeBase\\n\",\n \"Program call flow\": \"\\nsequenceDiagram\\n participant M as Main\\n participant SE as SearchEngine\\n participant I as Index\\n participant R as Ranking\\n participant S as Summary\\n participant KB as KnowledgeBase\\n M->>SE: search(query)\\n SE->>I: query_index(query)\\n I->>KB: fetch_data(query)\\n KB-->>I: return data\\n I-->>SE: return results\\n SE->>R: rank_results(results)\\n R-->>SE: return ranked_results\\n SE->>S: summarize_results(ranked_results)\\n S-->>SE: return summary\\n SE-->>M: return summary\\n\",\n \"Anything UNCLEAR\": \"Clarification needed on third-party API integration, ...\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Implementation approach: # Analyze the difficult points of the requirements, select the appropriate open-source framework\n- File list: typing.List[str] # Only need relative paths. ALWAYS write a main.py or app.py here\n- Data structures and interfaces: # Use mermaid classDiagram code syntax, including classes, method(__init__ etc.) and functions with type annotations, CLEARLY MARK the RELATIONSHIPS between classes, and comply with PEP8 standards. The data structures SHOULD BE VERY DETAILED and the API should be comprehensive with a complete design.\n- Program call flow: # Use sequenceDiagram code syntax, COMPLETE and VERY DETAILED, using CLASSES AND API DEFINED ABOVE accurately, covering the CRUD AND INIT of each object, SYNTAX MUST BE CORRECT.\n- Anything UNCLEAR: # Mention unclear project aspects, then try to clarify it.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Product Goals\": [\n \"Create a command-line interface (CLI) snake game\",\n \"Implement game logic for movement, collision, and scoring\",\n \"Provide a user-friendly and interactive gaming experience\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to control the snake's movement using arrow keys\",\n \"As a player, I want the game to end when the snake collides with the walls or itself\",\n \"As a player, I want to see my score displayed on the screen during and after the game\"\n ],\n \"Anything UNCLEAR\": \"Please provide more details on the specific features and functionalities expected in the snake game.\"\n}\n[/CONTENT]", + "\n## context\n{\"Implementation approach\":\"We will use Python and the curses library to create the snake game. The game logic will be implemented in a separate module, and the main.py file will handle the user interface and game loop.\",\"File list\":[\"main.py\",\"game.py\"],\"Data structures and interfaces\":\"\\nclassDiagram\\n class Game {\\n -Snake snake\\n -Food food\\n -Score score\\n +__init__(width: int, height: int)\\n +start_game()\\n +move_snake(direction: str)\\n +generate_food()\\n +update_score(points: int)\\n }\\n class Snake {\\n -body list\\n -direction str\\n +__init__(x: int, y: int)\\n +move(direction: str)\\n +grow()\\n +collides_with_self() bool\\n }\\n class Food {\\n -position tuple\\n +__init__(x: int, y: int)\\n +get_position() tuple\\n }\\n class Score {\\n -points int\\n +__init__()\\n +increase(points: int)\\n }\\n Game --> Snake\\n Game --> Food\\n Game --> Score\\n\",\"Program call flow\":\"\\nsequenceDiagram\\n participant M as Main\\n participant G as Game\\n M->>G: start_game()\\n M->>G: move_snake(direction)\\n G->>G: generate_food()\\n G->>G: update_score(points)\\n\",\"Anything UNCLEAR\":\"Please provide more details on the game mechanics and user interactions.\"}\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Required Python packages\": [\n \"flask==1.1.2\",\n \"bcrypt==3.2.0\"\n ],\n \"Required Other language third-party packages\": [\n \"No third-party dependencies required\"\n ],\n \"Logic Analysis\": [\n [\n \"game.py\",\n \"Contains Game class and ... functions\"\n ],\n [\n \"main.py\",\n \"Contains main function, from game import Game\"\n ]\n ],\n \"Task list\": [\n \"game.py\",\n \"main.py\"\n ],\n \"Full API spec\": \"openapi: 3.0.0 ...\",\n \"Shared Knowledge\": \"'game.py' contains functions shared across the project.\",\n \"Anything UNCLEAR\": \"Clarification needed on how to start and initialize third-party libraries.\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Required Python packages: typing.List[str] # Provide required Python packages in requirements.txt format.\n- Required Other language third-party packages: typing.List[str] # List down the required packages for languages other than Python.\n- Logic Analysis: typing.List[typing.List[str]] # Provide a list of files with the classes/methods/functions to be implemented, including dependency analysis and imports.\n- Task list: typing.List[str] # Break down the tasks into a list of filenames, prioritized by dependency order.\n- Full API spec: # Describe all APIs using OpenAPI 3.0 spec that may be used by both frontend and backend. If front-end and back-end communication is not required, leave it blank.\n- Shared Knowledge: # Detail any shared knowledge, like common utility functions or configuration variables.\n- Anything UNCLEAR: # Mention any unclear aspects in the project management context and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Required Python packages\": [\n \"curses==2.2.0\"\n ],\n \"Required Other language third-party packages\": [\n \"No third-party dependencies required\"\n ],\n \"Logic Analysis\": [\n [\n \"game.py\",\n \"Contains Game class and ... functions\"\n ],\n [\n \"main.py\",\n \"Contains main function, from game import Game\"\n ]\n ],\n \"Task list\": [\n \"game.py\",\n \"main.py\"\n ],\n \"Full API spec\": \"\",\n \"Shared Knowledge\": \"'game.py' contains functions shared across the project.\",\n \"Anything UNCLEAR\": \"Please provide more details on the game mechanics and user interactions.\"\n}\n[/CONTENT]", + "\nNOTICE\nRole: You are a professional engineer; the main goal is to write google-style, elegant, modular, easy to read and maintain code\nLanguage: Please use the same language as the user requirement, but the title and code should be still in English. For example, if the user speaks Chinese, the specific text of your answer should also be in Chinese.\nATTENTION: Use '##' to SPLIT SECTIONS, not '#'. Output format carefully referenced \"Format example\".\n\n# Context\n## Design\n{\"Implementation approach\":\"We will use Python and the curses library to create the snake game. The game logic will be implemented in a separate module, and the main.py file will handle the user interface and game loop.\",\"File list\":[\"main.py\",\"game.py\"],\"Data structures and interfaces\":\"\\nclassDiagram\\n class Game {\\n -Snake snake\\n -Food food\\n -Score score\\n +__init__(width: int, height: int)\\n +start_game()\\n +move_snake(direction: str)\\n +generate_food()\\n +update_score(points: int)\\n }\\n class Snake {\\n -body list\\n -direction str\\n +__init__(x: int, y: int)\\n +move(direction: str)\\n +grow()\\n +collides_with_self() bool\\n }\\n class Food {\\n -position tuple\\n +__init__(x: int, y: int)\\n +get_position() tuple\\n }\\n class Score {\\n -points int\\n +__init__()\\n +increase(points: int)\\n }\\n Game --> Snake\\n Game --> Food\\n Game --> Score\\n\",\"Program call flow\":\"\\nsequenceDiagram\\n participant M as Main\\n participant G as Game\\n M->>G: start_game()\\n M->>G: move_snake(direction)\\n G->>G: generate_food()\\n G->>G: update_score(points)\\n\",\"Anything UNCLEAR\":\"Please provide more details on the game mechanics and user interactions.\"}\n\n## Tasks\n{\"Required Python packages\":[\"curses==2.2.0\"],\"Required Other language third-party packages\":[\"No third-party dependencies required\"],\"Logic Analysis\":[[\"game.py\",\"Contains Game class and ... functions\"],[\"main.py\",\"Contains main function, from game import Game\"]],\"Task list\":[\"game.py\",\"main.py\"],\"Full API spec\":\"\",\"Shared Knowledge\":\"'game.py' contains functions shared across the project.\",\"Anything UNCLEAR\":\"Please provide more details on the game mechanics and user interactions.\"}\n\n## Legacy Code\n```Code\n\n```\n\n## Debug logs\n```text\n\n\n\n```\n\n## Bug Feedback logs\n```text\n\n```\n\n# Format example\n## Code: game.py\n```python\n## game.py\n...\n```\n\n# Instruction: Based on the context, follow \"Format example\", write code.\n\n## Code: game.py. Write code with triple quoto, based on the following attentions and context.\n1. Only One file: do your best to implement THIS ONLY ONE FILE.\n2. COMPLETE CODE: Your code will be part of the entire project, so please implement complete, reliable, reusable code snippets.\n3. Set default value: If there is any setting, ALWAYS SET A DEFAULT VALUE, ALWAYS USE STRONG TYPE AND EXPLICIT VARIABLE. AVOID circular import.\n4. Follow design: YOU MUST FOLLOW \"Data structures and interfaces\". DONT CHANGE ANY DESIGN. Do not use public member functions that do not exist in your design.\n5. CAREFULLY CHECK THAT YOU DONT MISS ANY NECESSARY CLASS/FUNCTION IN THIS FILE.\n6. Before using a external variable/module, make sure you import it first.\n7. Write out EVERY CODE DETAIL, DON'T LEAVE TODO.\n\n": "```python\n## game.py\n\nclass Snake:\n def __init__(self, x: int, y: int):\n self.body = [(x, y)]\n self.direction = 'right'\n\n def move(self, direction: str):\n if direction == 'up' and self.direction != 'down':\n self.direction = 'up'\n elif direction == 'down' and self.direction != 'up':\n self.direction = 'down'\n elif direction == 'left' and self.direction != 'right':\n self.direction = 'left'\n elif direction == 'right' and self.direction != 'left':\n self.direction = 'right'\n\n head = self.body[0]\n x, y = head\n if self.direction == 'up':\n new_head = (x, y - 1)\n elif self.direction == 'down':\n new_head = (x, y + 1)\n elif self.direction == 'left':\n new_head = (x - 1, y)\n elif self.direction == 'right':\n new_head = (x + 1, y)\n self.body.insert(0, new_head)\n\n def grow(self):\n tail = self.body[-1]\n x, y = tail\n if self.direction == 'up':\n new_tail = (x, y + 1)\n elif self.direction == 'down':\n new_tail = (x, y - 1)\n elif self.direction == 'left':\n new_tail = (x + 1, y)\n elif self.direction == 'right':\n new_tail = (x - 1, y)\n self.body.append(new_tail)\n\n def collides_with_self(self) -> bool:\n return len(self.body) != len(set(self.body))\n\nclass Food:\n def __init__(self, x: int, y: int):\n self.position = (x, y)\n\n def get_position(self) -> tuple:\n return self.position\n\nclass Score:\n def __init__(self):\n self.points = 0\n\n def increase(self, points: int):\n self.points += points\n\nclass Game:\n def __init__(self, width: int, height: int):\n self.snake = Snake(width // 2, height // 2)\n self.food = None\n self.score = Score()\n self.width = width\n self.height = height\n\n def start_game(self):\n self.generate_food()\n\n def move_snake(self, direction: str):\n self.snake.move(direction)\n if self.check_collision():\n return False\n self.check_eat()\n return True\n\n def generate_food(self):\n import random\n x = random.randint(0, self.width - 1)\n y = random.randint(0, self.height - 1)\n self.food = Food(x, y)\n\n def update_score(self, points: int):\n self.score.increase(points)\n\n def check_collision(self) -> bool:\n head = self.snake.body[0]\n if head[0] < 0 or head[0] >= self.width or head[1] < 0 or head[1] >= self.height:\n return True\n if self.snake.collides_with_self():\n return True\n return False\n\n def check_eat(self):\n if self.snake.body[0] == self.food.get_position():\n self.snake.grow()\n self.generate_food()\n self.update_score(10)\n```", + "\n# System\nRole: You are a professional software engineer, and your main task is to review and revise the code. You need to ensure that the code conforms to the google-style standards, is elegantly designed and modularized, easy to read and maintain.\nLanguage: Please use the same language as the user requirement, but the title and code should be still in English. For example, if the user speaks Chinese, the specific text of your answer should also be in Chinese.\nATTENTION: Use '##' to SPLIT SECTIONS, not '#'. Output format carefully referenced \"Format example\".\n\n# Context\n## System Design\n{\"Implementation approach\":\"We will use Python and the curses library to create the snake game. The game logic will be implemented in a separate module, and the main.py file will handle the user interface and game loop.\",\"File list\":[\"main.py\",\"game.py\"],\"Data structures and interfaces\":\"\\nclassDiagram\\n class Game {\\n -Snake snake\\n -Food food\\n -Score score\\n +__init__(width: int, height: int)\\n +start_game()\\n +move_snake(direction: str)\\n +generate_food()\\n +update_score(points: int)\\n }\\n class Snake {\\n -body list\\n -direction str\\n +__init__(x: int, y: int)\\n +move(direction: str)\\n +grow()\\n +collides_with_self() bool\\n }\\n class Food {\\n -position tuple\\n +__init__(x: int, y: int)\\n +get_position() tuple\\n }\\n class Score {\\n -points int\\n +__init__()\\n +increase(points: int)\\n }\\n Game --> Snake\\n Game --> Food\\n Game --> Score\\n\",\"Program call flow\":\"\\nsequenceDiagram\\n participant M as Main\\n participant G as Game\\n M->>G: start_game()\\n M->>G: move_snake(direction)\\n G->>G: generate_food()\\n G->>G: update_score(points)\\n\",\"Anything UNCLEAR\":\"Please provide more details on the game mechanics and user interactions.\"}\n\n## Tasks\n{\"Required Python packages\":[\"curses==2.2.0\"],\"Required Other language third-party packages\":[\"No third-party dependencies required\"],\"Logic Analysis\":[[\"game.py\",\"Contains Game class and ... functions\"],[\"main.py\",\"Contains main function, from game import Game\"]],\"Task list\":[\"game.py\",\"main.py\"],\"Full API spec\":\"\",\"Shared Knowledge\":\"'game.py' contains functions shared across the project.\",\"Anything UNCLEAR\":\"Please provide more details on the game mechanics and user interactions.\"}\n\n## Code Files\n\n\n\n## Code to be Reviewed: game.py\n```Code\n## game.py\n\nclass Snake:\n def __init__(self, x: int, y: int):\n self.body = [(x, y)]\n self.direction = 'right'\n\n def move(self, direction: str):\n if direction == 'up' and self.direction != 'down':\n self.direction = 'up'\n elif direction == 'down' and self.direction != 'up':\n self.direction = 'down'\n elif direction == 'left' and self.direction != 'right':\n self.direction = 'left'\n elif direction == 'right' and self.direction != 'left':\n self.direction = 'right'\n\n head = self.body[0]\n x, y = head\n if self.direction == 'up':\n new_head = (x, y - 1)\n elif self.direction == 'down':\n new_head = (x, y + 1)\n elif self.direction == 'left':\n new_head = (x - 1, y)\n elif self.direction == 'right':\n new_head = (x + 1, y)\n self.body.insert(0, new_head)\n\n def grow(self):\n tail = self.body[-1]\n x, y = tail\n if self.direction == 'up':\n new_tail = (x, y + 1)\n elif self.direction == 'down':\n new_tail = (x, y - 1)\n elif self.direction == 'left':\n new_tail = (x + 1, y)\n elif self.direction == 'right':\n new_tail = (x - 1, y)\n self.body.append(new_tail)\n\n def collides_with_self(self) -> bool:\n return len(self.body) != len(set(self.body))\n\nclass Food:\n def __init__(self, x: int, y: int):\n self.position = (x, y)\n\n def get_position(self) -> tuple:\n return self.position\n\nclass Score:\n def __init__(self):\n self.points = 0\n\n def increase(self, points: int):\n self.points += points\n\nclass Game:\n def __init__(self, width: int, height: int):\n self.snake = Snake(width // 2, height // 2)\n self.food = None\n self.score = Score()\n self.width = width\n self.height = height\n\n def start_game(self):\n self.generate_food()\n\n def move_snake(self, direction: str):\n self.snake.move(direction)\n if self.check_collision():\n return False\n self.check_eat()\n return True\n\n def generate_food(self):\n import random\n x = random.randint(0, self.width - 1)\n y = random.randint(0, self.height - 1)\n self.food = Food(x, y)\n\n def update_score(self, points: int):\n self.score.increase(points)\n\n def check_collision(self) -> bool:\n head = self.snake.body[0]\n if head[0] < 0 or head[0] >= self.width or head[1] < 0 or head[1] >= self.height:\n return True\n if self.snake.collides_with_self():\n return True\n return False\n\n def check_eat(self):\n if self.snake.body[0] == self.food.get_position():\n self.snake.grow()\n self.generate_food()\n self.update_score(10)\n\n```\n\n\n\n# Format example 1\n## Code Review: game.py\n1. No, we should fix the logic of class A due to ...\n2. ...\n3. ...\n4. No, function B is not implemented, ...\n5. ...\n6. ...\n\n## Actions\n1. Fix the `handle_events` method to update the game state only if a move is successful.\n ```python\n def handle_events(self):\n for event in pygame.event.get():\n if event.type == pygame.QUIT:\n return False\n if event.type == pygame.KEYDOWN:\n moved = False\n if event.key == pygame.K_UP:\n moved = self.game.move('UP')\n elif event.key == pygame.K_DOWN:\n moved = self.game.move('DOWN')\n elif event.key == pygame.K_LEFT:\n moved = self.game.move('LEFT')\n elif event.key == pygame.K_RIGHT:\n moved = self.game.move('RIGHT')\n if moved:\n # Update the game state only if a move was successful\n self.render()\n return True\n ```\n2. Implement function B\n\n## Code Review Result\nLBTM\n\n# Format example 2\n## Code Review: game.py\n1. Yes.\n2. Yes.\n3. Yes.\n4. Yes.\n5. Yes.\n6. Yes.\n\n## Actions\npass\n\n## Code Review Result\nLGTM\n\n\n\n# Instruction: Based on the actual code situation, follow one of the \"Format example\". Return only 1 file under review.\n\n## Code Review: Ordered List. Based on the \"Code to be Reviewed\", provide key, clear, concise, and specific answer. If any answer is no, explain how to fix it step by step.\n1. Is the code implemented as per the requirements? If not, how to achieve it? Analyse it step by step.\n2. Is the code logic completely correct? If there are errors, please indicate how to correct them.\n3. Does the existing code follow the \"Data structures and interfaces\"?\n4. Are all functions implemented? If there is no implementation, please indicate how to achieve it step by step.\n5. Have all necessary pre-dependencies been imported? If not, indicate which ones need to be imported\n6. Are methods from other files being reused correctly?\n\n## Actions: Ordered List. Things that should be done after CR, such as implementing class A and function B\n\n## Code Review Result: str. If the code doesn't have bugs, we don't need to rewrite it, so answer LGTM and stop. ONLY ANSWER LGTM/LBTM.\nLGTM/LBTM\n\n": "## Code Review: game.py\n1. Yes, the code is implemented as per the requirements. The Snake, Food, and Score classes are correctly implemented, and the Game class handles the game logic as specified.\n2. Yes, the code logic is correct. The movement, growth, collision detection, and scoring mechanisms are implemented accurately.\n3. Yes, the existing code follows the \"Data structures and interfaces\" as specified in the class diagram.\n4. Yes, all functions are implemented as required.\n5. No, the necessary pre-dependencies have not been imported. The `curses` package needs to be imported at the beginning of the file using `import curses`.\n6. Yes, methods from other files are being reused correctly.\n\n## Actions\n1. Import the necessary pre-dependency `curses` at the beginning of the file using `import curses`.\n\n## Code Review Result\nLBTM", + "\n# System\nRole: You are a professional software engineer, and your main task is to review and revise the code. You need to ensure that the code conforms to the google-style standards, is elegantly designed and modularized, easy to read and maintain.\nLanguage: Please use the same language as the user requirement, but the title and code should be still in English. For example, if the user speaks Chinese, the specific text of your answer should also be in Chinese.\nATTENTION: Use '##' to SPLIT SECTIONS, not '#'. Output format carefully referenced \"Format example\".\n\n# Context\n## System Design\n{\"Implementation approach\":\"We will use Python and the curses library to create the snake game. The game logic will be implemented in a separate module, and the main.py file will handle the user interface and game loop.\",\"File list\":[\"main.py\",\"game.py\"],\"Data structures and interfaces\":\"\\nclassDiagram\\n class Game {\\n -Snake snake\\n -Food food\\n -Score score\\n +__init__(width: int, height: int)\\n +start_game()\\n +move_snake(direction: str)\\n +generate_food()\\n +update_score(points: int)\\n }\\n class Snake {\\n -body list\\n -direction str\\n +__init__(x: int, y: int)\\n +move(direction: str)\\n +grow()\\n +collides_with_self() bool\\n }\\n class Food {\\n -position tuple\\n +__init__(x: int, y: int)\\n +get_position() tuple\\n }\\n class Score {\\n -points int\\n +__init__()\\n +increase(points: int)\\n }\\n Game --> Snake\\n Game --> Food\\n Game --> Score\\n\",\"Program call flow\":\"\\nsequenceDiagram\\n participant M as Main\\n participant G as Game\\n M->>G: start_game()\\n M->>G: move_snake(direction)\\n G->>G: generate_food()\\n G->>G: update_score(points)\\n\",\"Anything UNCLEAR\":\"Please provide more details on the game mechanics and user interactions.\"}\n\n## Tasks\n{\"Required Python packages\":[\"curses==2.2.0\"],\"Required Other language third-party packages\":[\"No third-party dependencies required\"],\"Logic Analysis\":[[\"game.py\",\"Contains Game class and ... functions\"],[\"main.py\",\"Contains main function, from game import Game\"]],\"Task list\":[\"game.py\",\"main.py\"],\"Full API spec\":\"\",\"Shared Knowledge\":\"'game.py' contains functions shared across the project.\",\"Anything UNCLEAR\":\"Please provide more details on the game mechanics and user interactions.\"}\n\n## Code Files\n\n\n\n## Code to be Reviewed: game.py\n```Code\n## game.py\n\nclass Snake:\n def __init__(self, x: int, y: int):\n self.body = [(x, y)]\n self.direction = 'right'\n\n def move(self, direction: str):\n if direction == 'up' and self.direction != 'down':\n self.direction = 'up'\n elif direction == 'down' and self.direction != 'up':\n self.direction = 'down'\n elif direction == 'left' and self.direction != 'right':\n self.direction = 'left'\n elif direction == 'right' and self.direction != 'left':\n self.direction = 'right'\n\n head = self.body[0]\n x, y = head\n if self.direction == 'up':\n new_head = (x, y - 1)\n elif self.direction == 'down':\n new_head = (x, y + 1)\n elif self.direction == 'left':\n new_head = (x - 1, y)\n elif self.direction == 'right':\n new_head = (x + 1, y)\n self.body.insert(0, new_head)\n\n def grow(self):\n tail = self.body[-1]\n x, y = tail\n if self.direction == 'up':\n new_tail = (x, y + 1)\n elif self.direction == 'down':\n new_tail = (x, y - 1)\n elif self.direction == 'left':\n new_tail = (x + 1, y)\n elif self.direction == 'right':\n new_tail = (x - 1, y)\n self.body.append(new_tail)\n\n def collides_with_self(self) -> bool:\n return len(self.body) != len(set(self.body))\n\nclass Food:\n def __init__(self, x: int, y: int):\n self.position = (x, y)\n\n def get_position(self) -> tuple:\n return self.position\n\nclass Score:\n def __init__(self):\n self.points = 0\n\n def increase(self, points: int):\n self.points += points\n\nclass Game:\n def __init__(self, width: int, height: int):\n self.snake = Snake(width // 2, height // 2)\n self.food = None\n self.score = Score()\n self.width = width\n self.height = height\n\n def start_game(self):\n self.generate_food()\n\n def move_snake(self, direction: str):\n self.snake.move(direction)\n if self.check_collision():\n return False\n self.check_eat()\n return True\n\n def generate_food(self):\n import random\n x = random.randint(0, self.width - 1)\n y = random.randint(0, self.height - 1)\n self.food = Food(x, y)\n\n def update_score(self, points: int):\n self.score.increase(points)\n\n def check_collision(self) -> bool:\n head = self.snake.body[0]\n if head[0] < 0 or head[0] >= self.width or head[1] < 0 or head[1] >= self.height:\n return True\n if self.snake.collides_with_self():\n return True\n return False\n\n def check_eat(self):\n if self.snake.body[0] == self.food.get_position():\n self.snake.grow()\n self.generate_food()\n self.update_score(10)\n\n```\n\n## Code Review: game.py\n1. Yes, the code is implemented as per the requirements. The Snake, Food, and Score classes are correctly implemented, and the Game class handles the game logic as specified.\n2. Yes, the code logic is correct. The movement, growth, collision detection, and scoring mechanisms are implemented accurately.\n3. Yes, the existing code follows the \"Data structures and interfaces\" as specified in the class diagram.\n4. Yes, all functions are implemented as required.\n5. No, the necessary pre-dependencies have not been imported. The `curses` package needs to be imported at the beginning of the file using `import curses`.\n6. Yes, methods from other files are being reused correctly.\n\n## Actions\n1. Import the necessary pre-dependency `curses` at the beginning of the file using `import curses`.\n\n## Code Review Result\nLBTM\n\n# Instruction: rewrite code based on the Code Review and Actions\n## Rewrite Code: CodeBlock. If it still has some bugs, rewrite game.py with triple quotes. Do your utmost to optimize THIS SINGLE FILE. Return all completed codes and prohibit the return of unfinished codes.\n```Code\n## game.py\n...\n```\n": "```Code\n## game.py\n\nimport curses\n\nclass Snake:\n def __init__(self, x: int, y: int):\n self.body = [(x, y)]\n self.direction = 'right'\n\n def move(self, direction: str):\n if direction == 'up' and self.direction != 'down':\n self.direction = 'up'\n elif direction == 'down' and self.direction != 'up':\n self.direction = 'down'\n elif direction == 'left' and self.direction != 'right':\n self.direction = 'left'\n elif direction == 'right' and self.direction != 'left':\n self.direction = 'right'\n\n head = self.body[0]\n x, y = head\n if self.direction == 'up':\n new_head = (x, y - 1)\n elif self.direction == 'down':\n new_head = (x, y + 1)\n elif self.direction == 'left':\n new_head = (x - 1, y)\n elif self.direction == 'right':\n new_head = (x + 1, y)\n self.body.insert(0, new_head)\n\n def grow(self):\n tail = self.body[-1]\n x, y = tail\n if self.direction == 'up':\n new_tail = (x, y + 1)\n elif self.direction == 'down':\n new_tail = (x, y - 1)\n elif self.direction == 'left':\n new_tail = (x + 1, y)\n elif self.direction == 'right':\n new_tail = (x - 1, y)\n self.body.append(new_tail)\n\n def collides_with_self(self) -> bool:\n return len(self.body) != len(set(self.body))\n\nclass Food:\n def __init__(self, x: int, y: int):\n self.position = (x, y)\n\n def get_position(self) -> tuple:\n return self.position\n\nclass Score:\n def __init__(self):\n self.points = 0\n\n def increase(self, points: int):\n self.points += points\n\nclass Game:\n def __init__(self, width: int, height: int):\n self.snake = Snake(width // 2, height // 2)\n self.food = None\n self.score = Score()\n self.width = width\n self.height = height\n\n def start_game(self):\n self.generate_food()\n\n def move_snake(self, direction: str):\n self.snake.move(direction)\n if self.check_collision():\n return False\n self.check_eat()\n return True\n\n def generate_food(self):\n import random\n x = random.randint(0, self.width - 1)\n y = random.randint(0, self.height - 1)\n self.food = Food(x, y)\n\n def update_score(self, points: int):\n self.score.increase(points)\n\n def check_collision(self) -> bool:\n head = self.snake.body[0]\n if head[0] < 0 or head[0] >= self.width or head[1] < 0 or head[1] >= self.height:\n return True\n if self.snake.collides_with_self():\n return True\n return False\n\n def check_eat(self):\n if self.snake.body[0] == self.food.get_position():\n self.snake.grow()\n self.generate_food()\n self.update_score(10)\n\n```", + "\n# System\nRole: You are a professional software engineer, and your main task is to review and revise the code. You need to ensure that the code conforms to the google-style standards, is elegantly designed and modularized, easy to read and maintain.\nLanguage: Please use the same language as the user requirement, but the title and code should be still in English. For example, if the user speaks Chinese, the specific text of your answer should also be in Chinese.\nATTENTION: Use '##' to SPLIT SECTIONS, not '#'. Output format carefully referenced \"Format example\".\n\n# Context\n## System Design\n{\"Implementation approach\":\"We will use Python and the curses library to create the snake game. The game logic will be implemented in a separate module, and the main.py file will handle the user interface and game loop.\",\"File list\":[\"main.py\",\"game.py\"],\"Data structures and interfaces\":\"\\nclassDiagram\\n class Game {\\n -Snake snake\\n -Food food\\n -Score score\\n +__init__(width: int, height: int)\\n +start_game()\\n +move_snake(direction: str)\\n +generate_food()\\n +update_score(points: int)\\n }\\n class Snake {\\n -body list\\n -direction str\\n +__init__(x: int, y: int)\\n +move(direction: str)\\n +grow()\\n +collides_with_self() bool\\n }\\n class Food {\\n -position tuple\\n +__init__(x: int, y: int)\\n +get_position() tuple\\n }\\n class Score {\\n -points int\\n +__init__()\\n +increase(points: int)\\n }\\n Game --> Snake\\n Game --> Food\\n Game --> Score\\n\",\"Program call flow\":\"\\nsequenceDiagram\\n participant M as Main\\n participant G as Game\\n M->>G: start_game()\\n M->>G: move_snake(direction)\\n G->>G: generate_food()\\n G->>G: update_score(points)\\n\",\"Anything UNCLEAR\":\"Please provide more details on the game mechanics and user interactions.\"}\n\n## Tasks\n{\"Required Python packages\":[\"curses==2.2.0\"],\"Required Other language third-party packages\":[\"No third-party dependencies required\"],\"Logic Analysis\":[[\"game.py\",\"Contains Game class and ... functions\"],[\"main.py\",\"Contains main function, from game import Game\"]],\"Task list\":[\"game.py\",\"main.py\"],\"Full API spec\":\"\",\"Shared Knowledge\":\"'game.py' contains functions shared across the project.\",\"Anything UNCLEAR\":\"Please provide more details on the game mechanics and user interactions.\"}\n\n## Code Files\n\n\n\n## Code to be Reviewed: game.py\n```Code\n## game.py\n\nimport curses\n\nclass Snake:\n def __init__(self, x: int, y: int):\n self.body = [(x, y)]\n self.direction = 'right'\n\n def move(self, direction: str):\n if direction == 'up' and self.direction != 'down':\n self.direction = 'up'\n elif direction == 'down' and self.direction != 'up':\n self.direction = 'down'\n elif direction == 'left' and self.direction != 'right':\n self.direction = 'left'\n elif direction == 'right' and self.direction != 'left':\n self.direction = 'right'\n\n head = self.body[0]\n x, y = head\n if self.direction == 'up':\n new_head = (x, y - 1)\n elif self.direction == 'down':\n new_head = (x, y + 1)\n elif self.direction == 'left':\n new_head = (x - 1, y)\n elif self.direction == 'right':\n new_head = (x + 1, y)\n self.body.insert(0, new_head)\n\n def grow(self):\n tail = self.body[-1]\n x, y = tail\n if self.direction == 'up':\n new_tail = (x, y + 1)\n elif self.direction == 'down':\n new_tail = (x, y - 1)\n elif self.direction == 'left':\n new_tail = (x + 1, y)\n elif self.direction == 'right':\n new_tail = (x - 1, y)\n self.body.append(new_tail)\n\n def collides_with_self(self) -> bool:\n return len(self.body) != len(set(self.body))\n\nclass Food:\n def __init__(self, x: int, y: int):\n self.position = (x, y)\n\n def get_position(self) -> tuple:\n return self.position\n\nclass Score:\n def __init__(self):\n self.points = 0\n\n def increase(self, points: int):\n self.points += points\n\nclass Game:\n def __init__(self, width: int, height: int):\n self.snake = Snake(width // 2, height // 2)\n self.food = None\n self.score = Score()\n self.width = width\n self.height = height\n\n def start_game(self):\n self.generate_food()\n\n def move_snake(self, direction: str):\n self.snake.move(direction)\n if self.check_collision():\n return False\n self.check_eat()\n return True\n\n def generate_food(self):\n import random\n x = random.randint(0, self.width - 1)\n y = random.randint(0, self.height - 1)\n self.food = Food(x, y)\n\n def update_score(self, points: int):\n self.score.increase(points)\n\n def check_collision(self) -> bool:\n head = self.snake.body[0]\n if head[0] < 0 or head[0] >= self.width or head[1] < 0 or head[1] >= self.height:\n return True\n if self.snake.collides_with_self():\n return True\n return False\n\n def check_eat(self):\n if self.snake.body[0] == self.food.get_position():\n self.snake.grow()\n self.generate_food()\n self.update_score(10)\n\n\n```\n\n\n\n# Format example 1\n## Code Review: game.py\n1. No, we should fix the logic of class A due to ...\n2. ...\n3. ...\n4. No, function B is not implemented, ...\n5. ...\n6. ...\n\n## Actions\n1. Fix the `handle_events` method to update the game state only if a move is successful.\n ```python\n def handle_events(self):\n for event in pygame.event.get():\n if event.type == pygame.QUIT:\n return False\n if event.type == pygame.KEYDOWN:\n moved = False\n if event.key == pygame.K_UP:\n moved = self.game.move('UP')\n elif event.key == pygame.K_DOWN:\n moved = self.game.move('DOWN')\n elif event.key == pygame.K_LEFT:\n moved = self.game.move('LEFT')\n elif event.key == pygame.K_RIGHT:\n moved = self.game.move('RIGHT')\n if moved:\n # Update the game state only if a move was successful\n self.render()\n return True\n ```\n2. Implement function B\n\n## Code Review Result\nLBTM\n\n# Format example 2\n## Code Review: game.py\n1. Yes.\n2. Yes.\n3. Yes.\n4. Yes.\n5. Yes.\n6. Yes.\n\n## Actions\npass\n\n## Code Review Result\nLGTM\n\n\n\n# Instruction: Based on the actual code situation, follow one of the \"Format example\". Return only 1 file under review.\n\n## Code Review: Ordered List. Based on the \"Code to be Reviewed\", provide key, clear, concise, and specific answer. If any answer is no, explain how to fix it step by step.\n1. Is the code implemented as per the requirements? If not, how to achieve it? Analyse it step by step.\n2. Is the code logic completely correct? If there are errors, please indicate how to correct them.\n3. Does the existing code follow the \"Data structures and interfaces\"?\n4. Are all functions implemented? If there is no implementation, please indicate how to achieve it step by step.\n5. Have all necessary pre-dependencies been imported? If not, indicate which ones need to be imported\n6. Are methods from other files being reused correctly?\n\n## Actions: Ordered List. Things that should be done after CR, such as implementing class A and function B\n\n## Code Review Result: str. If the code doesn't have bugs, we don't need to rewrite it, so answer LGTM and stop. ONLY ANSWER LGTM/LBTM.\nLGTM/LBTM\n\n": "## Code Review: game.py\n1. Yes, the code is implemented as per the requirements.\n2. Yes, the code logic is completely correct.\n3. Yes, the existing code follows the \"Data structures and interfaces\".\n4. Yes, all functions are implemented.\n5. Yes, all necessary pre-dependencies have been imported.\n6. Yes, methods from other files are being reused correctly.\n\n## Code Review Result\nLGTM", + "\nNOTICE\nRole: You are a professional engineer; the main goal is to write google-style, elegant, modular, easy to read and maintain code\nLanguage: Please use the same language as the user requirement, but the title and code should be still in English. For example, if the user speaks Chinese, the specific text of your answer should also be in Chinese.\nATTENTION: Use '##' to SPLIT SECTIONS, not '#'. Output format carefully referenced \"Format example\".\n\n# Context\n## Design\n{\"Implementation approach\":\"We will use Python and the curses library to create the snake game. The game logic will be implemented in a separate module, and the main.py file will handle the user interface and game loop.\",\"File list\":[\"main.py\",\"game.py\"],\"Data structures and interfaces\":\"\\nclassDiagram\\n class Game {\\n -Snake snake\\n -Food food\\n -Score score\\n +__init__(width: int, height: int)\\n +start_game()\\n +move_snake(direction: str)\\n +generate_food()\\n +update_score(points: int)\\n }\\n class Snake {\\n -body list\\n -direction str\\n +__init__(x: int, y: int)\\n +move(direction: str)\\n +grow()\\n +collides_with_self() bool\\n }\\n class Food {\\n -position tuple\\n +__init__(x: int, y: int)\\n +get_position() tuple\\n }\\n class Score {\\n -points int\\n +__init__()\\n +increase(points: int)\\n }\\n Game --> Snake\\n Game --> Food\\n Game --> Score\\n\",\"Program call flow\":\"\\nsequenceDiagram\\n participant M as Main\\n participant G as Game\\n M->>G: start_game()\\n M->>G: move_snake(direction)\\n G->>G: generate_food()\\n G->>G: update_score(points)\\n\",\"Anything UNCLEAR\":\"Please provide more details on the game mechanics and user interactions.\"}\n\n## Tasks\n{\"Required Python packages\":[\"curses==2.2.0\"],\"Required Other language third-party packages\":[\"No third-party dependencies required\"],\"Logic Analysis\":[[\"game.py\",\"Contains Game class and ... functions\"],[\"main.py\",\"Contains main function, from game import Game\"]],\"Task list\":[\"game.py\",\"main.py\"],\"Full API spec\":\"\",\"Shared Knowledge\":\"'game.py' contains functions shared across the project.\",\"Anything UNCLEAR\":\"Please provide more details on the game mechanics and user interactions.\"}\n\n## Legacy Code\n```Code\n----- game.py\n## game.py\n\nimport curses\n\nclass Snake:\n def __init__(self, x: int, y: int):\n self.body = [(x, y)]\n self.direction = 'right'\n\n def move(self, direction: str):\n if direction == 'up' and self.direction != 'down':\n self.direction = 'up'\n elif direction == 'down' and self.direction != 'up':\n self.direction = 'down'\n elif direction == 'left' and self.direction != 'right':\n self.direction = 'left'\n elif direction == 'right' and self.direction != 'left':\n self.direction = 'right'\n\n head = self.body[0]\n x, y = head\n if self.direction == 'up':\n new_head = (x, y - 1)\n elif self.direction == 'down':\n new_head = (x, y + 1)\n elif self.direction == 'left':\n new_head = (x - 1, y)\n elif self.direction == 'right':\n new_head = (x + 1, y)\n self.body.insert(0, new_head)\n\n def grow(self):\n tail = self.body[-1]\n x, y = tail\n if self.direction == 'up':\n new_tail = (x, y + 1)\n elif self.direction == 'down':\n new_tail = (x, y - 1)\n elif self.direction == 'left':\n new_tail = (x + 1, y)\n elif self.direction == 'right':\n new_tail = (x - 1, y)\n self.body.append(new_tail)\n\n def collides_with_self(self) -> bool:\n return len(self.body) != len(set(self.body))\n\nclass Food:\n def __init__(self, x: int, y: int):\n self.position = (x, y)\n\n def get_position(self) -> tuple:\n return self.position\n\nclass Score:\n def __init__(self):\n self.points = 0\n\n def increase(self, points: int):\n self.points += points\n\nclass Game:\n def __init__(self, width: int, height: int):\n self.snake = Snake(width // 2, height // 2)\n self.food = None\n self.score = Score()\n self.width = width\n self.height = height\n\n def start_game(self):\n self.generate_food()\n\n def move_snake(self, direction: str):\n self.snake.move(direction)\n if self.check_collision():\n return False\n self.check_eat()\n return True\n\n def generate_food(self):\n import random\n x = random.randint(0, self.width - 1)\n y = random.randint(0, self.height - 1)\n self.food = Food(x, y)\n\n def update_score(self, points: int):\n self.score.increase(points)\n\n def check_collision(self) -> bool:\n head = self.snake.body[0]\n if head[0] < 0 or head[0] >= self.width or head[1] < 0 or head[1] >= self.height:\n return True\n if self.snake.collides_with_self():\n return True\n return False\n\n def check_eat(self):\n if self.snake.body[0] == self.food.get_position():\n self.snake.grow()\n self.generate_food()\n self.update_score(10)\n\n\n```\n\n## Debug logs\n```text\n\n\n\n```\n\n## Bug Feedback logs\n```text\n\n```\n\n# Format example\n## Code: main.py\n```python\n## main.py\n...\n```\n\n# Instruction: Based on the context, follow \"Format example\", write code.\n\n## Code: main.py. Write code with triple quoto, based on the following attentions and context.\n1. Only One file: do your best to implement THIS ONLY ONE FILE.\n2. COMPLETE CODE: Your code will be part of the entire project, so please implement complete, reliable, reusable code snippets.\n3. Set default value: If there is any setting, ALWAYS SET A DEFAULT VALUE, ALWAYS USE STRONG TYPE AND EXPLICIT VARIABLE. AVOID circular import.\n4. Follow design: YOU MUST FOLLOW \"Data structures and interfaces\". DONT CHANGE ANY DESIGN. Do not use public member functions that do not exist in your design.\n5. CAREFULLY CHECK THAT YOU DONT MISS ANY NECESSARY CLASS/FUNCTION IN THIS FILE.\n6. Before using a external variable/module, make sure you import it first.\n7. Write out EVERY CODE DETAIL, DON'T LEAVE TODO.\n\n": "```python\n## main.py\n\nfrom game import Game\n\ndef main():\n width = 20 # Set default width\n height = 20 # Set default height\n game = Game(width, height)\n game.start_game()\n\n while True:\n # Get user input for direction\n direction = input(\"Enter direction (up/down/left/right): \")\n if direction in ['up', 'down', 'left', 'right']:\n if not game.move_snake(direction):\n print(\"Game over!\")\n break\n else:\n print(\"Invalid direction!\")\n\nif __name__ == \"__main__\":\n main()\n```", + "\n# System\nRole: You are a professional software engineer, and your main task is to review and revise the code. You need to ensure that the code conforms to the google-style standards, is elegantly designed and modularized, easy to read and maintain.\nLanguage: Please use the same language as the user requirement, but the title and code should be still in English. For example, if the user speaks Chinese, the specific text of your answer should also be in Chinese.\nATTENTION: Use '##' to SPLIT SECTIONS, not '#'. Output format carefully referenced \"Format example\".\n\n# Context\n## System Design\n{\"Implementation approach\":\"We will use Python and the curses library to create the snake game. The game logic will be implemented in a separate module, and the main.py file will handle the user interface and game loop.\",\"File list\":[\"main.py\",\"game.py\"],\"Data structures and interfaces\":\"\\nclassDiagram\\n class Game {\\n -Snake snake\\n -Food food\\n -Score score\\n +__init__(width: int, height: int)\\n +start_game()\\n +move_snake(direction: str)\\n +generate_food()\\n +update_score(points: int)\\n }\\n class Snake {\\n -body list\\n -direction str\\n +__init__(x: int, y: int)\\n +move(direction: str)\\n +grow()\\n +collides_with_self() bool\\n }\\n class Food {\\n -position tuple\\n +__init__(x: int, y: int)\\n +get_position() tuple\\n }\\n class Score {\\n -points int\\n +__init__()\\n +increase(points: int)\\n }\\n Game --> Snake\\n Game --> Food\\n Game --> Score\\n\",\"Program call flow\":\"\\nsequenceDiagram\\n participant M as Main\\n participant G as Game\\n M->>G: start_game()\\n M->>G: move_snake(direction)\\n G->>G: generate_food()\\n G->>G: update_score(points)\\n\",\"Anything UNCLEAR\":\"Please provide more details on the game mechanics and user interactions.\"}\n\n## Tasks\n{\"Required Python packages\":[\"curses==2.2.0\"],\"Required Other language third-party packages\":[\"No third-party dependencies required\"],\"Logic Analysis\":[[\"game.py\",\"Contains Game class and ... functions\"],[\"main.py\",\"Contains main function, from game import Game\"]],\"Task list\":[\"game.py\",\"main.py\"],\"Full API spec\":\"\",\"Shared Knowledge\":\"'game.py' contains functions shared across the project.\",\"Anything UNCLEAR\":\"Please provide more details on the game mechanics and user interactions.\"}\n\n## Code Files\n----- game.py\n## game.py\n\nimport curses\n\nclass Snake:\n def __init__(self, x: int, y: int):\n self.body = [(x, y)]\n self.direction = 'right'\n\n def move(self, direction: str):\n if direction == 'up' and self.direction != 'down':\n self.direction = 'up'\n elif direction == 'down' and self.direction != 'up':\n self.direction = 'down'\n elif direction == 'left' and self.direction != 'right':\n self.direction = 'left'\n elif direction == 'right' and self.direction != 'left':\n self.direction = 'right'\n\n head = self.body[0]\n x, y = head\n if self.direction == 'up':\n new_head = (x, y - 1)\n elif self.direction == 'down':\n new_head = (x, y + 1)\n elif self.direction == 'left':\n new_head = (x - 1, y)\n elif self.direction == 'right':\n new_head = (x + 1, y)\n self.body.insert(0, new_head)\n\n def grow(self):\n tail = self.body[-1]\n x, y = tail\n if self.direction == 'up':\n new_tail = (x, y + 1)\n elif self.direction == 'down':\n new_tail = (x, y - 1)\n elif self.direction == 'left':\n new_tail = (x + 1, y)\n elif self.direction == 'right':\n new_tail = (x - 1, y)\n self.body.append(new_tail)\n\n def collides_with_self(self) -> bool:\n return len(self.body) != len(set(self.body))\n\nclass Food:\n def __init__(self, x: int, y: int):\n self.position = (x, y)\n\n def get_position(self) -> tuple:\n return self.position\n\nclass Score:\n def __init__(self):\n self.points = 0\n\n def increase(self, points: int):\n self.points += points\n\nclass Game:\n def __init__(self, width: int, height: int):\n self.snake = Snake(width // 2, height // 2)\n self.food = None\n self.score = Score()\n self.width = width\n self.height = height\n\n def start_game(self):\n self.generate_food()\n\n def move_snake(self, direction: str):\n self.snake.move(direction)\n if self.check_collision():\n return False\n self.check_eat()\n return True\n\n def generate_food(self):\n import random\n x = random.randint(0, self.width - 1)\n y = random.randint(0, self.height - 1)\n self.food = Food(x, y)\n\n def update_score(self, points: int):\n self.score.increase(points)\n\n def check_collision(self) -> bool:\n head = self.snake.body[0]\n if head[0] < 0 or head[0] >= self.width or head[1] < 0 or head[1] >= self.height:\n return True\n if self.snake.collides_with_self():\n return True\n return False\n\n def check_eat(self):\n if self.snake.body[0] == self.food.get_position():\n self.snake.grow()\n self.generate_food()\n self.update_score(10)\n\n\n\n\n## Code to be Reviewed: main.py\n```Code\n## main.py\n\nfrom game import Game\n\ndef main():\n width = 20 # Set default width\n height = 20 # Set default height\n game = Game(width, height)\n game.start_game()\n\n while True:\n # Get user input for direction\n direction = input(\"Enter direction (up/down/left/right): \")\n if direction in ['up', 'down', 'left', 'right']:\n if not game.move_snake(direction):\n print(\"Game over!\")\n break\n else:\n print(\"Invalid direction!\")\n\nif __name__ == \"__main__\":\n main()\n\n```\n\n\n\n# Format example 1\n## Code Review: main.py\n1. No, we should fix the logic of class A due to ...\n2. ...\n3. ...\n4. No, function B is not implemented, ...\n5. ...\n6. ...\n\n## Actions\n1. Fix the `handle_events` method to update the game state only if a move is successful.\n ```python\n def handle_events(self):\n for event in pygame.event.get():\n if event.type == pygame.QUIT:\n return False\n if event.type == pygame.KEYDOWN:\n moved = False\n if event.key == pygame.K_UP:\n moved = self.game.move('UP')\n elif event.key == pygame.K_DOWN:\n moved = self.game.move('DOWN')\n elif event.key == pygame.K_LEFT:\n moved = self.game.move('LEFT')\n elif event.key == pygame.K_RIGHT:\n moved = self.game.move('RIGHT')\n if moved:\n # Update the game state only if a move was successful\n self.render()\n return True\n ```\n2. Implement function B\n\n## Code Review Result\nLBTM\n\n# Format example 2\n## Code Review: main.py\n1. Yes.\n2. Yes.\n3. Yes.\n4. Yes.\n5. Yes.\n6. Yes.\n\n## Actions\npass\n\n## Code Review Result\nLGTM\n\n\n\n# Instruction: Based on the actual code situation, follow one of the \"Format example\". Return only 1 file under review.\n\n## Code Review: Ordered List. Based on the \"Code to be Reviewed\", provide key, clear, concise, and specific answer. If any answer is no, explain how to fix it step by step.\n1. Is the code implemented as per the requirements? If not, how to achieve it? Analyse it step by step.\n2. Is the code logic completely correct? If there are errors, please indicate how to correct them.\n3. Does the existing code follow the \"Data structures and interfaces\"?\n4. Are all functions implemented? If there is no implementation, please indicate how to achieve it step by step.\n5. Have all necessary pre-dependencies been imported? If not, indicate which ones need to be imported\n6. Are methods from other files being reused correctly?\n\n## Actions: Ordered List. Things that should be done after CR, such as implementing class A and function B\n\n## Code Review Result: str. If the code doesn't have bugs, we don't need to rewrite it, so answer LGTM and stop. ONLY ANSWER LGTM/LBTM.\nLGTM/LBTM\n\n": "## Code Review: main.py\n1. The code is implemented as per the requirements. The main function initializes the game and handles user input for the snake's direction.\n2. The code logic is correct. It correctly handles user input and updates the game state based on the snake's movement.\n3. The existing code follows the \"Data structures and interfaces\" as defined in the class diagram.\n4. All necessary functions are implemented in the code.\n5. The necessary pre-dependencies have been imported. The required Python package \"curses==2.2.0\" has been imported.\n6. The methods from the \"game.py\" file are being reused correctly.\n\n## Actions\npass\n\n## Code Review Result\nLGTM", + "\n## context\n\n### Project Name\n20240110212717\n\n### Original Requirements\n['开发一个基于大语言模型与私有知识库的搜索引擎，希望可以基于大语言模型进行搜索总结']\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"zh_cn\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"开发一个基于大语言模型与私有知识库的搜索引擎，希望可以基于大语言模型进行搜索总结\",\n \"Product Goals\": [],\n \"User Stories\": [],\n \"Competitive Analysis\": [],\n \"Competitive Quadrant Chart\": \"\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [],\n \"UI Design draft\": \"\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]", + "\n## context\n\n### Project Name\n20240110212717\n\n### Original Requirements\n['']\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"\",\n \"Product Goals\": [],\n \"User Stories\": [],\n \"Competitive Analysis\": [],\n \"Competitive Quadrant Chart\": \"\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [],\n \"UI Design draft\": \"\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]", + "\n## context\n\n### Project Name\n20240110220803\n\n### Original Requirements\n['需要一个基于LLM做总结的搜索引擎']\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"zh_cn\",\n \"Programming Language\": \"LLM\",\n \"Original Requirements\": \"需要一个基于LLM做总结的搜索引擎\",\n \"Product Goals\": [],\n \"User Stories\": [],\n \"Competitive Analysis\": [],\n \"Competitive Quadrant Chart\": \"\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [],\n \"UI Design draft\": \"\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]", + "You are a python code to Mermaid Sequence Diagram translator in function detail#SYSTEM_MSG_END#```python\n#!/usr/bin/env python\n# -*- coding: utf-8 -*-\nimport asyncio\nimport shutil\nfrom pathlib import Path\n\nimport typer\n\nfrom metagpt.config2 import config\nfrom metagpt.const import METAGPT_ROOT\n\napp = typer.Typer(add_completion=False, pretty_exceptions_show_locals=False)\n\n\ndef generate_repo(\n idea,\n investment,\n n_round,\n code_review,\n run_tests,\n implement,\n project_name,\n inc,\n project_path,\n reqa_file,\n max_auto_summarize_code,\n recover_path,\n):\n \"\"\"Run the startup logic. Can be called from CLI or other Python scripts.\"\"\"\n from metagpt.roles import (\n Architect,\n Engineer,\n ProductManager,\n ProjectManager,\n QaEngineer,\n )\n from metagpt.team import Team\n\n config.update_via_cli(project_path, project_name, inc, reqa_file, max_auto_summarize_code)\n\n if not recover_path:\n company = Team()\n company.hire(\n [\n ProductManager(),\n Architect(),\n ProjectManager(),\n ]\n )\n\n if implement or code_review:\n company.hire([Engineer(n_borg=5, use_code_review=code_review)])\n\n if run_tests:\n company.hire([QaEngineer()])\n else:\n stg_path = Path(recover_path)\n if not stg_path.exists() or not str(stg_path).endswith(\"team\"):\n raise FileNotFoundError(f\"{recover_path} not exists or not endswith `team`\")\n\n company = Team.deserialize(stg_path=stg_path)\n idea = company.idea\n\n company.invest(investment)\n company.run_project(idea)\n asyncio.run(company.run(n_round=n_round))\n\n\n@app.command(\"\", help=\"Start a new project.\")\ndef startup(\n idea: str = typer.Argument(None, help=\"Your innovative idea, such as 'Create a 2048 game.'\"),\n investment: float = typer.Option(default=3.0, help=\"Dollar amount to invest in the AI company.\"),\n n_round: int = typer.Option(default=5, help=\"Number of rounds for the simulation.\"),\n code_review: bool = typer.Option(default=True, help=\"Whether to use code review.\"),\n run_tests: bool = typer.Option(default=False, help=\"Whether to enable QA for adding & running tests.\"),\n implement: bool = typer.Option(default=True, help=\"Enable or disable code implementation.\"),\n project_name: str = typer.Option(default=\"\", help=\"Unique project name, such as 'game_2048'.\"),\n inc: bool = typer.Option(default=False, help=\"Incremental mode. Use it to coop with existing repo.\"),\n project_path: str = typer.Option(\n default=\"\",\n help=\"Specify the directory path of the old version project to fulfill the incremental requirements.\",\n ),\n reqa_file: str = typer.Option(\n default=\"\", help=\"Specify the source file name for rewriting the quality assurance code.\"\n ),\n max_auto_summarize_code: int = typer.Option(\n default=0,\n help=\"The maximum number of times the 'SummarizeCode' action is automatically invoked, with -1 indicating \"\n \"unlimited. This parameter is used for debugging the workflow.\",\n ),\n recover_path: str = typer.Option(default=None, help=\"recover the project from existing serialized storage\"),\n init_config: bool = typer.Option(default=False, help=\"Initialize the configuration file for MetaGPT.\"),\n):\n \"\"\"Run a startup. Be a boss.\"\"\"\n if init_config:\n copy_config_to()\n return\n\n if idea is None:\n typer.echo(\"Missing argument 'IDEA'. Run 'metagpt --help' for more information.\")\n raise typer.Exit()\n\n return generate_repo(\n idea,\n investment,\n n_round,\n code_review,\n run_tests,\n implement,\n project_name,\n inc,\n project_path,\n reqa_file,\n max_auto_summarize_code,\n recover_path,\n )\n\n\ndef copy_config_to(config_path=METAGPT_ROOT / \"config\" / \"config2.yaml\"):\n \"\"\"Initialize the configuration file for MetaGPT.\"\"\"\n target_path = Path.home() / \".metagpt\" / \"config2.yaml\"\n\n # 创建目标目录（如果不存在）\n target_path.parent.mkdir(parents=True, exist_ok=True)\n\n # 如果目标文件已经存在，则重命名为 .bak\n if target_path.exists():\n backup_path = target_path.with_suffix(\".bak\")\n target_path.rename(backup_path)\n print(f\"Existing configuration file backed up at {backup_path}\")\n\n # 复制文件\n shutil.copy(str(config_path), target_path)\n print(f\"Configuration file initialized at {target_path}\")\n\n\nif __name__ == \"__main__\":\n app()\n\n```\n\n---\nTranslate the code above into Mermaid Sequence Diagram.": "```mermaid\nsequenceDiagram\n participant app as app\n participant typer as typer\n participant generate_repo as generate_repo\n participant Team as company\n participant ProductManager as ProductManager\n participant Architect as Architect\n participant ProjectManager as ProjectManager\n participant Engineer as Engineer\n participant QaEngineer as QaEngineer\n\n app -> typer: startup()\n typer -> generate_repo: generate_repo()\n generate_repo -> config: config.update_via_cli()\n generate_repo -> company: company.hire([ProductManager, Architect, ProjectManager])\n generate_repo -> company: company.hire([Engineer])\n generate_repo -> company: company.hire([QaEngineer])\n generate_repo -> company: company.invest()\n generate_repo -> company: company.run_project()\n generate_repo -> company: asyncio.run(company.run())\n\n Note right of generate_repo: If recover_path is provided,
deserialize Team from recover_path\n```", + "You are a python code to Mermaid Sequence Diagram translator in function detail#SYSTEM_MSG_END#```python\n#!/usr/bin/env python\n\nfrom __future__ import annotations\n\nimport asyncio\nimport json\nfrom concurrent import futures\nfrom typing import Literal, overload\n\nfrom metagpt.config2 import config\n\ntry:\n from duckduckgo_search import DDGS\nexcept ImportError:\n raise ImportError(\n \"To use this module, you should have the `duckduckgo_search` Python package installed. \"\n \"You can install it by running the command: `pip install -e.[search-ddg]`\"\n )\n\n\nclass DDGAPIWrapper:\n \"\"\"Wrapper around duckduckgo_search API.\n\n To use this module, you should have the `duckduckgo_search` Python package installed.\n \"\"\"\n\n def __init__(\n self,\n *,\n loop: asyncio.AbstractEventLoop | None = None,\n executor: futures.Executor | None = None,\n ):\n kwargs = {}\n if config.proxy:\n kwargs[\"proxies\"] = config.proxy\n self.loop = loop\n self.executor = executor\n self.ddgs = DDGS(**kwargs)\n\n @overload\n def run(\n self,\n query: str,\n max_results: int = 8,\n as_string: Literal[True] = True,\n focus: list[str] | None = None,\n ) -> str:\n ...\n\n @overload\n def run(\n self,\n query: str,\n max_results: int = 8,\n as_string: Literal[False] = False,\n focus: list[str] | None = None,\n ) -> list[dict[str, str]]:\n ...\n\n async def run(\n self,\n query: str,\n max_results: int = 8,\n as_string: bool = True,\n ) -> str | list[dict]:\n \"\"\"Return the results of a Google search using the official Google API\n\n Args:\n query: The search query.\n max_results: The number of results to return.\n as_string: A boolean flag to determine the return type of the results. If True, the function will\n return a formatted string with the search results. If False, it will return a list of dictionaries\n containing detailed information about each search result.\n\n Returns:\n The results of the search.\n \"\"\"\n loop = self.loop or asyncio.get_event_loop()\n future = loop.run_in_executor(\n self.executor,\n self._search_from_ddgs,\n query,\n max_results,\n )\n search_results = await future\n\n # Return the list of search result URLs\n if as_string:\n return json.dumps(search_results, ensure_ascii=False)\n return search_results\n\n def _search_from_ddgs(self, query: str, max_results: int):\n return [\n {\"link\": i[\"href\"], \"snippet\": i[\"body\"], \"title\": i[\"title\"]}\n for (_, i) in zip(range(max_results), self.ddgs.text(query))\n ]\n\n\nif __name__ == \"__main__\":\n import fire\n\n fire.Fire(DDGAPIWrapper().run)\n\n```\n\n---\nTranslate the code above into Mermaid Sequence Diagram.": "```mermaid\nsequenceDiagram\n participant User\n participant DDGAPIWrapper\n participant asyncio\n participant futures\n participant DDGS\n participant config\n\n User->>DDGAPIWrapper: run(query, max_results, as_string)\n DDGAPIWrapper->>asyncio: get_event_loop()\n asyncio->>DDGAPIWrapper: loop\n alt config.proxy\n DDGAPIWrapper->>config: proxy\n end\n DDGAPIWrapper->>futures: Executor\n futures->>DDGAPIWrapper: executor\n DDGAPIWrapper->>DDGS: __init__(**kwargs)\n DDGAPIWrapper->>asyncio: run_in_executor(executor, _search_from_ddgs, query, max_results)\n asyncio->>DDGAPIWrapper: future\n DDGAPIWrapper->>DDGS: text(query)\n DDGS-->>DDGAPIWrapper: search results\n DDGAPIWrapper-->>asyncio: search_results\n asyncio-->>DDGAPIWrapper: await future\n DDGAPIWrapper-->>User: search results\n```", + "You are a python code to Mermaid Sequence Diagram translator in function detail#SYSTEM_MSG_END#```python\n#!/usr/bin/env python\n# -*- coding: utf-8 -*-\n\"\"\"\n@Time : 2023/5/23 18:27\n@Author : alexanderwu\n@File : search_engine_serpapi.py\n\"\"\"\nfrom typing import Any, Dict, Optional, Tuple\n\nimport aiohttp\nfrom pydantic import BaseModel, ConfigDict, Field, field_validator\n\nfrom metagpt.config2 import config\n\n\nclass SerpAPIWrapper(BaseModel):\n model_config = ConfigDict(arbitrary_types_allowed=True)\n\n search_engine: Any = None #: :meta private:\n params: dict = Field(\n default_factory=lambda: {\n \"engine\": \"google\",\n \"google_domain\": \"google.com\",\n \"gl\": \"us\",\n \"hl\": \"en\",\n }\n )\n # should add `validate_default=True` to check with default value\n serpapi_api_key: Optional[str] = Field(default=None, validate_default=True)\n aiosession: Optional[aiohttp.ClientSession] = None\n\n @field_validator(\"serpapi_api_key\", mode=\"before\")\n @classmethod\n def check_serpapi_api_key(cls, val: str):\n val = val or config.search[\"serpapi\"].api_key\n if not val:\n raise ValueError(\n \"To use, make sure you provide the serpapi_api_key when constructing an object. Alternatively, \"\n \"ensure that the environment variable SERPAPI_API_KEY is set with your API key. You can obtain \"\n \"an API key from https://serpapi.com/.\"\n )\n return val\n\n async def run(self, query, max_results: int = 8, as_string: bool = True, **kwargs: Any) -> str:\n \"\"\"Run query through SerpAPI and parse result async.\"\"\"\n result = await self.results(query, max_results)\n return self._process_response(result, as_string=as_string)\n\n async def results(self, query: str, max_results: int) -> dict:\n \"\"\"Use aiohttp to run query through SerpAPI and return the results async.\"\"\"\n\n def construct_url_and_params() -> Tuple[str, Dict[str, str]]:\n params = self.get_params(query)\n params[\"source\"] = \"python\"\n params[\"num\"] = max_results\n params[\"output\"] = \"json\"\n url = \"https://serpapi.com/search\"\n return url, params\n\n url, params = construct_url_and_params()\n if not self.aiosession:\n async with aiohttp.ClientSession() as session:\n async with session.get(url, params=params) as response:\n res = await response.json()\n else:\n async with self.aiosession.get(url, params=params) as response:\n res = await response.json()\n\n return res\n\n def get_params(self, query: str) -> Dict[str, str]:\n \"\"\"Get parameters for SerpAPI.\"\"\"\n _params = {\n \"api_key\": self.serpapi_api_key,\n \"q\": query,\n }\n params = {**self.params, **_params}\n return params\n\n @staticmethod\n def _process_response(res: dict, as_string: bool) -> str:\n \"\"\"Process response from SerpAPI.\"\"\"\n # logger.debug(res)\n focus = [\"title\", \"snippet\", \"link\"]\n get_focused = lambda x: {i: j for i, j in x.items() if i in focus}\n\n if \"error\" in res.keys():\n raise ValueError(f\"Got error from SerpAPI: {res['error']}\")\n if \"answer_box\" in res.keys() and \"answer\" in res[\"answer_box\"].keys():\n toret = res[\"answer_box\"][\"answer\"]\n elif \"answer_box\" in res.keys() and \"snippet\" in res[\"answer_box\"].keys():\n toret = res[\"answer_box\"][\"snippet\"]\n elif \"answer_box\" in res.keys() and \"snippet_highlighted_words\" in res[\"answer_box\"].keys():\n toret = res[\"answer_box\"][\"snippet_highlighted_words\"][0]\n elif \"sports_results\" in res.keys() and \"game_spotlight\" in res[\"sports_results\"].keys():\n toret = res[\"sports_results\"][\"game_spotlight\"]\n elif \"knowledge_graph\" in res.keys() and \"description\" in res[\"knowledge_graph\"].keys():\n toret = res[\"knowledge_graph\"][\"description\"]\n elif \"snippet\" in res[\"organic_results\"][0].keys():\n toret = res[\"organic_results\"][0][\"snippet\"]\n else:\n toret = \"No good search result found\"\n\n toret_l = []\n if \"answer_box\" in res.keys() and \"snippet\" in res[\"answer_box\"].keys():\n toret_l += [get_focused(res[\"answer_box\"])]\n if res.get(\"organic_results\"):\n toret_l += [get_focused(i) for i in res.get(\"organic_results\")]\n\n return str(toret) + \"\\n\" + str(toret_l) if as_string else toret_l\n\n\nif __name__ == \"__main__\":\n import fire\n\n fire.Fire(SerpAPIWrapper().run)\n\n```\n\n---\nTranslate the code above into Mermaid Sequence Diagram.": "```mermaid\nsequenceDiagram\n participant SerpAPIWrapper\n participant aiohttp\n participant session\n participant response\n participant fire\n\n Note over SerpAPIWrapper: Initialization\n SerpAPIWrapper->>SerpAPIWrapper: __init__\n\n Note over SerpAPIWrapper: Run query through SerpAPI\n SerpAPIWrapper->>SerpAPIWrapper: run(query, max_results, as_string, **kwargs)\n SerpAPIWrapper->>SerpAPIWrapper: results(query, max_results)\n SerpAPIWrapper->>SerpAPIWrapper: get_params(query)\n SerpAPIWrapper->>aiohttp: session.get(url, params)\n aiohttp->>session: get(url, params)\n session->>response: response.json()\n response-->>session: res\n session-->>aiohttp: res\n aiohttp-->>SerpAPIWrapper: res\n SerpAPIWrapper-->>SerpAPIWrapper: _process_response(result, as_string)\n\n Note over SerpAPIWrapper: Use aiohttp to run query through SerpAPI\n SerpAPIWrapper->>SerpAPIWrapper: results(query, max_results)\n SerpAPIWrapper->>SerpAPIWrapper: get_params(query)\n SerpAPIWrapper->>aiohttp: ClientSession()\n aiohttp->>session: get(url, params)\n session->>response: response.json()\n response-->>session: res\n session-->>aiohttp: res\n aiohttp-->>SerpAPIWrapper: res\n\n Note over SerpAPIWrapper: Get parameters for SerpAPI\n SerpAPIWrapper->>SerpAPIWrapper: get_params(query)\n\n Note over SerpAPIWrapper: Process response from SerpAPI\n SerpAPIWrapper->>SerpAPIWrapper: _process_response(res, as_string)\n\n Note over fire: Main function\n fire->>SerpAPIWrapper: run\n\n```", + "You are a python code to Mermaid Sequence Diagram translator in function detail#SYSTEM_MSG_END#```python\n#!/usr/bin/env python\n# -*- coding: utf-8 -*-\n\"\"\"\n@Time : 2023/5/23 18:27\n@Author : alexanderwu\n@File : search_engine_serpapi.py\n\"\"\"\nimport json\nfrom typing import Any, Dict, Optional, Tuple\n\nimport aiohttp\nfrom pydantic import BaseModel, ConfigDict, Field, field_validator\n\nfrom metagpt.config2 import config\n\n\nclass SerperWrapper(BaseModel):\n model_config = ConfigDict(arbitrary_types_allowed=True)\n\n search_engine: Any = None #: :meta private:\n payload: dict = Field(default_factory=lambda: {\"page\": 1, \"num\": 10})\n serper_api_key: Optional[str] = Field(default=None, validate_default=True)\n aiosession: Optional[aiohttp.ClientSession] = None\n\n @field_validator(\"serper_api_key\", mode=\"before\")\n @classmethod\n def check_serper_api_key(cls, val: str):\n val = val or config.search[\"serper\"].api_key\n if not val:\n raise ValueError(\n \"To use, make sure you provide the serper_api_key when constructing an object. Alternatively, \"\n \"ensure that the environment variable SERPER_API_KEY is set with your API key. You can obtain \"\n \"an API key from https://serper.dev/.\"\n )\n return val\n\n async def run(self, query: str, max_results: int = 8, as_string: bool = True, **kwargs: Any) -> str:\n \"\"\"Run query through Serper and parse result async.\"\"\"\n if isinstance(query, str):\n return self._process_response((await self.results([query], max_results))[0], as_string=as_string)\n else:\n results = [self._process_response(res, as_string) for res in await self.results(query, max_results)]\n return \"\\n\".join(results) if as_string else results\n\n async def results(self, queries: list[str], max_results: int = 8) -> dict:\n \"\"\"Use aiohttp to run query through Serper and return the results async.\"\"\"\n\n def construct_url_and_payload_and_headers() -> Tuple[str, Dict[str, str]]:\n payloads = self.get_payloads(queries, max_results)\n url = \"https://google.serper.dev/search\"\n headers = self.get_headers()\n return url, payloads, headers\n\n url, payloads, headers = construct_url_and_payload_and_headers()\n if not self.aiosession:\n async with aiohttp.ClientSession() as session:\n async with session.post(url, data=payloads, headers=headers) as response:\n res = await response.json()\n else:\n async with self.aiosession.get.post(url, data=payloads, headers=headers) as response:\n res = await response.json()\n\n return res\n\n def get_payloads(self, queries: list[str], max_results: int) -> Dict[str, str]:\n \"\"\"Get payloads for Serper.\"\"\"\n payloads = []\n for query in queries:\n _payload = {\n \"q\": query,\n \"num\": max_results,\n }\n payloads.append({**self.payload, **_payload})\n return json.dumps(payloads, sort_keys=True)\n\n def get_headers(self) -> Dict[str, str]:\n headers = {\"X-API-KEY\": self.serper_api_key, \"Content-Type\": \"application/json\"}\n return headers\n\n @staticmethod\n def _process_response(res: dict, as_string: bool = False) -> str:\n \"\"\"Process response from SerpAPI.\"\"\"\n # logger.debug(res)\n focus = [\"title\", \"snippet\", \"link\"]\n\n def get_focused(x):\n return {i: j for i, j in x.items() if i in focus}\n\n if \"error\" in res.keys():\n raise ValueError(f\"Got error from SerpAPI: {res['error']}\")\n if \"answer_box\" in res.keys() and \"answer\" in res[\"answer_box\"].keys():\n toret = res[\"answer_box\"][\"answer\"]\n elif \"answer_box\" in res.keys() and \"snippet\" in res[\"answer_box\"].keys():\n toret = res[\"answer_box\"][\"snippet\"]\n elif \"answer_box\" in res.keys() and \"snippet_highlighted_words\" in res[\"answer_box\"].keys():\n toret = res[\"answer_box\"][\"snippet_highlighted_words\"][0]\n elif \"sports_results\" in res.keys() and \"game_spotlight\" in res[\"sports_results\"].keys():\n toret = res[\"sports_results\"][\"game_spotlight\"]\n elif \"knowledge_graph\" in res.keys() and \"description\" in res[\"knowledge_graph\"].keys():\n toret = res[\"knowledge_graph\"][\"description\"]\n elif \"snippet\" in res[\"organic\"][0].keys():\n toret = res[\"organic\"][0][\"snippet\"]\n else:\n toret = \"No good search result found\"\n\n toret_l = []\n if \"answer_box\" in res.keys() and \"snippet\" in res[\"answer_box\"].keys():\n toret_l += [get_focused(res[\"answer_box\"])]\n if res.get(\"organic\"):\n toret_l += [get_focused(i) for i in res.get(\"organic\")]\n\n return str(toret) + \"\\n\" + str(toret_l) if as_string else toret_l\n\n\nif __name__ == \"__main__\":\n import fire\n\n fire.Fire(SerperWrapper().run)\n\n```\n\n---\nTranslate the code above into Mermaid Sequence Diagram.": "```mermaid\nsequenceDiagram\n participant User\n participant SerperWrapper\n participant aiohttp\n participant pydantic\n participant config\n\n User ->> SerperWrapper: run(query: str, max_results: int, as_string: bool, **kwargs: Any)\n SerperWrapper ->> SerperWrapper: _process_response(response: dict, as_string: bool)\n SerperWrapper ->> SerperWrapper: get_payloads(queries: list[str], max_results: int)\n SerperWrapper ->> SerperWrapper: get_headers()\n SerperWrapper ->> aiohttp: ClientSession.post(url, data, headers)\n aiohttp ->> SerperWrapper: response\n SerperWrapper ->> aiohttp: ClientSession.get.post(url, data, headers)\n aiohttp ->> SerperWrapper: response\n SerperWrapper ->> aiohttp: ClientSession.post(url, data, headers)\n aiohttp ->> SerperWrapper: response\n SerperWrapper ->> User: str\n```", + "You are a python code to Mermaid Sequence Diagram translator in function detail#SYSTEM_MSG_END#```python\n#!/usr/bin/env python\n# -*- coding: utf-8 -*-\nfrom __future__ import annotations\n\nimport asyncio\nimport json\nfrom concurrent import futures\nfrom typing import Optional\nfrom urllib.parse import urlparse\n\nimport httplib2\nfrom pydantic import BaseModel, ConfigDict, Field, field_validator\n\nfrom metagpt.config2 import config\nfrom metagpt.logs import logger\n\ntry:\n from googleapiclient.discovery import build\n from googleapiclient.errors import HttpError\nexcept ImportError:\n raise ImportError(\n \"To use this module, you should have the `google-api-python-client` Python package installed. \"\n \"You can install it by running the command: `pip install -e.[search-google]`\"\n )\n\n\nclass GoogleAPIWrapper(BaseModel):\n model_config = ConfigDict(arbitrary_types_allowed=True)\n\n google_api_key: Optional[str] = Field(default=None, validate_default=True)\n google_cse_id: Optional[str] = Field(default=None, validate_default=True)\n loop: Optional[asyncio.AbstractEventLoop] = None\n executor: Optional[futures.Executor] = None\n\n @field_validator(\"google_api_key\", mode=\"before\")\n @classmethod\n def check_google_api_key(cls, val: str):\n val = val or config.search[\"google\"].api_key\n if not val:\n raise ValueError(\n \"To use, make sure you provide the google_api_key when constructing an object. Alternatively, \"\n \"ensure that the environment variable GOOGLE_API_KEY is set with your API key. You can obtain \"\n \"an API key from https://console.cloud.google.com/apis/credentials.\"\n )\n return val\n\n @field_validator(\"google_cse_id\", mode=\"before\")\n @classmethod\n def check_google_cse_id(cls, val: str):\n val = val or config.search[\"google\"].cse_id\n if not val:\n raise ValueError(\n \"To use, make sure you provide the google_cse_id when constructing an object. Alternatively, \"\n \"ensure that the environment variable GOOGLE_CSE_ID is set with your API key. You can obtain \"\n \"an API key from https://programmablesearchengine.google.com/controlpanel/create.\"\n )\n return val\n\n @property\n def google_api_client(self):\n build_kwargs = {\"developerKey\": self.google_api_key}\n if config.proxy:\n parse_result = urlparse(config.proxy)\n proxy_type = parse_result.scheme\n if proxy_type == \"https\":\n proxy_type = \"http\"\n build_kwargs[\"http\"] = httplib2.Http(\n proxy_info=httplib2.ProxyInfo(\n getattr(httplib2.socks, f\"PROXY_TYPE_{proxy_type.upper()}\"),\n parse_result.hostname,\n parse_result.port,\n ),\n )\n service = build(\"customsearch\", \"v1\", **build_kwargs)\n return service.cse()\n\n async def run(\n self,\n query: str,\n max_results: int = 8,\n as_string: bool = True,\n focus: list[str] | None = None,\n ) -> str | list[dict]:\n \"\"\"Return the results of a Google search using the official Google API.\n\n Args:\n query: The search query.\n max_results: The number of results to return.\n as_string: A boolean flag to determine the return type of the results. If True, the function will\n return a formatted string with the search results. If False, it will return a list of dictionaries\n containing detailed information about each search result.\n focus: Specific information to be focused on from each search result.\n\n Returns:\n The results of the search.\n \"\"\"\n loop = self.loop or asyncio.get_event_loop()\n future = loop.run_in_executor(\n self.executor, self.google_api_client.list(q=query, num=max_results, cx=self.google_cse_id).execute\n )\n try:\n result = await future\n # Extract the search result items from the response\n search_results = result.get(\"items\", [])\n\n except HttpError as e:\n # Handle errors in the API call\n logger.exception(f\"fail to search {query} for {e}\")\n search_results = []\n\n focus = focus or [\"snippet\", \"link\", \"title\"]\n details = [{i: j for i, j in item_dict.items() if i in focus} for item_dict in search_results]\n # Return the list of search result URLs\n if as_string:\n return safe_google_results(details)\n\n return details\n\n\ndef safe_google_results(results: str | list) -> str:\n \"\"\"Return the results of a google search in a safe format.\n\n Args:\n results: The search results.\n\n Returns:\n The results of the search.\n \"\"\"\n if isinstance(results, list):\n safe_message = json.dumps([result for result in results])\n else:\n safe_message = results.encode(\"utf-8\", \"ignore\").decode(\"utf-8\")\n return safe_message\n\n\nif __name__ == \"__main__\":\n import fire\n\n fire.Fire(GoogleAPIWrapper().run)\n\n```\n\n---\nTranslate the code above into Mermaid Sequence Diagram.": "```mermaid\nsequenceDiagram\n participant BaseModel\n participant ConfigDict\n participant Field\n participant field_validator\n participant asyncio\n participant futures\n participant urlparse\n participant httplib2\n participant logger\n participant build\n participant HttpError\n participant GoogleAPIWrapper\n participant fire\n\n BaseModel ->> ConfigDict: model_config\n BaseModel ->> Field: google_api_key\n BaseModel ->> Field: google_cse_id\n BaseModel ->> Field: loop\n BaseModel ->> Field: executor\n Field ->> field_validator: check_google_api_key\n Field ->> field_validator: check_google_cse_id\n GoogleAPIWrapper ->> urlparse: parse_result\n urlparse ->> httplib2: Http\n urlparse ->> httplib2: ProxyInfo\n httplib2 ->> logger: exception\n build ->> GoogleAPIWrapper: google_api_client\n GoogleAPIWrapper ->> asyncio: run\n asyncio ->> futures: run_in_executor\n futures ->> GoogleAPIWrapper: google_api_client.list\n GoogleAPIWrapper ->> HttpError: HttpError\n HttpError ->> logger: exception\n GoogleAPIWrapper ->> safe_google_results: safe_google_results\n safe_google_results -->> GoogleAPIWrapper: safe_message\n GoogleAPIWrapper -->> fire: run\n```", + "You are a python code to Mermaid Sequence Diagram translator in function detail#SYSTEM_MSG_END#```python\nNone\n```\n\n---\nTranslate the code above into Mermaid Sequence Diagram.": "```mermaid\nsequenceDiagram\n participant PythonCode\n PythonCode->>Mermaid: None\n```", + "You are a python code to Mermaid Sequence Diagram translator in function detail#SYSTEM_MSG_END#```python\n#!/usr/bin/env python\n\"\"\"\n@Modified By: mashenquan, 2023/8/22. A definition has been provided for the return value of _think: returning false indicates that further reasoning cannot continue.\n@Modified By: mashenquan, 2023-11-1. According to Chapter 2.2.1 and 2.2.2 of RFC 116, change the data type of\n the `cause_by` value in the `Message` to a string to support the new message distribution feature.\n\"\"\"\n\nimport asyncio\nimport re\n\nfrom pydantic import BaseModel\n\nfrom metagpt.actions import Action, CollectLinks, ConductResearch, WebBrowseAndSummarize\nfrom metagpt.actions.research import get_research_system_text\nfrom metagpt.const import RESEARCH_PATH\nfrom metagpt.logs import logger\nfrom metagpt.roles.role import Role, RoleReactMode\nfrom metagpt.schema import Message\n\n\nclass Report(BaseModel):\n topic: str\n links: dict[str, list[str]] = None\n summaries: list[tuple[str, str]] = None\n content: str = \"\"\n\n\nclass Researcher(Role):\n name: str = \"David\"\n profile: str = \"Researcher\"\n goal: str = \"Gather information and conduct research\"\n constraints: str = \"Ensure accuracy and relevance of information\"\n language: str = \"en-us\"\n\n def __init__(self, **kwargs):\n super().__init__(**kwargs)\n self.set_actions(\n [CollectLinks(name=self.name), WebBrowseAndSummarize(name=self.name), ConductResearch(name=self.name)]\n )\n self._set_react_mode(react_mode=RoleReactMode.BY_ORDER.value)\n if self.language not in (\"en-us\", \"zh-cn\"):\n logger.warning(f\"The language `{self.language}` has not been tested, it may not work.\")\n\n async def _think(self) -> bool:\n if self.rc.todo is None:\n self._set_state(0)\n return True\n\n if self.rc.state + 1 < len(self.states):\n self._set_state(self.rc.state + 1)\n else:\n self.set_todo(None)\n return False\n\n async def _act(self) -> Message:\n logger.info(f\"{self._setting}: to do {self.rc.todo}({self.rc.todo.name})\")\n todo = self.rc.todo\n msg = self.rc.memory.get(k=1)[0]\n if isinstance(msg.instruct_content, Report):\n instruct_content = msg.instruct_content\n topic = instruct_content.topic\n else:\n topic = msg.content\n\n research_system_text = self.research_system_text(topic, todo)\n if isinstance(todo, CollectLinks):\n links = await todo.run(topic, 4, 4)\n ret = Message(\n content=\"\", instruct_content=Report(topic=topic, links=links), role=self.profile, cause_by=todo\n )\n elif isinstance(todo, WebBrowseAndSummarize):\n links = instruct_content.links\n todos = (todo.run(*url, query=query, system_text=research_system_text) for (query, url) in links.items())\n summaries = await asyncio.gather(*todos)\n summaries = list((url, summary) for i in summaries for (url, summary) in i.items() if summary)\n ret = Message(\n content=\"\", instruct_content=Report(topic=topic, summaries=summaries), role=self.profile, cause_by=todo\n )\n else:\n summaries = instruct_content.summaries\n summary_text = \"\\n---\\n\".join(f\"url: {url}\\nsummary: {summary}\" for (url, summary) in summaries)\n content = await self.rc.todo.run(topic, summary_text, system_text=research_system_text)\n ret = Message(\n content=\"\",\n instruct_content=Report(topic=topic, content=content),\n role=self.profile,\n cause_by=self.rc.todo,\n )\n self.rc.memory.add(ret)\n return ret\n\n def research_system_text(self, topic, current_task: Action) -> str:\n \"\"\"BACKWARD compatible\n This allows sub-class able to define its own system prompt based on topic.\n return the previous implementation to have backward compatible\n Args:\n topic:\n language:\n\n Returns: str\n \"\"\"\n return get_research_system_text(topic, self.language)\n\n async def react(self) -> Message:\n msg = await super().react()\n report = msg.instruct_content\n self.write_report(report.topic, report.content)\n return msg\n\n def write_report(self, topic: str, content: str):\n filename = re.sub(r'[\\\\/:\"*?<>|]+', \" \", topic)\n filename = filename.replace(\"\\n\", \"\")\n if not RESEARCH_PATH.exists():\n RESEARCH_PATH.mkdir(parents=True)\n filepath = RESEARCH_PATH / f\"{filename}.md\"\n filepath.write_text(content)\n\n\nif __name__ == \"__main__\":\n import fire\n\n async def main(topic: str, language=\"en-us\"):\n role = Researcher(language=language)\n await role.run(topic)\n\n fire.Fire(main)\n\n```\n\n---\nTranslate the code above into Mermaid Sequence Diagram.": "```mermaid\nsequenceDiagram\n participant Role\n participant CollectLinks\n participant WebBrowseAndSummarize\n participant ConductResearch\n participant Message\n participant Report\n\n Role->>Role: Gather information and conduct research\n Role->>Role: Ensure accuracy and relevance of information\n Role->>Role: Set react mode to BY_ORDER\n\n Role->>Role: to do {todo}({todo.name})\n Role->>CollectLinks: run(topic, 4, 4)\n CollectLinks-->>Role: links\n Role->>Message: Report(topic, links)\n Role->>Role: Add message to memory\n\n Role->>WebBrowseAndSummarize: run(url, query, system_text)\n WebBrowseAndSummarize-->>Role: summaries\n Role->>Message: Report(topic, summaries)\n Role->>Role: Add message to memory\n\n Role->>ConductResearch: run(topic, summary_text, system_text)\n ConductResearch-->>Role: content\n Role->>Message: Report(topic, content)\n Role->>Role: Add message to memory\n\n Role->>Role: React\n Role->>Role: Write report to file\n```", + "You are a python code to Mermaid Sequence Diagram translator in function detail#SYSTEM_MSG_END#```python\n\"\"\"Code Docstring Generator.\n\nThis script provides a tool to automatically generate docstrings for Python code. It uses the specified style to create\ndocstrings for the given code and system text.\n\nUsage:\n python3 -m metagpt.actions.write_docstring [--overwrite] [--style=]\n\nArguments:\n filename The path to the Python file for which you want to generate docstrings.\n\nOptions:\n --overwrite If specified, overwrite the original file with the code containing docstrings.\n --style= Specify the style of the generated docstrings.\n Valid values: 'google', 'numpy', or 'sphinx'.\n Default: 'google'\n\nExample:\n python3 -m metagpt.actions.write_docstring ./metagpt/startup.py --overwrite False --style=numpy\n\nThis script uses the 'fire' library to create a command-line interface. It generates docstrings for the given Python code using\nthe specified docstring style and adds them to the code.\n\"\"\"\nfrom __future__ import annotations\n\nimport ast\nfrom pathlib import Path\nfrom typing import Literal, Optional\n\nfrom metagpt.actions.action import Action\nfrom metagpt.utils.common import OutputParser, aread, awrite\nfrom metagpt.utils.pycst import merge_docstring\n\nPYTHON_DOCSTRING_SYSTEM = \"\"\"### Requirements\n1. Add docstrings to the given code following the {style} style.\n2. Replace the function body with an Ellipsis object(...) to reduce output.\n3. If the types are already annotated, there is no need to include them in the docstring.\n4. Extract only class, function or the docstrings for the module parts from the given Python code, avoiding any other text.\n\n### Input Example\n```python\ndef function_with_pep484_type_annotations(param1: int) -> bool:\n return isinstance(param1, int)\n\nclass ExampleError(Exception):\n def __init__(self, msg: str):\n self.msg = msg\n```\n\n### Output Example\n```python\n{example}\n```\n\"\"\"\n\n# https://www.sphinx-doc.org/en/master/usage/extensions/napoleon.html\n\nPYTHON_DOCSTRING_EXAMPLE_GOOGLE = '''\ndef function_with_pep484_type_annotations(param1: int) -> bool:\n \"\"\"Example function with PEP 484 type annotations.\n\n Extended description of function.\n\n Args:\n param1: The first parameter.\n\n Returns:\n The return value. True for success, False otherwise.\n \"\"\"\n ...\n\nclass ExampleError(Exception):\n \"\"\"Exceptions are documented in the same way as classes.\n\n The __init__ method was documented in the class level docstring.\n\n Args:\n msg: Human readable string describing the exception.\n\n Attributes:\n msg: Human readable string describing the exception.\n \"\"\"\n ...\n'''\n\nPYTHON_DOCSTRING_EXAMPLE_NUMPY = '''\ndef function_with_pep484_type_annotations(param1: int) -> bool:\n \"\"\"\n Example function with PEP 484 type annotations.\n\n Extended description of function.\n\n Parameters\n ----------\n param1\n The first parameter.\n\n Returns\n -------\n bool\n The return value. True for success, False otherwise.\n \"\"\"\n ...\n\nclass ExampleError(Exception):\n \"\"\"\n Exceptions are documented in the same way as classes.\n\n The __init__ method was documented in the class level docstring.\n\n Parameters\n ----------\n msg\n Human readable string describing the exception.\n\n Attributes\n ----------\n msg\n Human readable string describing the exception.\n \"\"\"\n ...\n'''\n\nPYTHON_DOCSTRING_EXAMPLE_SPHINX = '''\ndef function_with_pep484_type_annotations(param1: int) -> bool:\n \"\"\"Example function with PEP 484 type annotations.\n\n Extended description of function.\n\n :param param1: The first parameter.\n :type param1: int\n\n :return: The return value. True for success, False otherwise.\n :rtype: bool\n \"\"\"\n ...\n\nclass ExampleError(Exception):\n \"\"\"Exceptions are documented in the same way as classes.\n\n The __init__ method was documented in the class level docstring.\n\n :param msg: Human-readable string describing the exception.\n :type msg: str\n \"\"\"\n ...\n'''\n\n_python_docstring_style = {\n \"google\": PYTHON_DOCSTRING_EXAMPLE_GOOGLE.strip(),\n \"numpy\": PYTHON_DOCSTRING_EXAMPLE_NUMPY.strip(),\n \"sphinx\": PYTHON_DOCSTRING_EXAMPLE_SPHINX.strip(),\n}\n\n\nclass WriteDocstring(Action):\n \"\"\"This class is used to write docstrings for code.\n\n Attributes:\n desc: A string describing the action.\n \"\"\"\n\n desc: str = \"Write docstring for code.\"\n i_context: Optional[str] = None\n\n async def run(\n self,\n code: str,\n system_text: str = PYTHON_DOCSTRING_SYSTEM,\n style: Literal[\"google\", \"numpy\", \"sphinx\"] = \"google\",\n ) -> str:\n \"\"\"Writes docstrings for the given code and system text in the specified style.\n\n Args:\n code: A string of Python code.\n system_text: A string of system text.\n style: A string specifying the style of the docstring. Can be 'google', 'numpy', or 'sphinx'.\n\n Returns:\n The Python code with docstrings added.\n \"\"\"\n system_text = system_text.format(style=style, example=_python_docstring_style[style])\n simplified_code = _simplify_python_code(code)\n documented_code = await self._aask(f\"```python\\n{simplified_code}\\n```\", [system_text])\n documented_code = OutputParser.parse_python_code(documented_code)\n return merge_docstring(code, documented_code)\n\n @staticmethod\n async def write_docstring(\n filename: str | Path, overwrite: bool = False, style: Literal[\"google\", \"numpy\", \"sphinx\"] = \"google\"\n ) -> str:\n data = await aread(str(filename))\n code = await WriteDocstring().run(data, style=style)\n if overwrite:\n await awrite(filename, code)\n return code\n\n\ndef _simplify_python_code(code: str) -> None:\n \"\"\"Simplifies the given Python code by removing expressions and the last if statement.\n\n Args:\n code: A string of Python code.\n\n Returns:\n The simplified Python code.\n \"\"\"\n code_tree = ast.parse(code)\n code_tree.body = [i for i in code_tree.body if not isinstance(i, ast.Expr)]\n if isinstance(code_tree.body[-1], ast.If):\n code_tree.body.pop()\n return ast.unparse(code_tree)\n\n\nif __name__ == \"__main__\":\n import fire\n\n fire.Fire(WriteDocstring.write_docstring)\n\n```\n\n---\nTranslate the code above into Mermaid Sequence Diagram.": "```mermaid\nsequenceDiagram\n participant User\n participant \"WriteDocstring\" as WD\n participant \"OutputParser\" as OP\n participant \"aread\" as AR\n participant \"awrite\" as AW\n\n User ->> WD: write_docstring(filename, overwrite, style)\n WD ->> AR: aread(filename)\n AR -->> WD: data\n WD ->> WD: run(data, style)\n WD ->> OP: parse_python_code(documented_code)\n OP -->> WD: documented_code\n WD ->> WD: merge_docstring(code, documented_code)\n WD ->> AW: awrite(filename, code)\n AW -->> WD: code\n WD -->> User: code\n```", + "\n## context\n\n### Project Name\n20240110221009\n\n### Original Requirements\n['开发一个基于大语言模型与私有知识库的搜索引擎，希望可以基于大语言模型进行搜索总结']\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"zh_cn\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"开发一个基于大语言模型与私有知识库的搜索引擎，希望可以基于大语言模型进行搜索总结\",\n \"Product Goals\": [],\n \"User Stories\": [],\n \"Competitive Analysis\": [],\n \"Competitive Quadrant Chart\": \"\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [],\n \"UI Design draft\": \"\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]", + "\n## context\n\n### Project Name\n20240110221525\n\n### Original Requirements\n['需要一个基于LLM做总结的搜索引擎']\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"zh_cn\",\n \"Programming Language\": \"LLM\",\n \"Original Requirements\": \"需要一个基于LLM做总结的搜索引擎\",\n \"Product Goals\": [],\n \"User Stories\": [],\n \"Competitive Analysis\": [],\n \"Competitive Quadrant Chart\": \"\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [],\n \"UI Design draft\": \"\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]", + "\n## context\n\n### Project Name\n20240110221737\n\n### Original Requirements\n['开发一个基于大语言模型与私有知识库的搜索引擎，希望可以基于大语言模型进行搜索总结']\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"zh_cn\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"开发一个基于大语言模型与私有知识库的搜索引擎，希望可以基于大语言模型进行搜索总结\",\n \"Product Goals\": [],\n \"User Stories\": [],\n \"Competitive Analysis\": [],\n \"Competitive Quadrant Chart\": \"\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [],\n \"UI Design draft\": \"\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]", + "\n## context\n\n### Project Name\n20240110221737\n\n### Original Requirements\n['']\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"\",\n \"Product Goals\": [],\n \"User Stories\": [],\n \"Competitive Analysis\": [],\n \"Competitive Quadrant Chart\": \"\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [],\n \"UI Design draft\": \"\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]", + "\n## context\n\n### Project Name\n20240111154514\n\n### Original Requirements\n['需要一个基于LLM做总结的搜索引擎']\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"zh_cn\",\n \"Programming Language\": \"LLM\",\n \"Original Requirements\": \"需要一个基于LLM做总结的搜索引擎\",\n \"Product Goals\": [],\n \"User Stories\": [],\n \"Competitive Analysis\": [],\n \"Competitive Quadrant Chart\": \"\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [],\n \"UI Design draft\": \"\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]", + "You are a python code to Mermaid Sequence Diagram translator in function detail#SYSTEM_MSG_END#```python\n#!/usr/bin/env python\n# -*- coding: utf-8 -*-\n\"\"\"\n@Time : 2023/5/23 18:27\n@Author : alexanderwu\n@File : search_engine_serpapi.py\n\"\"\"\nfrom typing import Any, Dict, Optional, Tuple\n\nimport aiohttp\nfrom pydantic import BaseModel, ConfigDict, Field, field_validator\n\nfrom metagpt.config2 import config\n\n\nclass SerpAPIWrapper(BaseModel):\n model_config = ConfigDict(arbitrary_types_allowed=True)\n\n search_engine: Any = None #: :meta private:\n params: dict = Field(\n default_factory=lambda: {\n \"engine\": \"google\",\n \"google_domain\": \"google.com\",\n \"gl\": \"us\",\n \"hl\": \"en\",\n }\n )\n # should add `validate_default=True` to check with default value\n serpapi_api_key: Optional[str] = Field(default=None, validate_default=True)\n aiosession: Optional[aiohttp.ClientSession] = None\n\n @field_validator(\"serpapi_api_key\", mode=\"before\")\n @classmethod\n def check_serpapi_api_key(cls, val: str):\n val = val or config.search.api_key\n if not val:\n raise ValueError(\n \"To use, make sure you provide the serpapi_api_key when constructing an object. Alternatively, \"\n \"ensure that the environment variable SERPAPI_API_KEY is set with your API key. You can obtain \"\n \"an API key from https://serpapi.com/.\"\n )\n return val\n\n async def run(self, query, max_results: int = 8, as_string: bool = True, **kwargs: Any) -> str:\n \"\"\"Run query through SerpAPI and parse result async.\"\"\"\n result = await self.results(query, max_results)\n return self._process_response(result, as_string=as_string)\n\n async def results(self, query: str, max_results: int) -> dict:\n \"\"\"Use aiohttp to run query through SerpAPI and return the results async.\"\"\"\n\n def construct_url_and_params() -> Tuple[str, Dict[str, str]]:\n params = self.get_params(query)\n params[\"source\"] = \"python\"\n params[\"num\"] = max_results\n params[\"output\"] = \"json\"\n url = \"https://serpapi.com/search\"\n return url, params\n\n url, params = construct_url_and_params()\n if not self.aiosession:\n async with aiohttp.ClientSession() as session:\n async with session.get(url, params=params) as response:\n res = await response.json()\n else:\n async with self.aiosession.get(url, params=params) as response:\n res = await response.json()\n\n return res\n\n def get_params(self, query: str) -> Dict[str, str]:\n \"\"\"Get parameters for SerpAPI.\"\"\"\n _params = {\n \"api_key\": self.serpapi_api_key,\n \"q\": query,\n }\n params = {**self.params, **_params}\n return params\n\n @staticmethod\n def _process_response(res: dict, as_string: bool) -> str:\n \"\"\"Process response from SerpAPI.\"\"\"\n # logger.debug(res)\n focus = [\"title\", \"snippet\", \"link\"]\n get_focused = lambda x: {i: j for i, j in x.items() if i in focus}\n\n if \"error\" in res.keys():\n raise ValueError(f\"Got error from SerpAPI: {res['error']}\")\n if \"answer_box\" in res.keys() and \"answer\" in res[\"answer_box\"].keys():\n toret = res[\"answer_box\"][\"answer\"]\n elif \"answer_box\" in res.keys() and \"snippet\" in res[\"answer_box\"].keys():\n toret = res[\"answer_box\"][\"snippet\"]\n elif \"answer_box\" in res.keys() and \"snippet_highlighted_words\" in res[\"answer_box\"].keys():\n toret = res[\"answer_box\"][\"snippet_highlighted_words\"][0]\n elif \"sports_results\" in res.keys() and \"game_spotlight\" in res[\"sports_results\"].keys():\n toret = res[\"sports_results\"][\"game_spotlight\"]\n elif \"knowledge_graph\" in res.keys() and \"description\" in res[\"knowledge_graph\"].keys():\n toret = res[\"knowledge_graph\"][\"description\"]\n elif \"snippet\" in res[\"organic_results\"][0].keys():\n toret = res[\"organic_results\"][0][\"snippet\"]\n else:\n toret = \"No good search result found\"\n\n toret_l = []\n if \"answer_box\" in res.keys() and \"snippet\" in res[\"answer_box\"].keys():\n toret_l += [get_focused(res[\"answer_box\"])]\n if res.get(\"organic_results\"):\n toret_l += [get_focused(i) for i in res.get(\"organic_results\")]\n\n return str(toret) + \"\\n\" + str(toret_l) if as_string else toret_l\n\n\nif __name__ == \"__main__\":\n import fire\n\n fire.Fire(SerpAPIWrapper().run)\n\n```\n\n---\nTranslate the code above into Mermaid Sequence Diagram.": "```mermaid\nsequenceDiagram\n participant SerpAPIWrapper\n participant aiohttp\n participant config\n participant session\n participant response\n participant fire\n\n Note over SerpAPIWrapper: Initialization\n SerpAPIWrapper->>config: get search.api_key\n config-->>SerpAPIWrapper: return search.api_key\n SerpAPIWrapper->>SerpAPIWrapper: check_serpapi_api_key()\n SerpAPIWrapper->>SerpAPIWrapper: get_params()\n SerpAPIWrapper->>SerpAPIWrapper: results()\n SerpAPIWrapper->>aiohttp: ClientSession()\n aiohttp->>session: get(url, params)\n session->>response: json()\n response-->>session: return json response\n session-->>aiohttp: return json response\n aiohttp-->>SerpAPIWrapper: return json response\n SerpAPIWrapper-->>SerpAPIWrapper: _process_response()\n SerpAPIWrapper-->>fire: run()\n```", + "You are a python code to Mermaid Sequence Diagram translator in function detail#SYSTEM_MSG_END#```python\n#!/usr/bin/env python\n# -*- coding: utf-8 -*-\n\"\"\"\n@Time : 2023/5/23 18:27\n@Author : alexanderwu\n@File : search_engine_serpapi.py\n\"\"\"\nimport json\nfrom typing import Any, Dict, Optional, Tuple\n\nimport aiohttp\nfrom pydantic import BaseModel, ConfigDict, Field, field_validator\n\nfrom metagpt.config2 import config\n\n\nclass SerperWrapper(BaseModel):\n model_config = ConfigDict(arbitrary_types_allowed=True)\n\n search_engine: Any = None #: :meta private:\n payload: dict = Field(default_factory=lambda: {\"page\": 1, \"num\": 10})\n serper_api_key: Optional[str] = Field(default=None, validate_default=True)\n aiosession: Optional[aiohttp.ClientSession] = None\n\n @field_validator(\"serper_api_key\", mode=\"before\")\n @classmethod\n def check_serper_api_key(cls, val: str):\n val = val or config.search.api_key\n if not val:\n raise ValueError(\n \"To use, make sure you provide the serper_api_key when constructing an object. Alternatively, \"\n \"ensure that the environment variable SERPER_API_KEY is set with your API key. You can obtain \"\n \"an API key from https://serper.dev/.\"\n )\n return val\n\n async def run(self, query: str, max_results: int = 8, as_string: bool = True, **kwargs: Any) -> str:\n \"\"\"Run query through Serper and parse result async.\"\"\"\n if isinstance(query, str):\n return self._process_response((await self.results([query], max_results))[0], as_string=as_string)\n else:\n results = [self._process_response(res, as_string) for res in await self.results(query, max_results)]\n return \"\\n\".join(results) if as_string else results\n\n async def results(self, queries: list[str], max_results: int = 8) -> dict:\n \"\"\"Use aiohttp to run query through Serper and return the results async.\"\"\"\n\n def construct_url_and_payload_and_headers() -> Tuple[str, Dict[str, str]]:\n payloads = self.get_payloads(queries, max_results)\n url = \"https://google.serper.dev/search\"\n headers = self.get_headers()\n return url, payloads, headers\n\n url, payloads, headers = construct_url_and_payload_and_headers()\n if not self.aiosession:\n async with aiohttp.ClientSession() as session:\n async with session.post(url, data=payloads, headers=headers) as response:\n res = await response.json()\n else:\n async with self.aiosession.get.post(url, data=payloads, headers=headers) as response:\n res = await response.json()\n\n return res\n\n def get_payloads(self, queries: list[str], max_results: int) -> Dict[str, str]:\n \"\"\"Get payloads for Serper.\"\"\"\n payloads = []\n for query in queries:\n _payload = {\n \"q\": query,\n \"num\": max_results,\n }\n payloads.append({**self.payload, **_payload})\n return json.dumps(payloads, sort_keys=True)\n\n def get_headers(self) -> Dict[str, str]:\n headers = {\"X-API-KEY\": self.serper_api_key, \"Content-Type\": \"application/json\"}\n return headers\n\n @staticmethod\n def _process_response(res: dict, as_string: bool = False) -> str:\n \"\"\"Process response from SerpAPI.\"\"\"\n # logger.debug(res)\n focus = [\"title\", \"snippet\", \"link\"]\n\n def get_focused(x):\n return {i: j for i, j in x.items() if i in focus}\n\n if \"error\" in res.keys():\n raise ValueError(f\"Got error from SerpAPI: {res['error']}\")\n if \"answer_box\" in res.keys() and \"answer\" in res[\"answer_box\"].keys():\n toret = res[\"answer_box\"][\"answer\"]\n elif \"answer_box\" in res.keys() and \"snippet\" in res[\"answer_box\"].keys():\n toret = res[\"answer_box\"][\"snippet\"]\n elif \"answer_box\" in res.keys() and \"snippet_highlighted_words\" in res[\"answer_box\"].keys():\n toret = res[\"answer_box\"][\"snippet_highlighted_words\"][0]\n elif \"sports_results\" in res.keys() and \"game_spotlight\" in res[\"sports_results\"].keys():\n toret = res[\"sports_results\"][\"game_spotlight\"]\n elif \"knowledge_graph\" in res.keys() and \"description\" in res[\"knowledge_graph\"].keys():\n toret = res[\"knowledge_graph\"][\"description\"]\n elif \"snippet\" in res[\"organic\"][0].keys():\n toret = res[\"organic\"][0][\"snippet\"]\n else:\n toret = \"No good search result found\"\n\n toret_l = []\n if \"answer_box\" in res.keys() and \"snippet\" in res[\"answer_box\"].keys():\n toret_l += [get_focused(res[\"answer_box\"])]\n if res.get(\"organic\"):\n toret_l += [get_focused(i) for i in res.get(\"organic\")]\n\n return str(toret) + \"\\n\" + str(toret_l) if as_string else toret_l\n\n\nif __name__ == \"__main__\":\n import fire\n\n fire.Fire(SerperWrapper().run)\n\n```\n\n---\nTranslate the code above into Mermaid Sequence Diagram.": "```mermaid\nsequenceDiagram\n participant User\n participant SerperWrapper\n participant aiohttp\n participant config\n\n User ->> SerperWrapper: run(query, max_results, as_string, **kwargs)\n SerperWrapper ->> SerperWrapper: _process_response(response, as_string)\n SerperWrapper ->> SerperWrapper: results(queries, max_results)\n SerperWrapper ->> aiohttp: post(url, data, headers)\n aiohttp ->> SerperWrapper: response\n SerperWrapper ->> User: return result\n SerperWrapper ->> config: search.api_key\n```", + "You are a python code to Mermaid Sequence Diagram translator in function detail#SYSTEM_MSG_END#```python\n#!/usr/bin/env python\n# -*- coding: utf-8 -*-\nfrom __future__ import annotations\n\nimport asyncio\nimport json\nfrom concurrent import futures\nfrom typing import Optional\nfrom urllib.parse import urlparse\n\nimport httplib2\nfrom pydantic import BaseModel, ConfigDict, Field, field_validator\n\nfrom metagpt.config2 import config\nfrom metagpt.logs import logger\n\ntry:\n from googleapiclient.discovery import build\n from googleapiclient.errors import HttpError\nexcept ImportError:\n raise ImportError(\n \"To use this module, you should have the `google-api-python-client` Python package installed. \"\n \"You can install it by running the command: `pip install -e.[search-google]`\"\n )\n\n\nclass GoogleAPIWrapper(BaseModel):\n model_config = ConfigDict(arbitrary_types_allowed=True)\n\n google_api_key: Optional[str] = Field(default=None, validate_default=True)\n google_cse_id: Optional[str] = Field(default=None, validate_default=True)\n loop: Optional[asyncio.AbstractEventLoop] = None\n executor: Optional[futures.Executor] = None\n\n @field_validator(\"google_api_key\", mode=\"before\")\n @classmethod\n def check_google_api_key(cls, val: str):\n val = val or config.search.api_key\n if not val:\n raise ValueError(\n \"To use, make sure you provide the google_api_key when constructing an object. Alternatively, \"\n \"ensure that the environment variable GOOGLE_API_KEY is set with your API key. You can obtain \"\n \"an API key from https://console.cloud.google.com/apis/credentials.\"\n )\n return val\n\n @field_validator(\"google_cse_id\", mode=\"before\")\n @classmethod\n def check_google_cse_id(cls, val: str):\n val = val or config.search.cse_id\n if not val:\n raise ValueError(\n \"To use, make sure you provide the google_cse_id when constructing an object. Alternatively, \"\n \"ensure that the environment variable GOOGLE_CSE_ID is set with your API key. You can obtain \"\n \"an API key from https://programmablesearchengine.google.com/controlpanel/create.\"\n )\n return val\n\n @property\n def google_api_client(self):\n build_kwargs = {\"developerKey\": self.google_api_key}\n if config.proxy:\n parse_result = urlparse(config.proxy)\n proxy_type = parse_result.scheme\n if proxy_type == \"https\":\n proxy_type = \"http\"\n build_kwargs[\"http\"] = httplib2.Http(\n proxy_info=httplib2.ProxyInfo(\n getattr(httplib2.socks, f\"PROXY_TYPE_{proxy_type.upper()}\"),\n parse_result.hostname,\n parse_result.port,\n ),\n )\n service = build(\"customsearch\", \"v1\", **build_kwargs)\n return service.cse()\n\n async def run(\n self,\n query: str,\n max_results: int = 8,\n as_string: bool = True,\n focus: list[str] | None = None,\n ) -> str | list[dict]:\n \"\"\"Return the results of a Google search using the official Google API.\n\n Args:\n query: The search query.\n max_results: The number of results to return.\n as_string: A boolean flag to determine the return type of the results. If True, the function will\n return a formatted string with the search results. If False, it will return a list of dictionaries\n containing detailed information about each search result.\n focus: Specific information to be focused on from each search result.\n\n Returns:\n The results of the search.\n \"\"\"\n loop = self.loop or asyncio.get_event_loop()\n future = loop.run_in_executor(\n self.executor, self.google_api_client.list(q=query, num=max_results, cx=self.google_cse_id).execute\n )\n try:\n result = await future\n # Extract the search result items from the response\n search_results = result.get(\"items\", [])\n\n except HttpError as e:\n # Handle errors in the API call\n logger.exception(f\"fail to search {query} for {e}\")\n search_results = []\n\n focus = focus or [\"snippet\", \"link\", \"title\"]\n details = [{i: j for i, j in item_dict.items() if i in focus} for item_dict in search_results]\n # Return the list of search result URLs\n if as_string:\n return safe_google_results(details)\n\n return details\n\n\ndef safe_google_results(results: str | list) -> str:\n \"\"\"Return the results of a google search in a safe format.\n\n Args:\n results: The search results.\n\n Returns:\n The results of the search.\n \"\"\"\n if isinstance(results, list):\n safe_message = json.dumps([result for result in results])\n else:\n safe_message = results.encode(\"utf-8\", \"ignore\").decode(\"utf-8\")\n return safe_message\n\n\nif __name__ == \"__main__\":\n import fire\n\n fire.Fire(GoogleAPIWrapper().run)\n\n```\n\n---\nTranslate the code above into Mermaid Sequence Diagram.": "```mermaid\nsequenceDiagram\n participant BaseModel\n participant httplib2\n participant asyncio\n participant futures\n participant urlparse\n participant json\n participant config\n participant logger\n participant googleapiclient.discovery\n participant googleapiclient.errors\n participant fire\n\n BaseModel->>ConfigDict: model_config\n BaseModel->>Optional: google_api_key\n BaseModel->>Optional: google_cse_id\n BaseModel->>Optional: loop\n BaseModel->>Optional: executor\n BaseModel->>googleapiclient.discovery: check_google_api_key\n BaseModel->>googleapiclient.discovery: check_google_cse_id\n BaseModel->>googleapiclient.discovery: google_api_client\n BaseModel->>asyncio: run\n asyncio->>futures: run_in_executor\n futures->>googleapiclient.discovery: list\n googleapiclient.discovery->>googleapiclient.discovery: execute\n googleapiclient.discovery-->>futures: result\n futures-->>asyncio: result\n asyncio-->>BaseModel: result\n BaseModel-->>BaseModel: safe_google_results\n BaseModel-->>BaseModel: run\n BaseModel-->>fire: run\n```", + "\n## context\n\n### Project Name\n20240111154819\n\n### Original Requirements\n['开发一个基于大语言模型与私有知识库的搜索引擎，希望可以基于大语言模型进行搜索总结']\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"zh_cn\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"开发一个基于大语言模型与私有知识库的搜索引擎，希望可以基于大语言模型进行搜索总结\",\n \"Product Goals\": [],\n \"User Stories\": [],\n \"Competitive Analysis\": [],\n \"Competitive Quadrant Chart\": \"\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [],\n \"UI Design draft\": \"\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]", + "\n## context\n\n### Project Name\n20240111154819\n\n### Original Requirements\n['']\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"\",\n \"Product Goals\": [],\n \"User Stories\": [],\n \"Competitive Analysis\": [],\n \"Competitive Quadrant Chart\": \"\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [],\n \"UI Design draft\": \"\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]", + "\n## context\n\n### Project Name\n20240111180901\n\n### Original Requirements\n['需要一个基于LLM做总结的搜索引擎']\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"zh_cn\",\n \"Programming Language\": \"LLM\",\n \"Original Requirements\": \"需要一个基于LLM做总结的搜索引擎\",\n \"Product Goals\": [],\n \"User Stories\": [],\n \"Competitive Analysis\": [],\n \"Competitive Quadrant Chart\": \"\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [],\n \"UI Design draft\": \"\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]", + "\n## context\n\n### Project Name\n20240111181214\n\n### Original Requirements\n['需要一个基于LLM做总结的搜索引擎']\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"zh_cn\",\n \"Programming Language\": \"LLM\",\n \"Original Requirements\": \"需要一个基于LLM做总结的搜索引擎\",\n \"Product Goals\": [],\n \"User Stories\": [],\n \"Competitive Analysis\": [],\n \"Competitive Quadrant Chart\": \"\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [],\n \"UI Design draft\": \"\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]", + "\n## context\n\n### Project Name\n20240111181426\n\n### Original Requirements\n['开发一个基于大语言模型与私有知识库的搜索引擎，希望可以基于大语言模型进行搜索总结']\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"zh_cn\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"开发一个基于大语言模型与私有知识库的搜索引擎，希望可以基于大语言模型进行搜索总结\",\n \"Product Goals\": [\n \"提供高效的搜索功能\",\n \"整合私有知识库\",\n \"生成准确的搜索总结\"\n ],\n \"User Stories\": [\n \"作为用户，我希望能够快速找到所需信息\",\n \"作为用户，我希望搜索结果能够涵盖私有知识库内容\",\n \"作为用户，我希望搜索总结能够准确反映所需信息\"\n ],\n \"Competitive Analysis\": [\n \"搜索引擎A：搜索速度快，但不支持私有知识库整合\",\n \"搜索引擎B：支持私有知识库整合，但搜索总结不够准确\",\n \"搜索引擎C：准确的搜索总结，但不支持私有知识库整合\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"搜索引擎比较\\\"\\n x-axis \\\"低速度\\\" --> \\\"高速度\\\"\\n y-axis \\\"低准确性\\\" --> \\\"高准确性\\\"\\n quadrant-1 \\\"速度快，准确性低\\\"\\n quadrant-2 \\\"速度慢，准确性低\\\"\\n quadrant-3 \\\"速度慢，准确性高\\\"\\n quadrant-4 \\\"速度快，准确性高\\\"\\n \\\"搜索引擎A\\\": [0.8, 0.3]\\n \\\"搜索引擎B\\\": [0.4, 0.2]\\n \\\"搜索引擎C\\\": [0.2, 0.9]\\n \\\"我们的目标产品\\\": [0.7, 0.8]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"基于大语言模型的搜索功能\"\n ],\n [\n \"P0\",\n \"私有知识库整合\"\n ],\n [\n \"P1\",\n \"搜索总结生成\"\n ]\n ],\n \"UI Design draft\": \"简洁的搜索界面，包含私有知识库搜索选项。\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]", + "\n## context\n\n### Project Name\n20240111181426\n\n### Original Requirements\n['']\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"\",\n \"Product Goals\": [],\n \"User Stories\": [],\n \"Competitive Analysis\": [],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]", + "## History Messages\n0: Human: Topic: climate change. Under 80 words per message.\n\n## Actions\nLanguage: Please use the same language as Human INPUT.\nSay your opinion with emotion and don't repeat it": "I believe that climate change is a critical issue that requires urgent action. It's alarming to see the impact of human activities on the environment and the devastating consequences it has on ecosystems and communities. We need to prioritize sustainable practices and reduce our carbon footprint to mitigate the effects of climate change. It's essential for the well-being of future generations and the health of our planet.", + "## History Messages\n0: Alex(Democratic candidate): I believe that climate change is a critical issue that requires urgent action. It's alarming to see the impact of human activities on the environment and the devastating consequences it has on ecosystems and communities. We need to prioritize sustainable practices and reduce our carbon footprint to mitigate the effects of climate change. It's essential for the well-being of future generations and the health of our planet.\n\n## Actions\nLanguage: Please use the same language as Human INPUT.\nSay your opinion with emotion and don't repeat it": "I completely agree with Alex. Climate change is a critical issue that demands immediate attention. The impact of human activities on the environment is devastating, and urgent action is necessary to prioritize sustainable practices and reduce our carbon footprint. It's essential for the well-being of future generations and the health of our planet.", + "## History Messages\n0: Bob(Republican candidate): I completely agree with Alex. Climate change is a critical issue that demands immediate attention. The impact of human activities on the environment is devastating, and urgent action is necessary to prioritize sustainable practices and reduce our carbon footprint. It's essential for the well-being of future generations and the health of our planet.\n1: Alex(Democratic candidate): I believe that climate change is a critical issue that requires urgent action. It's alarming to see the impact of human activities on the environment and the devastating consequences it has on ecosystems and communities. We need to prioritize sustainable practices and reduce our carbon footprint to mitigate the effects of climate change. It's essential for the well-being of future generations and the health of our planet.\n2: Human: Topic: climate change. Under 80 words per message.\n\n## Actions\nLanguage: Please use the same language as Human INPUT.\nSay your opinion with emotion and don't repeat it": "I completely agree with both Bob and Alex. Climate change is a critical issue that demands immediate attention. The impact of human activities on the environment is devastating, and urgent action is necessary to prioritize sustainable practices and reduce our carbon footprint. It's essential for the well-being of future generations and the health of our planet.", + "\n## context\n\n### Project Name\n20240112110621\n\n### Original Requirements\n['需要一个基于LLM做总结的搜索引擎']\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"zh_cn\",\n \"Programming Language\": \"LLM\",\n \"Original Requirements\": \"需要一个基于LLM做总结的搜索引擎\",\n \"Product Goals\": [],\n \"User Stories\": [],\n \"Competitive Analysis\": [],\n \"Competitive Quadrant Chart\": \"\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [],\n \"UI Design draft\": \"\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]", + "You are a python code to Mermaid Sequence Diagram translator in function detail#SYSTEM_MSG_END#```python\n#!/usr/bin/env python\n# -*- coding: utf-8 -*-\nimport asyncio\nimport shutil\nfrom pathlib import Path\n\nimport typer\n\nfrom metagpt.config2 import config\nfrom metagpt.const import CONFIG_ROOT, METAGPT_ROOT\n\napp = typer.Typer(add_completion=False, pretty_exceptions_show_locals=False)\n\n\ndef generate_repo(\n idea,\n investment,\n n_round,\n code_review,\n run_tests,\n implement,\n project_name,\n inc,\n project_path,\n reqa_file,\n max_auto_summarize_code,\n recover_path,\n):\n \"\"\"Run the startup logic. Can be called from CLI or other Python scripts.\"\"\"\n from metagpt.roles import (\n Architect,\n Engineer,\n ProductManager,\n ProjectManager,\n QaEngineer,\n )\n from metagpt.team import Team\n\n config.update_via_cli(project_path, project_name, inc, reqa_file, max_auto_summarize_code)\n\n if not recover_path:\n company = Team()\n company.hire(\n [\n ProductManager(),\n Architect(),\n ProjectManager(),\n ]\n )\n\n if implement or code_review:\n company.hire([Engineer(n_borg=5, use_code_review=code_review)])\n\n if run_tests:\n company.hire([QaEngineer()])\n else:\n stg_path = Path(recover_path)\n if not stg_path.exists() or not str(stg_path).endswith(\"team\"):\n raise FileNotFoundError(f\"{recover_path} not exists or not endswith `team`\")\n\n company = Team.deserialize(stg_path=stg_path)\n idea = company.idea\n\n company.invest(investment)\n company.run_project(idea)\n asyncio.run(company.run(n_round=n_round))\n\n\n@app.command(\"\", help=\"Start a new project.\")\ndef startup(\n idea: str = typer.Argument(None, help=\"Your innovative idea, such as 'Create a 2048 game.'\"),\n investment: float = typer.Option(default=3.0, help=\"Dollar amount to invest in the AI company.\"),\n n_round: int = typer.Option(default=5, help=\"Number of rounds for the simulation.\"),\n code_review: bool = typer.Option(default=True, help=\"Whether to use code review.\"),\n run_tests: bool = typer.Option(default=False, help=\"Whether to enable QA for adding & running tests.\"),\n implement: bool = typer.Option(default=True, help=\"Enable or disable code implementation.\"),\n project_name: str = typer.Option(default=\"\", help=\"Unique project name, such as 'game_2048'.\"),\n inc: bool = typer.Option(default=False, help=\"Incremental mode. Use it to coop with existing repo.\"),\n project_path: str = typer.Option(\n default=\"\",\n help=\"Specify the directory path of the old version project to fulfill the incremental requirements.\",\n ),\n reqa_file: str = typer.Option(\n default=\"\", help=\"Specify the source file name for rewriting the quality assurance code.\"\n ),\n max_auto_summarize_code: int = typer.Option(\n default=0,\n help=\"The maximum number of times the 'SummarizeCode' action is automatically invoked, with -1 indicating \"\n \"unlimited. This parameter is used for debugging the workflow.\",\n ),\n recover_path: str = typer.Option(default=None, help=\"recover the project from existing serialized storage\"),\n init_config: bool = typer.Option(default=False, help=\"Initialize the configuration file for MetaGPT.\"),\n):\n \"\"\"Run a startup. Be a boss.\"\"\"\n if init_config:\n copy_config_to()\n return\n\n if idea is None:\n typer.echo(\"Missing argument 'IDEA'. Run 'metagpt --help' for more information.\")\n raise typer.Exit()\n\n return generate_repo(\n idea,\n investment,\n n_round,\n code_review,\n run_tests,\n implement,\n project_name,\n inc,\n project_path,\n reqa_file,\n max_auto_summarize_code,\n recover_path,\n )\n\n\ndef copy_config_to(config_path=METAGPT_ROOT / \"config\" / \"config2.yaml\"):\n \"\"\"Initialize the configuration file for MetaGPT.\"\"\"\n target_path = CONFIG_ROOT / \"config2.yaml\"\n\n # 创建目标目录（如果不存在）\n target_path.parent.mkdir(parents=True, exist_ok=True)\n\n # 如果目标文件已经存在，则重命名为 .bak\n if target_path.exists():\n backup_path = target_path.with_suffix(\".bak\")\n target_path.rename(backup_path)\n print(f\"Existing configuration file backed up at {backup_path}\")\n\n # 复制文件\n shutil.copy(str(config_path), target_path)\n print(f\"Configuration file initialized at {target_path}\")\n\n\nif __name__ == \"__main__\":\n app()\n\n```\n\n---\nTranslate the code above into Mermaid Sequence Diagram.": "```mermaid\nsequenceDiagram\n participant app\n participant generate_repo\n participant copy_config_to\n participant Team\n participant ProductManager\n participant Architect\n participant ProjectManager\n participant Engineer\n participant QaEngineer\n\n app -> generate_repo: startup()\n generate_repo -> config: update_via_cli()\n generate_repo -> Team: hire()\n Team -> ProductManager: hire()\n Team -> Architect: hire()\n Team -> ProjectManager: hire()\n generate_repo -> Engineer: hire()\n generate_repo -> QaEngineer: hire()\n generate_repo -> Team: invest()\n generate_repo -> Team: run_project()\n generate_repo -> Team: run()\n\n app -> copy_config_to: copy_config_to()\n copy_config_to -> config: update_via_cli()\n```", + "You are a python code to Mermaid Sequence Diagram translator in function detail#SYSTEM_MSG_END#```python\n#!/usr/bin/env python\n# -*- coding: utf-8 -*-\n\"\"\"\n@Time : 2023/12/14 11:40\n@Author : alexanderwu\n@File : write_prd_an.py\n\"\"\"\nfrom typing import List\n\nfrom metagpt.actions.action_node import ActionNode\n\nLANGUAGE = ActionNode(\n key=\"Language\",\n expected_type=str,\n instruction=\"Provide the language used in the project, typically matching the user's requirement language.\",\n example=\"en_us\",\n)\n\nPROGRAMMING_LANGUAGE = ActionNode(\n key=\"Programming Language\",\n expected_type=str,\n instruction=\"Python/JavaScript or other mainstream programming language.\",\n example=\"Python\",\n)\n\nORIGINAL_REQUIREMENTS = ActionNode(\n key=\"Original Requirements\",\n expected_type=str,\n instruction=\"Place the original user's requirements here.\",\n example=\"Create a 2048 game\",\n)\n\nPROJECT_NAME = ActionNode(\n key=\"Project Name\",\n expected_type=str,\n instruction='According to the content of \"Original Requirements,\" name the project using snake case style , '\n \"like 'game_2048' or 'simple_crm.\",\n example=\"game_2048\",\n)\n\nPRODUCT_GOALS = ActionNode(\n key=\"Product Goals\",\n expected_type=List[str],\n instruction=\"Provide up to three clear, orthogonal product goals.\",\n example=[\"Create an engaging user experience\", \"Improve accessibility, be responsive\", \"More beautiful UI\"],\n)\n\nUSER_STORIES = ActionNode(\n key=\"User Stories\",\n expected_type=List[str],\n instruction=\"Provide up to 3 to 5 scenario-based user stories.\",\n example=[\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\",\n ],\n)\n\nCOMPETITIVE_ANALYSIS = ActionNode(\n key=\"Competitive Analysis\",\n expected_type=List[str],\n instruction=\"Provide 5 to 7 competitive products.\",\n example=[\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\",\n ],\n)\n\nCOMPETITIVE_QUADRANT_CHART = ActionNode(\n key=\"Competitive Quadrant Chart\",\n expected_type=str,\n instruction=\"Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\",\n example=\"\"\"quadrantChart\n title \"Reach and engagement of campaigns\"\n x-axis \"Low Reach\" --> \"High Reach\"\n y-axis \"Low Engagement\" --> \"High Engagement\"\n quadrant-1 \"We should expand\"\n quadrant-2 \"Need to promote\"\n quadrant-3 \"Re-evaluate\"\n quadrant-4 \"May be improved\"\n \"Campaign A\": [0.3, 0.6]\n \"Campaign B\": [0.45, 0.23]\n \"Campaign C\": [0.57, 0.69]\n \"Campaign D\": [0.78, 0.34]\n \"Campaign E\": [0.40, 0.34]\n \"Campaign F\": [0.35, 0.78]\n \"Our Target Product\": [0.5, 0.6]\"\"\",\n)\n\nREQUIREMENT_ANALYSIS = ActionNode(\n key=\"Requirement Analysis\",\n expected_type=str,\n instruction=\"Provide a detailed analysis of the requirements.\",\n example=\"\",\n)\n\nREQUIREMENT_POOL = ActionNode(\n key=\"Requirement Pool\",\n expected_type=List[List[str]],\n instruction=\"List down the top-5 requirements with their priority (P0, P1, P2).\",\n example=[[\"P0\", \"The main code ...\"], [\"P0\", \"The game algorithm ...\"]],\n)\n\nUI_DESIGN_DRAFT = ActionNode(\n key=\"UI Design draft\",\n expected_type=str,\n instruction=\"Provide a simple description of UI elements, functions, style, and layout.\",\n example=\"Basic function description with a simple style and layout.\",\n)\n\nANYTHING_UNCLEAR = ActionNode(\n key=\"Anything UNCLEAR\",\n expected_type=str,\n instruction=\"Mention any aspects of the project that are unclear and try to clarify them.\",\n example=\"\",\n)\n\nISSUE_TYPE = ActionNode(\n key=\"issue_type\",\n expected_type=str,\n instruction=\"Answer BUG/REQUIREMENT. If it is a bugfix, answer BUG, otherwise answer Requirement\",\n example=\"BUG\",\n)\n\nIS_RELATIVE = ActionNode(\n key=\"is_relative\",\n expected_type=str,\n instruction=\"Answer YES/NO. If the requirement is related to the old PRD, answer YES, otherwise NO\",\n example=\"YES\",\n)\n\nREASON = ActionNode(\n key=\"reason\", expected_type=str, instruction=\"Explain the reasoning process from question to answer\", example=\"...\"\n)\n\n\nNODES = [\n LANGUAGE,\n PROGRAMMING_LANGUAGE,\n ORIGINAL_REQUIREMENTS,\n PROJECT_NAME,\n PRODUCT_GOALS,\n USER_STORIES,\n COMPETITIVE_ANALYSIS,\n COMPETITIVE_QUADRANT_CHART,\n REQUIREMENT_ANALYSIS,\n REQUIREMENT_POOL,\n UI_DESIGN_DRAFT,\n ANYTHING_UNCLEAR,\n]\n\nWRITE_PRD_NODE = ActionNode.from_children(\"WritePRD\", NODES)\nWP_ISSUE_TYPE_NODE = ActionNode.from_children(\"WP_ISSUE_TYPE\", [ISSUE_TYPE, REASON])\nWP_IS_RELATIVE_NODE = ActionNode.from_children(\"WP_IS_RELATIVE\", [IS_RELATIVE, REASON])\n\n```\n\n---\nTranslate the code above into Mermaid Sequence Diagram.": "```mermaid\nclassDef actionNode fill:#f9f,stroke:#333,stroke-width:2px;\nclassDef actionNodeTitle fill:#f9f,stroke:#333,stroke-width:2px,font-weight:bold;\nclassDef actionNodeExample fill:#f9f,stroke:#333,stroke-width:2px,font-style:italic;\n\nclass ActionNodeTitle actionNodeTitle\nclass ActionNodeExample actionNodeExample\n\nActionNodeTitle:::Language --> \"Language\"\nActionNodeExample:::Language --> \"Provide the language used in the project, typically matching the user's requirement language.\\nExample: en_us\"\n\nActionNodeTitle:::ProgrammingLanguage --> \"Programming Language\"\nActionNodeExample:::ProgrammingLanguage --> \"Python/JavaScript or other mainstream programming language.\\nExample: Python\"\n\nActionNodeTitle:::OriginalRequirements --> \"Original Requirements\"\nActionNodeExample:::OriginalRequirements --> \"Place the original user's requirements here.\\nExample: Create a 2048 game\"\n\nActionNodeTitle:::ProjectName --> \"Project Name\"\nActionNodeExample:::ProjectName --> 'According to the content of \"Original Requirements,\" name the project using snake case style , like \\'game_2048\\' or \\'simple_crm.\\nExample: game_2048'\n\nActionNodeTitle:::ProductGoals --> \"Product Goals\"\nActionNodeExample:::ProductGoals --> \"Provide up to three clear, orthogonal product goals.\\nExample:\\n- Create an engaging user experience\\n- Improve accessibility, be responsive\\n- More beautiful UI\"\n\nActionNodeTitle:::UserStories --> \"User Stories\"\nActionNodeExample:::UserStories --> \"Provide up to 3 to 5 scenario-based user stories.\\nExample:\\n- As a player, I want to be able to choose difficulty levels\\n- As a player, I want to see my score after each game\\n- As a player, I want to get restart button when I lose\\n- As a player, I want to see beautiful UI that make me feel good\\n- As a player, I want to play game via mobile phone\"\n\nActionNodeTitle:::CompetitiveAnalysis --> \"Competitive Analysis\"\nActionNodeExample:::CompetitiveAnalysis --> \"Provide 5 to 7 competitive products.\\nExample:\\n- 2048 Game A: Simple interface, lacks responsive features\\n- play2048.co: Beautiful and responsive UI with my best score shown\\n- 2048game.com: Responsive UI with my best score shown, but many ads\"\n\nActionNodeTitle:::CompetitiveQuadrantChart --> \"Competitive Quadrant Chart\"\nActionNodeExample:::CompetitiveQuadrantChart --> \"Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\\nExample:\\nquadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\"\n\nActionNodeTitle:::RequirementAnalysis --> \"Requirement Analysis\"\nActionNodeExample:::RequirementAnalysis --> \"Provide a detailed analysis of the requirements.\\nExample: \"\n\nActionNodeTitle:::RequirementPool --> \"Requirement Pool\"\nActionNodeExample:::RequirementPool --> \"List down the top-5 requirements with their priority (P0, P1, P2).\\nExample:\\n- P0: The main code ...\\n- P0: The game algorithm ...\"\n\nActionNodeTitle:::UIDesignDraft --> \"UI Design draft\"\nActionNodeExample:::UIDesignDraft --> \"Provide a simple description of UI elements, functions, style, and layout.\\nExample: Basic function description with a simple style and layout.\"\n\nActionNodeTitle:::AnythingUNCLEAR --> \"Anything UNCLEAR\"\nActionNodeExample:::AnythingUNCLEAR --> \"Mention any aspects of the project that are unclear and try to clarify them.\\nExample: \"\n\nActionNodeTitle:::issue_type --> \"issue_type\"\nActionNodeExample:::issue_type --> \"Answer BUG/REQUIREMENT. If it is a bugfix, answer BUG, otherwise answer Requirement\\nExample: BUG\"\n\nActionNodeTitle:::is_relative --> \"is_relative\"\nActionNodeExample:::is_relative --> \"Answer YES/NO. If the requirement is related to the old PRD, answer YES, otherwise NO\\nExample: YES\"\n\nActionNodeTitle:::reason --> \"reason\"\nActionNodeExample:::reason --> \"Explain the reasoning process from question to answer\\nExample: ...\"\n\nActionNodeTitle:::WritePRD --> \"WritePRD\"\nActionNodeExample:::WritePRD --> \"Language\\nProgramming Language\\nOriginal Requirements\\nProject Name\\nProduct Goals\\nUser Stories\\nCompetitive Analysis\\nCompetitive Quadrant Chart\\nRequirement Analysis\\nRequirement Pool\\nUI Design draft\\nAnything UNCLEAR\"\n\nActionNodeTitle:::WP_ISSUE_TYPE --> \"WP_ISSUE_TYPE\"\nActionNodeExample:::WP_ISSUE_TYPE --> \"issue_type\\nreason\"\n\nActionNodeTitle:::WP_IS_RELATIVE --> \"WP_IS_RELATIVE\"\nActionNodeExample:::WP_IS_RELATIVE --> \"is_relative\\nreason\"\n```", + "\n## context\n\n### Project Name\n20240112110833\n\n### Original Requirements\n['开发一个基于大语言模型与私有知识库的搜索引擎，希望可以基于大语言模型进行搜索总结']\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"zh_cn\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"开发一个基于大语言模型与私有知识库的搜索引擎，希望可以基于大语言模型进行搜索总结\",\n \"Product Goals\": [],\n \"User Stories\": [],\n \"Competitive Analysis\": [],\n \"Competitive Quadrant Chart\": \"\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [],\n \"UI Design draft\": \"\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]", + "\n## context\n\n### Project Name\n20240112110833\n\n### Original Requirements\n['']\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"\",\n \"Product Goals\": [],\n \"User Stories\": [],\n \"Competitive Analysis\": [],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]", + "\n## context\n\n### Project Name\n\n\n### Original Requirements\n需要一个基于LLM做总结的搜索引擎\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Project Name\": \"game_2048\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Project Name: # According to the content of \"Original Requirements,\" name the project using snake case style , like 'game_2048' or 'simple_crm.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"zh_cn\",\n \"Programming Language\": \"LLM\",\n \"Original Requirements\": \"需要一个基于LLM做总结的搜索引擎\",\n \"Project Name\": \"llm_summary_search_engine\",\n \"Product Goals\": [\n \"提供准确的搜索结果\",\n \"提高搜索引擎的效率\",\n \"优化用户体验\"\n ],\n \"User Stories\": [\n \"作为用户，我希望能够快速找到我需要的信息\",\n \"作为用户，我希望搜索结果准确无误\",\n \"作为用户，我希望搜索引擎能够智能推荐相关内容\"\n ],\n \"Competitive Analysis\": [\n \"搜索引擎A: 提供准确的搜索结果，但界面简陋\",\n \"搜索引擎B: 界面美观，但搜索结果不够准确\",\n \"搜索引擎C: 搜索结果准确，但速度较慢\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"搜索引擎比较\\\"\\n x-axis \\\"低准确性\\\" --> \\\"高准确性\\\"\\n y-axis \\\"低速度\\\" --> \\\"高速度\\\"\\n quadrant-1 \\\"需要改进\\\"\\n quadrant-2 \\\"值得推广\\\"\\n quadrant-3 \\\"重新评估\\\"\\n quadrant-4 \\\"需要扩展\\\"\\n \\\"搜索引擎A\\\": [0.8, 0.3]\\n \\\"搜索引擎B\\\": [0.5, 0.7]\\n \\\"搜索引擎C\\\": [0.9, 0.2]\\n \\\"我们的搜索引擎\\\": [0.7, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"实现LLM技术进行文本摘要\"\n ],\n [\n \"P0\",\n \"构建搜索算法，提高搜索效率\"\n ],\n [\n \"P1\",\n \"设计智能推荐系统\"\n ]\n ],\n \"UI Design draft\": \"简洁的搜索框，清晰的搜索结果页面，智能推荐模块\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]", + "\n## context\n\n### Project Name\n\n\n### Original Requirements\nMake a cli snake game\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Project Name\": \"game_2048\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Project Name: # According to the content of \"Original Requirements,\" name the project using snake case style , like 'game_2048' or 'simple_crm.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Make a cli snake game\",\n \"Project Name\": \"cli_snake_game\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility and responsiveness\",\n \"Enhance the game with additional features\"\n ],\n \"User Stories\": [\n \"As a player, I want to control the snake using arrow keys\",\n \"As a player, I want to see my score during the game\",\n \"As a player, I want to have the option to restart the game\",\n \"As a player, I want to see a visually appealing UI\",\n \"As a player, I want to play the game on different platforms\"\n ],\n \"Competitive Analysis\": [\n \"Snake Game A: Simple interface, lacks responsive features\",\n \"SnakeGame.co: Beautiful and responsive UI with high scores displayed\",\n \"SnakeGame.com: Responsive UI with high scores shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of snake games\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Snake Game A\\\": [0.3, 0.6]\\n \\\"SnakeGame.co\\\": [0.45, 0.23]\\n \\\"SnakeGame.com\\\": [0.57, 0.69]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code for controlling the snake and game logic\"\n ],\n [\n \"P1\",\n \"Implementing the scoring system and UI\"\n ],\n [\n \"P2\",\n \"Adding platform compatibility and restart functionality\"\n ]\n ],\n \"UI Design draft\": \"The game will have a simple and intuitive UI with clear controls and a visually appealing design.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]", + "\n## context\n{\"Language\":\"en_us\",\"Programming Language\":\"Python\",\"Original Requirements\":\"Make a cli snake game\",\"Project Name\":\"cli_snake_game\",\"Product Goals\":[\"Create an engaging user experience\",\"Improve accessibility and responsiveness\",\"Enhance the game with additional features\"],\"User Stories\":[\"As a player, I want to control the snake using arrow keys\",\"As a player, I want to see my score during the game\",\"As a player, I want to have the option to restart the game\",\"As a player, I want to see a visually appealing UI\",\"As a player, I want to play the game on different platforms\"],\"Competitive Analysis\":[\"Snake Game A: Simple interface, lacks responsive features\",\"SnakeGame.co: Beautiful and responsive UI with high scores displayed\",\"SnakeGame.com: Responsive UI with high scores shown, but many ads\"],\"Competitive Quadrant Chart\":\"quadrantChart\\n title \\\"Reach and engagement of snake games\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Snake Game A\\\": [0.3, 0.6]\\n \\\"SnakeGame.co\\\": [0.45, 0.23]\\n \\\"SnakeGame.com\\\": [0.57, 0.69]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\"Requirement Analysis\":\"\",\"Requirement Pool\":[[\"P0\",\"The main code for controlling the snake and game logic\"],[\"P1\",\"Implementing the scoring system and UI\"],[\"P2\",\"Adding platform compatibility and restart functionality\"]],\"UI Design draft\":\"The game will have a simple and intuitive UI with clear controls and a visually appealing design.\",\"Anything UNCLEAR\":\"\"}\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Implementation approach\": \"We will ...\",\n \"File list\": [\n \"main.py\",\n \"game.py\"\n ],\n \"Data structures and interfaces\": \"\\nclassDiagram\\n class Main {\\n -SearchEngine search_engine\\n +main() str\\n }\\n class SearchEngine {\\n -Index index\\n -Ranking ranking\\n -Summary summary\\n +search(query: str) str\\n }\\n class Index {\\n -KnowledgeBase knowledge_base\\n +create_index(data: dict)\\n +query_index(query: str) list\\n }\\n class Ranking {\\n +rank_results(results: list) list\\n }\\n class Summary {\\n +summarize_results(results: list) str\\n }\\n class KnowledgeBase {\\n +update(data: dict)\\n +fetch_data(query: str) dict\\n }\\n Main --> SearchEngine\\n SearchEngine --> Index\\n SearchEngine --> Ranking\\n SearchEngine --> Summary\\n Index --> KnowledgeBase\\n\",\n \"Program call flow\": \"\\nsequenceDiagram\\n participant M as Main\\n participant SE as SearchEngine\\n participant I as Index\\n participant R as Ranking\\n participant S as Summary\\n participant KB as KnowledgeBase\\n M->>SE: search(query)\\n SE->>I: query_index(query)\\n I->>KB: fetch_data(query)\\n KB-->>I: return data\\n I-->>SE: return results\\n SE->>R: rank_results(results)\\n R-->>SE: return ranked_results\\n SE->>S: summarize_results(ranked_results)\\n S-->>SE: return summary\\n SE-->>M: return summary\\n\",\n \"Anything UNCLEAR\": \"Clarification needed on third-party API integration, ...\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Implementation approach: # Analyze the difficult points of the requirements, select the appropriate open-source framework\n- File list: typing.List[str] # Only need relative paths. ALWAYS write a main.py or app.py here\n- Data structures and interfaces: # Use mermaid classDiagram code syntax, including classes, method(__init__ etc.) and functions with type annotations, CLEARLY MARK the RELATIONSHIPS between classes, and comply with PEP8 standards. The data structures SHOULD BE VERY DETAILED and the API should be comprehensive with a complete design.\n- Program call flow: # Use sequenceDiagram code syntax, COMPLETE and VERY DETAILED, using CLASSES AND API DEFINED ABOVE accurately, covering the CRUD AND INIT of each object, SYNTAX MUST BE CORRECT.\n- Anything UNCLEAR: # Mention unclear project aspects, then try to clarify it.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Implementation approach\": \"We will use the Pygame library to create the game and handle user input. The game logic will be implemented using Python classes and functions.\",\n \"File list\": [\n \"main.py\",\n \"game.py\"\n ],\n \"Data structures and interfaces\": \"\\nclassDiagram\\n class Game {\\n -snake: list\\n -food: tuple\\n -score: int\\n +__init__()\\n +update()\\n +draw()\\n +handle_input()\\n +restart()\\n }\\n class Snake {\\n -body: list\\n -direction: str\\n +__init__(start_pos: tuple, length: int, direction: str)\\n +move()\\n +change_direction(new_direction: str)\\n +grow()\\n }\\n Game --> Snake\\n\",\n \"Program call flow\": \"\\nsequenceDiagram\\n participant M as Main\\n participant G as Game\\n M->>G: initialize game\\n G->>G: update game state\\n G->>G: draw game\\n G->>G: handle user input\\n G->>G: restart game\\n\",\n \"Anything UNCLEAR\": \"Clarification needed on the scoring system and how the game over condition will be handled.\"\n}\n[/CONTENT]", + "\n## context\n{\"Implementation approach\":\"We will use the Pygame library to create the game and handle user input. The game logic will be implemented using Python classes and functions.\",\"File list\":[\"main.py\",\"game.py\"],\"Data structures and interfaces\":\"\\nclassDiagram\\n class Game {\\n -snake: list\\n -food: tuple\\n -score: int\\n +__init__()\\n +update()\\n +draw()\\n +handle_input()\\n +restart()\\n }\\n class Snake {\\n -body: list\\n -direction: str\\n +__init__(start_pos: tuple, length: int, direction: str)\\n +move()\\n +change_direction(new_direction: str)\\n +grow()\\n }\\n Game --> Snake\\n\",\"Program call flow\":\"\\nsequenceDiagram\\n participant M as Main\\n participant G as Game\\n M->>G: initialize game\\n G->>G: update game state\\n G->>G: draw game\\n G->>G: handle user input\\n G->>G: restart game\\n\",\"Anything UNCLEAR\":\"Clarification needed on the scoring system and how the game over condition will be handled.\"}\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Required Python packages\": [\n \"flask==1.1.2\",\n \"bcrypt==3.2.0\"\n ],\n \"Required Other language third-party packages\": [\n \"No third-party dependencies required\"\n ],\n \"Logic Analysis\": [\n [\n \"game.py\",\n \"Contains Game class and ... functions\"\n ],\n [\n \"main.py\",\n \"Contains main function, from game import Game\"\n ]\n ],\n \"Task list\": [\n \"game.py\",\n \"main.py\"\n ],\n \"Full API spec\": \"openapi: 3.0.0 ...\",\n \"Shared Knowledge\": \"'game.py' contains functions shared across the project.\",\n \"Anything UNCLEAR\": \"Clarification needed on how to start and initialize third-party libraries.\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Required Python packages: typing.List[str] # Provide required Python packages in requirements.txt format.\n- Required Other language third-party packages: typing.List[str] # List down the required packages for languages other than Python.\n- Logic Analysis: typing.List[typing.List[str]] # Provide a list of files with the classes/methods/functions to be implemented, including dependency analysis and imports.\n- Task list: typing.List[str] # Break down the tasks into a list of filenames, prioritized by dependency order.\n- Full API spec: # Describe all APIs using OpenAPI 3.0 spec that may be used by both frontend and backend. If front-end and back-end communication is not required, leave it blank.\n- Shared Knowledge: # Detail any shared knowledge, like common utility functions or configuration variables.\n- Anything UNCLEAR: # Mention any unclear aspects in the project management context and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Required Python packages\": [\n \"pygame==2.0.1\"\n ],\n \"Required Other language third-party packages\": [\n \"No third-party dependencies required\"\n ],\n \"Logic Analysis\": [\n [\n \"game.py\",\n \"Contains Game class and ... functions\"\n ],\n [\n \"main.py\",\n \"Contains main function, from game import Game\"\n ]\n ],\n \"Task list\": [\n \"game.py\",\n \"main.py\"\n ],\n \"Full API spec\": \"\",\n \"Shared Knowledge\": \"'game.py' contains functions shared across the project.\",\n \"Anything UNCLEAR\": \"Clarification needed on the scoring system and how the game over condition will be handled.\"\n}\n[/CONTENT]", + "\nNOTICE\nRole: You are a professional engineer; the main goal is to write google-style, elegant, modular, easy to read and maintain code\nLanguage: Please use the same language as the user requirement, but the title and code should be still in English. For example, if the user speaks Chinese, the specific text of your answer should also be in Chinese.\nATTENTION: Use '##' to SPLIT SECTIONS, not '#'. Output format carefully referenced \"Format example\".\n\n# Context\n## Design\n{\"Implementation approach\":\"We will use the Pygame library to create the game and handle user input. The game logic will be implemented using Python classes and functions.\",\"File list\":[\"main.py\",\"game.py\"],\"Data structures and interfaces\":\"\\nclassDiagram\\n class Game {\\n -snake: list\\n -food: tuple\\n -score: int\\n +__init__()\\n +update()\\n +draw()\\n +handle_input()\\n +restart()\\n }\\n class Snake {\\n -body: list\\n -direction: str\\n +__init__(start_pos: tuple, length: int, direction: str)\\n +move()\\n +change_direction(new_direction: str)\\n +grow()\\n }\\n Game --> Snake\\n\",\"Program call flow\":\"\\nsequenceDiagram\\n participant M as Main\\n participant G as Game\\n M->>G: initialize game\\n G->>G: update game state\\n G->>G: draw game\\n G->>G: handle user input\\n G->>G: restart game\\n\",\"Anything UNCLEAR\":\"Clarification needed on the scoring system and how the game over condition will be handled.\"}\n\n## Tasks\n{\"Required Python packages\":[\"pygame==2.0.1\"],\"Required Other language third-party packages\":[\"No third-party dependencies required\"],\"Logic Analysis\":[[\"game.py\",\"Contains Game class and ... functions\"],[\"main.py\",\"Contains main function, from game import Game\"]],\"Task list\":[\"game.py\",\"main.py\"],\"Full API spec\":\"\",\"Shared Knowledge\":\"'game.py' contains functions shared across the project.\",\"Anything UNCLEAR\":\"Clarification needed on the scoring system and how the game over condition will be handled.\"}\n\n## Legacy Code\n```Code\n\n```\n\n## Debug logs\n```text\n\n\n\n```\n\n## Bug Feedback logs\n```text\n\n```\n\n# Format example\n## Code: game.py\n```python\n## game.py\n...\n```\n\n# Instruction: Based on the context, follow \"Format example\", write code.\n\n## Code: game.py. Write code with triple quoto, based on the following attentions and context.\n1. Only One file: do your best to implement THIS ONLY ONE FILE.\n2. COMPLETE CODE: Your code will be part of the entire project, so please implement complete, reliable, reusable code snippets.\n3. Set default value: If there is any setting, ALWAYS SET A DEFAULT VALUE, ALWAYS USE STRONG TYPE AND EXPLICIT VARIABLE. AVOID circular import.\n4. Follow design: YOU MUST FOLLOW \"Data structures and interfaces\". DONT CHANGE ANY DESIGN. Do not use public member functions that do not exist in your design.\n5. CAREFULLY CHECK THAT YOU DONT MISS ANY NECESSARY CLASS/FUNCTION IN THIS FILE.\n6. Before using a external variable/module, make sure you import it first.\n7. Write out EVERY CODE DETAIL, DON'T LEAVE TODO.\n\n": "```python\n## game.py\n\nclass Game:\n def __init__(self):\n self.snake = []\n self.food = (0, 0)\n self.score = 0\n\n def update(self):\n # Update game state\n pass\n\n def draw(self):\n # Draw game\n pass\n\n def handle_input(self):\n # Handle user input\n pass\n\n def restart(self):\n # Restart game\n pass\n\n\nclass Snake:\n def __init__(self, start_pos: tuple, length: int, direction: str):\n self.body = []\n self.direction = direction\n\n def move(self):\n # Move snake\n pass\n\n def change_direction(self, new_direction: str):\n # Change snake direction\n pass\n\n def grow(self):\n # Grow snake\n pass\n```", + "\n# System\nRole: You are a professional software engineer, and your main task is to review and revise the code. You need to ensure that the code conforms to the google-style standards, is elegantly designed and modularized, easy to read and maintain.\nLanguage: Please use the same language as the user requirement, but the title and code should be still in English. For example, if the user speaks Chinese, the specific text of your answer should also be in Chinese.\nATTENTION: Use '##' to SPLIT SECTIONS, not '#'. Output format carefully referenced \"Format example\".\n\n# Context\n## System Design\n{\"Implementation approach\":\"We will use the Pygame library to create the game and handle user input. The game logic will be implemented using Python classes and functions.\",\"File list\":[\"main.py\",\"game.py\"],\"Data structures and interfaces\":\"\\nclassDiagram\\n class Game {\\n -snake: list\\n -food: tuple\\n -score: int\\n +__init__()\\n +update()\\n +draw()\\n +handle_input()\\n +restart()\\n }\\n class Snake {\\n -body: list\\n -direction: str\\n +__init__(start_pos: tuple, length: int, direction: str)\\n +move()\\n +change_direction(new_direction: str)\\n +grow()\\n }\\n Game --> Snake\\n\",\"Program call flow\":\"\\nsequenceDiagram\\n participant M as Main\\n participant G as Game\\n M->>G: initialize game\\n G->>G: update game state\\n G->>G: draw game\\n G->>G: handle user input\\n G->>G: restart game\\n\",\"Anything UNCLEAR\":\"Clarification needed on the scoring system and how the game over condition will be handled.\"}\n\n## Tasks\n{\"Required Python packages\":[\"pygame==2.0.1\"],\"Required Other language third-party packages\":[\"No third-party dependencies required\"],\"Logic Analysis\":[[\"game.py\",\"Contains Game class and ... functions\"],[\"main.py\",\"Contains main function, from game import Game\"]],\"Task list\":[\"game.py\",\"main.py\"],\"Full API spec\":\"\",\"Shared Knowledge\":\"'game.py' contains functions shared across the project.\",\"Anything UNCLEAR\":\"Clarification needed on the scoring system and how the game over condition will be handled.\"}\n\n## Code Files\n\n\n\n## Code to be Reviewed: game.py\n```Code\n## game.py\n\nclass Game:\n def __init__(self):\n self.snake = []\n self.food = (0, 0)\n self.score = 0\n\n def update(self):\n # Update game state\n pass\n\n def draw(self):\n # Draw game\n pass\n\n def handle_input(self):\n # Handle user input\n pass\n\n def restart(self):\n # Restart game\n pass\n\n\nclass Snake:\n def __init__(self, start_pos: tuple, length: int, direction: str):\n self.body = []\n self.direction = direction\n\n def move(self):\n # Move snake\n pass\n\n def change_direction(self, new_direction: str):\n # Change snake direction\n pass\n\n def grow(self):\n # Grow snake\n pass\n\n```\n\n\n\n# Format example 1\n## Code Review: game.py\n1. No, we should fix the logic of class A due to ...\n2. ...\n3. ...\n4. No, function B is not implemented, ...\n5. ...\n6. ...\n\n## Actions\n1. Fix the `handle_events` method to update the game state only if a move is successful.\n ```python\n def handle_events(self):\n for event in pygame.event.get():\n if event.type == pygame.QUIT:\n return False\n if event.type == pygame.KEYDOWN:\n moved = False\n if event.key == pygame.K_UP:\n moved = self.game.move('UP')\n elif event.key == pygame.K_DOWN:\n moved = self.game.move('DOWN')\n elif event.key == pygame.K_LEFT:\n moved = self.game.move('LEFT')\n elif event.key == pygame.K_RIGHT:\n moved = self.game.move('RIGHT')\n if moved:\n # Update the game state only if a move was successful\n self.render()\n return True\n ```\n2. Implement function B\n\n## Code Review Result\nLBTM\n\n# Format example 2\n## Code Review: game.py\n1. Yes.\n2. Yes.\n3. Yes.\n4. Yes.\n5. Yes.\n6. Yes.\n\n## Actions\npass\n\n## Code Review Result\nLGTM\n\n\n\n# Instruction: Based on the actual code situation, follow one of the \"Format example\". Return only 1 file under review.\n\n## Code Review: Ordered List. Based on the \"Code to be Reviewed\", provide key, clear, concise, and specific answer. If any answer is no, explain how to fix it step by step.\n1. Is the code implemented as per the requirements? If not, how to achieve it? Analyse it step by step.\n2. Is the code logic completely correct? If there are errors, please indicate how to correct them.\n3. Does the existing code follow the \"Data structures and interfaces\"?\n4. Are all functions implemented? If there is no implementation, please indicate how to achieve it step by step.\n5. Have all necessary pre-dependencies been imported? If not, indicate which ones need to be imported\n6. Are methods from other files being reused correctly?\n\n## Actions: Ordered List. Things that should be done after CR, such as implementing class A and function B\n\n## Code Review Result: str. If the code doesn't have bugs, we don't need to rewrite it, so answer LGTM and stop. ONLY ANSWER LGTM/LBTM.\nLGTM/LBTM\n\n": "## Code Review: game.py\n1. No, the code is not implemented as per the requirements. The Game class should have a method to handle game over condition and update the scoring system.\n2. No, the code logic is not completely correct. The Snake class is missing the implementation of the move, change_direction, and grow methods.\n3. Yes, the existing code follows the \"Data structures and interfaces\" provided.\n4. No, the move, change_direction, and grow methods of the Snake class are not implemented. To achieve this, the methods need to be implemented with the appropriate logic for moving the snake, changing its direction, and growing its body.\n5. Yes, all necessary pre-dependencies have been imported.\n6. No, there are no methods from other files being reused in the current code.\n\n## Actions\n1. Implement a method in the Game class to handle the game over condition and update the scoring system.\n2. Implement the move, change_direction, and grow methods in the Snake class.\n3. Reuse methods from other files if necessary.\n\n## Code Review Result\nLBTM", + "\n# System\nRole: You are a professional software engineer, and your main task is to review and revise the code. You need to ensure that the code conforms to the google-style standards, is elegantly designed and modularized, easy to read and maintain.\nLanguage: Please use the same language as the user requirement, but the title and code should be still in English. For example, if the user speaks Chinese, the specific text of your answer should also be in Chinese.\nATTENTION: Use '##' to SPLIT SECTIONS, not '#'. Output format carefully referenced \"Format example\".\n\n# Context\n## System Design\n{\"Implementation approach\":\"We will use the Pygame library to create the game and handle user input. The game logic will be implemented using Python classes and functions.\",\"File list\":[\"main.py\",\"game.py\"],\"Data structures and interfaces\":\"\\nclassDiagram\\n class Game {\\n -snake: list\\n -food: tuple\\n -score: int\\n +__init__()\\n +update()\\n +draw()\\n +handle_input()\\n +restart()\\n }\\n class Snake {\\n -body: list\\n -direction: str\\n +__init__(start_pos: tuple, length: int, direction: str)\\n +move()\\n +change_direction(new_direction: str)\\n +grow()\\n }\\n Game --> Snake\\n\",\"Program call flow\":\"\\nsequenceDiagram\\n participant M as Main\\n participant G as Game\\n M->>G: initialize game\\n G->>G: update game state\\n G->>G: draw game\\n G->>G: handle user input\\n G->>G: restart game\\n\",\"Anything UNCLEAR\":\"Clarification needed on the scoring system and how the game over condition will be handled.\"}\n\n## Tasks\n{\"Required Python packages\":[\"pygame==2.0.1\"],\"Required Other language third-party packages\":[\"No third-party dependencies required\"],\"Logic Analysis\":[[\"game.py\",\"Contains Game class and ... functions\"],[\"main.py\",\"Contains main function, from game import Game\"]],\"Task list\":[\"game.py\",\"main.py\"],\"Full API spec\":\"\",\"Shared Knowledge\":\"'game.py' contains functions shared across the project.\",\"Anything UNCLEAR\":\"Clarification needed on the scoring system and how the game over condition will be handled.\"}\n\n## Code Files\n\n\n\n## Code to be Reviewed: game.py\n```Code\n## game.py\n\nclass Game:\n def __init__(self):\n self.snake = []\n self.food = (0, 0)\n self.score = 0\n\n def update(self):\n # Update game state\n pass\n\n def draw(self):\n # Draw game\n pass\n\n def handle_input(self):\n # Handle user input\n pass\n\n def restart(self):\n # Restart game\n pass\n\n\nclass Snake:\n def __init__(self, start_pos: tuple, length: int, direction: str):\n self.body = []\n self.direction = direction\n\n def move(self):\n # Move snake\n pass\n\n def change_direction(self, new_direction: str):\n # Change snake direction\n pass\n\n def grow(self):\n # Grow snake\n pass\n\n```\n\n## Code Review: game.py\n1. No, the code is not implemented as per the requirements. The Game class should have a method to handle game over condition and update the scoring system.\n2. No, the code logic is not completely correct. The Snake class is missing the implementation of the move, change_direction, and grow methods.\n3. Yes, the existing code follows the \"Data structures and interfaces\" provided.\n4. No, the move, change_direction, and grow methods of the Snake class are not implemented. To achieve this, the methods need to be implemented with the appropriate logic for moving the snake, changing its direction, and growing its body.\n5. Yes, all necessary pre-dependencies have been imported.\n6. No, there are no methods from other files being reused in the current code.\n\n## Actions\n1. Implement a method in the Game class to handle the game over condition and update the scoring system.\n2. Implement the move, change_direction, and grow methods in the Snake class.\n3. Reuse methods from other files if necessary.\n\n## Code Review Result\nLBTM\n\n# Instruction: rewrite code based on the Code Review and Actions\n## Rewrite Code: CodeBlock. If it still has some bugs, rewrite game.py with triple quotes. Do your utmost to optimize THIS SINGLE FILE. Return all completed codes and prohibit the return of unfinished codes.\n```Code\n## game.py\n...\n```\n": "```Code\n## game.py\n\nclass Game:\n def __init__(self):\n self.snake = []\n self.food = (0, 0)\n self.score = 0\n\n def update(self):\n # Update game state\n pass\n\n def draw(self):\n # Draw game\n pass\n\n def handle_input(self):\n # Handle user input\n pass\n\n def restart(self):\n # Restart game\n pass\n\n def handle_game_over(self):\n # Handle game over condition\n pass\n\n def update_score(self):\n # Update scoring system\n pass\n\n\nclass Snake:\n def __init__(self, start_pos: tuple, length: int, direction: str):\n self.body = []\n self.direction = direction\n\n def move(self):\n # Move snake\n pass\n\n def change_direction(self, new_direction: str):\n # Change snake direction\n pass\n\n def grow(self):\n # Grow snake\n pass\n```", + "\n# System\nRole: You are a professional software engineer, and your main task is to review and revise the code. You need to ensure that the code conforms to the google-style standards, is elegantly designed and modularized, easy to read and maintain.\nLanguage: Please use the same language as the user requirement, but the title and code should be still in English. For example, if the user speaks Chinese, the specific text of your answer should also be in Chinese.\nATTENTION: Use '##' to SPLIT SECTIONS, not '#'. Output format carefully referenced \"Format example\".\n\n# Context\n## System Design\n{\"Implementation approach\":\"We will use the Pygame library to create the game and handle user input. The game logic will be implemented using Python classes and functions.\",\"File list\":[\"main.py\",\"game.py\"],\"Data structures and interfaces\":\"\\nclassDiagram\\n class Game {\\n -snake: list\\n -food: tuple\\n -score: int\\n +__init__()\\n +update()\\n +draw()\\n +handle_input()\\n +restart()\\n }\\n class Snake {\\n -body: list\\n -direction: str\\n +__init__(start_pos: tuple, length: int, direction: str)\\n +move()\\n +change_direction(new_direction: str)\\n +grow()\\n }\\n Game --> Snake\\n\",\"Program call flow\":\"\\nsequenceDiagram\\n participant M as Main\\n participant G as Game\\n M->>G: initialize game\\n G->>G: update game state\\n G->>G: draw game\\n G->>G: handle user input\\n G->>G: restart game\\n\",\"Anything UNCLEAR\":\"Clarification needed on the scoring system and how the game over condition will be handled.\"}\n\n## Tasks\n{\"Required Python packages\":[\"pygame==2.0.1\"],\"Required Other language third-party packages\":[\"No third-party dependencies required\"],\"Logic Analysis\":[[\"game.py\",\"Contains Game class and ... functions\"],[\"main.py\",\"Contains main function, from game import Game\"]],\"Task list\":[\"game.py\",\"main.py\"],\"Full API spec\":\"\",\"Shared Knowledge\":\"'game.py' contains functions shared across the project.\",\"Anything UNCLEAR\":\"Clarification needed on the scoring system and how the game over condition will be handled.\"}\n\n## Code Files\n\n\n\n## Code to be Reviewed: game.py\n```Code\n## game.py\n\nclass Game:\n def __init__(self):\n self.snake = []\n self.food = (0, 0)\n self.score = 0\n\n def update(self):\n # Update game state\n pass\n\n def draw(self):\n # Draw game\n pass\n\n def handle_input(self):\n # Handle user input\n pass\n\n def restart(self):\n # Restart game\n pass\n\n def handle_game_over(self):\n # Handle game over condition\n pass\n\n def update_score(self):\n # Update scoring system\n pass\n\n\nclass Snake:\n def __init__(self, start_pos: tuple, length: int, direction: str):\n self.body = []\n self.direction = direction\n\n def move(self):\n # Move snake\n pass\n\n def change_direction(self, new_direction: str):\n # Change snake direction\n pass\n\n def grow(self):\n # Grow snake\n pass\n\n```\n\n\n\n# Format example 1\n## Code Review: game.py\n1. No, we should fix the logic of class A due to ...\n2. ...\n3. ...\n4. No, function B is not implemented, ...\n5. ...\n6. ...\n\n## Actions\n1. Fix the `handle_events` method to update the game state only if a move is successful.\n ```python\n def handle_events(self):\n for event in pygame.event.get():\n if event.type == pygame.QUIT:\n return False\n if event.type == pygame.KEYDOWN:\n moved = False\n if event.key == pygame.K_UP:\n moved = self.game.move('UP')\n elif event.key == pygame.K_DOWN:\n moved = self.game.move('DOWN')\n elif event.key == pygame.K_LEFT:\n moved = self.game.move('LEFT')\n elif event.key == pygame.K_RIGHT:\n moved = self.game.move('RIGHT')\n if moved:\n # Update the game state only if a move was successful\n self.render()\n return True\n ```\n2. Implement function B\n\n## Code Review Result\nLBTM\n\n# Format example 2\n## Code Review: game.py\n1. Yes.\n2. Yes.\n3. Yes.\n4. Yes.\n5. Yes.\n6. Yes.\n\n## Actions\npass\n\n## Code Review Result\nLGTM\n\n\n\n# Instruction: Based on the actual code situation, follow one of the \"Format example\". Return only 1 file under review.\n\n## Code Review: Ordered List. Based on the \"Code to be Reviewed\", provide key, clear, concise, and specific answer. If any answer is no, explain how to fix it step by step.\n1. Is the code implemented as per the requirements? If not, how to achieve it? Analyse it step by step.\n2. Is the code logic completely correct? If there are errors, please indicate how to correct them.\n3. Does the existing code follow the \"Data structures and interfaces\"?\n4. Are all functions implemented? If there is no implementation, please indicate how to achieve it step by step.\n5. Have all necessary pre-dependencies been imported? If not, indicate which ones need to be imported\n6. Are methods from other files being reused correctly?\n\n## Actions: Ordered List. Things that should be done after CR, such as implementing class A and function B\n\n## Code Review Result: str. If the code doesn't have bugs, we don't need to rewrite it, so answer LGTM and stop. ONLY ANSWER LGTM/LBTM.\nLGTM/LBTM\n\n": "## Code Review: game.py\n1. No, the code is not implemented as per the requirements. The `handle_game_over` and `update_score` methods are missing, and the `handle_input` method needs to be implemented to handle user input.\n2. The code logic is not completely correct. The `Snake` class `__init__` method does not initialize the `body` attribute with the `start_pos` and `length` parameters.\n3. The existing code follows the \"Data structures and interfaces\" for the `Game` and `Snake` classes.\n4. The `handle_game_over` and `update_score` methods are not implemented. The `handle_input` method needs to be implemented to handle user input.\n5. The necessary pre-dependencies have been imported.\n6. The methods from other files are not being reused in the existing code.\n\n## Actions\n1. Implement the `handle_game_over` method to handle the game over condition.\n2. Implement the `update_score` method to update the scoring system.\n3. Implement the `handle_input` method to handle user input and update the game state accordingly.\n\n## Code Review Result\nLBTM", + "\n# System\nRole: You are a professional software engineer, and your main task is to review and revise the code. You need to ensure that the code conforms to the google-style standards, is elegantly designed and modularized, easy to read and maintain.\nLanguage: Please use the same language as the user requirement, but the title and code should be still in English. For example, if the user speaks Chinese, the specific text of your answer should also be in Chinese.\nATTENTION: Use '##' to SPLIT SECTIONS, not '#'. Output format carefully referenced \"Format example\".\n\n# Context\n## System Design\n{\"Implementation approach\":\"We will use the Pygame library to create the game and handle user input. The game logic will be implemented using Python classes and functions.\",\"File list\":[\"main.py\",\"game.py\"],\"Data structures and interfaces\":\"\\nclassDiagram\\n class Game {\\n -snake: list\\n -food: tuple\\n -score: int\\n +__init__()\\n +update()\\n +draw()\\n +handle_input()\\n +restart()\\n }\\n class Snake {\\n -body: list\\n -direction: str\\n +__init__(start_pos: tuple, length: int, direction: str)\\n +move()\\n +change_direction(new_direction: str)\\n +grow()\\n }\\n Game --> Snake\\n\",\"Program call flow\":\"\\nsequenceDiagram\\n participant M as Main\\n participant G as Game\\n M->>G: initialize game\\n G->>G: update game state\\n G->>G: draw game\\n G->>G: handle user input\\n G->>G: restart game\\n\",\"Anything UNCLEAR\":\"Clarification needed on the scoring system and how the game over condition will be handled.\"}\n\n## Tasks\n{\"Required Python packages\":[\"pygame==2.0.1\"],\"Required Other language third-party packages\":[\"No third-party dependencies required\"],\"Logic Analysis\":[[\"game.py\",\"Contains Game class and ... functions\"],[\"main.py\",\"Contains main function, from game import Game\"]],\"Task list\":[\"game.py\",\"main.py\"],\"Full API spec\":\"\",\"Shared Knowledge\":\"'game.py' contains functions shared across the project.\",\"Anything UNCLEAR\":\"Clarification needed on the scoring system and how the game over condition will be handled.\"}\n\n## Code Files\n\n\n\n## Code to be Reviewed: game.py\n```Code\n## game.py\n\nclass Game:\n def __init__(self):\n self.snake = []\n self.food = (0, 0)\n self.score = 0\n\n def update(self):\n # Update game state\n pass\n\n def draw(self):\n # Draw game\n pass\n\n def handle_input(self):\n # Handle user input\n pass\n\n def restart(self):\n # Restart game\n pass\n\n def handle_game_over(self):\n # Handle game over condition\n pass\n\n def update_score(self):\n # Update scoring system\n pass\n\n\nclass Snake:\n def __init__(self, start_pos: tuple, length: int, direction: str):\n self.body = []\n self.direction = direction\n\n def move(self):\n # Move snake\n pass\n\n def change_direction(self, new_direction: str):\n # Change snake direction\n pass\n\n def grow(self):\n # Grow snake\n pass\n\n```\n\n## Code Review: game.py\n1. No, the code is not implemented as per the requirements. The `handle_game_over` and `update_score` methods are missing, and the `handle_input` method needs to be implemented to handle user input.\n2. The code logic is not completely correct. The `Snake` class `__init__` method does not initialize the `body` attribute with the `start_pos` and `length` parameters.\n3. The existing code follows the \"Data structures and interfaces\" for the `Game` and `Snake` classes.\n4. The `handle_game_over` and `update_score` methods are not implemented. The `handle_input` method needs to be implemented to handle user input.\n5. The necessary pre-dependencies have been imported.\n6. The methods from other files are not being reused in the existing code.\n\n## Actions\n1. Implement the `handle_game_over` method to handle the game over condition.\n2. Implement the `update_score` method to update the scoring system.\n3. Implement the `handle_input` method to handle user input and update the game state accordingly.\n\n## Code Review Result\nLBTM\n\n# Instruction: rewrite code based on the Code Review and Actions\n## Rewrite Code: CodeBlock. If it still has some bugs, rewrite game.py with triple quotes. Do your utmost to optimize THIS SINGLE FILE. Return all completed codes and prohibit the return of unfinished codes.\n```Code\n## game.py\n...\n```\n": "```Code\n## game.py\n\nclass Game:\n def __init__(self):\n self.snake = []\n self.food = (0, 0)\n self.score = 0\n\n def update(self):\n # Update game state\n pass\n\n def draw(self):\n # Draw game\n pass\n\n def handle_input(self):\n # Handle user input\n pass\n\n def restart(self):\n # Restart game\n pass\n\n def handle_game_over(self):\n # Handle game over condition\n pass\n\n def update_score(self):\n # Update scoring system\n pass\n\n\nclass Snake:\n def __init__(self, start_pos: tuple, length: int, direction: str):\n self.body = [start_pos] # Initialize body with start_pos\n self.direction = direction\n\n def move(self):\n # Move snake\n pass\n\n def change_direction(self, new_direction: str):\n # Change snake direction\n pass\n\n def grow(self):\n # Grow snake\n pass\n```", + "\nNOTICE\nRole: You are a professional engineer; the main goal is to write google-style, elegant, modular, easy to read and maintain code\nLanguage: Please use the same language as the user requirement, but the title and code should be still in English. For example, if the user speaks Chinese, the specific text of your answer should also be in Chinese.\nATTENTION: Use '##' to SPLIT SECTIONS, not '#'. Output format carefully referenced \"Format example\".\n\n# Context\n## Design\n{\"Implementation approach\":\"We will use the Pygame library to create the game and handle user input. The game logic will be implemented using Python classes and functions.\",\"File list\":[\"main.py\",\"game.py\"],\"Data structures and interfaces\":\"\\nclassDiagram\\n class Game {\\n -snake: list\\n -food: tuple\\n -score: int\\n +__init__()\\n +update()\\n +draw()\\n +handle_input()\\n +restart()\\n }\\n class Snake {\\n -body: list\\n -direction: str\\n +__init__(start_pos: tuple, length: int, direction: str)\\n +move()\\n +change_direction(new_direction: str)\\n +grow()\\n }\\n Game --> Snake\\n\",\"Program call flow\":\"\\nsequenceDiagram\\n participant M as Main\\n participant G as Game\\n M->>G: initialize game\\n G->>G: update game state\\n G->>G: draw game\\n G->>G: handle user input\\n G->>G: restart game\\n\",\"Anything UNCLEAR\":\"Clarification needed on the scoring system and how the game over condition will be handled.\"}\n\n## Tasks\n{\"Required Python packages\":[\"pygame==2.0.1\"],\"Required Other language third-party packages\":[\"No third-party dependencies required\"],\"Logic Analysis\":[[\"game.py\",\"Contains Game class and ... functions\"],[\"main.py\",\"Contains main function, from game import Game\"]],\"Task list\":[\"game.py\",\"main.py\"],\"Full API spec\":\"\",\"Shared Knowledge\":\"'game.py' contains functions shared across the project.\",\"Anything UNCLEAR\":\"Clarification needed on the scoring system and how the game over condition will be handled.\"}\n\n## Legacy Code\n```Code\n----- game.py\n## game.py\n\nclass Game:\n def __init__(self):\n self.snake = []\n self.food = (0, 0)\n self.score = 0\n\n def update(self):\n # Update game state\n pass\n\n def draw(self):\n # Draw game\n pass\n\n def handle_input(self):\n # Handle user input\n pass\n\n def restart(self):\n # Restart game\n pass\n\n def handle_game_over(self):\n # Handle game over condition\n pass\n\n def update_score(self):\n # Update scoring system\n pass\n\n\nclass Snake:\n def __init__(self, start_pos: tuple, length: int, direction: str):\n self.body = [start_pos] # Initialize body with start_pos\n self.direction = direction\n\n def move(self):\n # Move snake\n pass\n\n def change_direction(self, new_direction: str):\n # Change snake direction\n pass\n\n def grow(self):\n # Grow snake\n pass\n\n```\n\n## Debug logs\n```text\n\n\n\n```\n\n## Bug Feedback logs\n```text\n\n```\n\n# Format example\n## Code: main.py\n```python\n## main.py\n...\n```\n\n# Instruction: Based on the context, follow \"Format example\", write code.\n\n## Code: main.py. Write code with triple quoto, based on the following attentions and context.\n1. Only One file: do your best to implement THIS ONLY ONE FILE.\n2. COMPLETE CODE: Your code will be part of the entire project, so please implement complete, reliable, reusable code snippets.\n3. Set default value: If there is any setting, ALWAYS SET A DEFAULT VALUE, ALWAYS USE STRONG TYPE AND EXPLICIT VARIABLE. AVOID circular import.\n4. Follow design: YOU MUST FOLLOW \"Data structures and interfaces\". DONT CHANGE ANY DESIGN. Do not use public member functions that do not exist in your design.\n5. CAREFULLY CHECK THAT YOU DONT MISS ANY NECESSARY CLASS/FUNCTION IN THIS FILE.\n6. Before using a external variable/module, make sure you import it first.\n7. Write out EVERY CODE DETAIL, DON'T LEAVE TODO.\n\n": "```python\n## main.py\n\nfrom game import Game\n\ndef main():\n game = Game()\n running = True\n\n while running:\n game.handle_input()\n game.update()\n game.draw()\n\n if game.handle_game_over():\n game.restart()\n\nif __name__ == \"__main__\":\n main()\n```", + "\n# System\nRole: You are a professional software engineer, and your main task is to review and revise the code. You need to ensure that the code conforms to the google-style standards, is elegantly designed and modularized, easy to read and maintain.\nLanguage: Please use the same language as the user requirement, but the title and code should be still in English. For example, if the user speaks Chinese, the specific text of your answer should also be in Chinese.\nATTENTION: Use '##' to SPLIT SECTIONS, not '#'. Output format carefully referenced \"Format example\".\n\n# Context\n## System Design\n{\"Implementation approach\":\"We will use the Pygame library to create the game and handle user input. The game logic will be implemented using Python classes and functions.\",\"File list\":[\"main.py\",\"game.py\"],\"Data structures and interfaces\":\"\\nclassDiagram\\n class Game {\\n -snake: list\\n -food: tuple\\n -score: int\\n +__init__()\\n +update()\\n +draw()\\n +handle_input()\\n +restart()\\n }\\n class Snake {\\n -body: list\\n -direction: str\\n +__init__(start_pos: tuple, length: int, direction: str)\\n +move()\\n +change_direction(new_direction: str)\\n +grow()\\n }\\n Game --> Snake\\n\",\"Program call flow\":\"\\nsequenceDiagram\\n participant M as Main\\n participant G as Game\\n M->>G: initialize game\\n G->>G: update game state\\n G->>G: draw game\\n G->>G: handle user input\\n G->>G: restart game\\n\",\"Anything UNCLEAR\":\"Clarification needed on the scoring system and how the game over condition will be handled.\"}\n\n## Tasks\n{\"Required Python packages\":[\"pygame==2.0.1\"],\"Required Other language third-party packages\":[\"No third-party dependencies required\"],\"Logic Analysis\":[[\"game.py\",\"Contains Game class and ... functions\"],[\"main.py\",\"Contains main function, from game import Game\"]],\"Task list\":[\"game.py\",\"main.py\"],\"Full API spec\":\"\",\"Shared Knowledge\":\"'game.py' contains functions shared across the project.\",\"Anything UNCLEAR\":\"Clarification needed on the scoring system and how the game over condition will be handled.\"}\n\n## Code Files\n----- game.py\n## game.py\n\nclass Game:\n def __init__(self):\n self.snake = []\n self.food = (0, 0)\n self.score = 0\n\n def update(self):\n # Update game state\n pass\n\n def draw(self):\n # Draw game\n pass\n\n def handle_input(self):\n # Handle user input\n pass\n\n def restart(self):\n # Restart game\n pass\n\n def handle_game_over(self):\n # Handle game over condition\n pass\n\n def update_score(self):\n # Update scoring system\n pass\n\n\nclass Snake:\n def __init__(self, start_pos: tuple, length: int, direction: str):\n self.body = [start_pos] # Initialize body with start_pos\n self.direction = direction\n\n def move(self):\n # Move snake\n pass\n\n def change_direction(self, new_direction: str):\n # Change snake direction\n pass\n\n def grow(self):\n # Grow snake\n pass\n\n\n\n## Code to be Reviewed: main.py\n```Code\n## main.py\n\nfrom game import Game\n\ndef main():\n game = Game()\n running = True\n\n while running:\n game.handle_input()\n game.update()\n game.draw()\n\n if game.handle_game_over():\n game.restart()\n\nif __name__ == \"__main__\":\n main()\n\n```\n\n\n\n# Format example 1\n## Code Review: main.py\n1. No, we should fix the logic of class A due to ...\n2. ...\n3. ...\n4. No, function B is not implemented, ...\n5. ...\n6. ...\n\n## Actions\n1. Fix the `handle_events` method to update the game state only if a move is successful.\n ```python\n def handle_events(self):\n for event in pygame.event.get():\n if event.type == pygame.QUIT:\n return False\n if event.type == pygame.KEYDOWN:\n moved = False\n if event.key == pygame.K_UP:\n moved = self.game.move('UP')\n elif event.key == pygame.K_DOWN:\n moved = self.game.move('DOWN')\n elif event.key == pygame.K_LEFT:\n moved = self.game.move('LEFT')\n elif event.key == pygame.K_RIGHT:\n moved = self.game.move('RIGHT')\n if moved:\n # Update the game state only if a move was successful\n self.render()\n return True\n ```\n2. Implement function B\n\n## Code Review Result\nLBTM\n\n# Format example 2\n## Code Review: main.py\n1. Yes.\n2. Yes.\n3. Yes.\n4. Yes.\n5. Yes.\n6. Yes.\n\n## Actions\npass\n\n## Code Review Result\nLGTM\n\n\n\n# Instruction: Based on the actual code situation, follow one of the \"Format example\". Return only 1 file under review.\n\n## Code Review: Ordered List. Based on the \"Code to be Reviewed\", provide key, clear, concise, and specific answer. If any answer is no, explain how to fix it step by step.\n1. Is the code implemented as per the requirements? If not, how to achieve it? Analyse it step by step.\n2. Is the code logic completely correct? If there are errors, please indicate how to correct them.\n3. Does the existing code follow the \"Data structures and interfaces\"?\n4. Are all functions implemented? If there is no implementation, please indicate how to achieve it step by step.\n5. Have all necessary pre-dependencies been imported? If not, indicate which ones need to be imported\n6. Are methods from other files being reused correctly?\n\n## Actions: Ordered List. Things that should be done after CR, such as implementing class A and function B\n\n## Code Review Result: str. If the code doesn't have bugs, we don't need to rewrite it, so answer LGTM and stop. ONLY ANSWER LGTM/LBTM.\nLGTM/LBTM\n\n": "## Code Review: main.py\n1. The code is implemented as per the requirements, but the game over condition and scoring system need to be clarified and implemented.\n \n## Actions\n1. Clarify and implement the game over condition and scoring system in the `Game` class.\n\n## Code Review Result\nLBTM", + "\n# System\nRole: You are a professional software engineer, and your main task is to review and revise the code. You need to ensure that the code conforms to the google-style standards, is elegantly designed and modularized, easy to read and maintain.\nLanguage: Please use the same language as the user requirement, but the title and code should be still in English. For example, if the user speaks Chinese, the specific text of your answer should also be in Chinese.\nATTENTION: Use '##' to SPLIT SECTIONS, not '#'. Output format carefully referenced \"Format example\".\n\n# Context\n## System Design\n{\"Implementation approach\":\"We will use the Pygame library to create the game and handle user input. The game logic will be implemented using Python classes and functions.\",\"File list\":[\"main.py\",\"game.py\"],\"Data structures and interfaces\":\"\\nclassDiagram\\n class Game {\\n -snake: list\\n -food: tuple\\n -score: int\\n +__init__()\\n +update()\\n +draw()\\n +handle_input()\\n +restart()\\n }\\n class Snake {\\n -body: list\\n -direction: str\\n +__init__(start_pos: tuple, length: int, direction: str)\\n +move()\\n +change_direction(new_direction: str)\\n +grow()\\n }\\n Game --> Snake\\n\",\"Program call flow\":\"\\nsequenceDiagram\\n participant M as Main\\n participant G as Game\\n M->>G: initialize game\\n G->>G: update game state\\n G->>G: draw game\\n G->>G: handle user input\\n G->>G: restart game\\n\",\"Anything UNCLEAR\":\"Clarification needed on the scoring system and how the game over condition will be handled.\"}\n\n## Tasks\n{\"Required Python packages\":[\"pygame==2.0.1\"],\"Required Other language third-party packages\":[\"No third-party dependencies required\"],\"Logic Analysis\":[[\"game.py\",\"Contains Game class and ... functions\"],[\"main.py\",\"Contains main function, from game import Game\"]],\"Task list\":[\"game.py\",\"main.py\"],\"Full API spec\":\"\",\"Shared Knowledge\":\"'game.py' contains functions shared across the project.\",\"Anything UNCLEAR\":\"Clarification needed on the scoring system and how the game over condition will be handled.\"}\n\n## Code Files\n----- game.py\n## game.py\n\nclass Game:\n def __init__(self):\n self.snake = []\n self.food = (0, 0)\n self.score = 0\n\n def update(self):\n # Update game state\n pass\n\n def draw(self):\n # Draw game\n pass\n\n def handle_input(self):\n # Handle user input\n pass\n\n def restart(self):\n # Restart game\n pass\n\n def handle_game_over(self):\n # Handle game over condition\n pass\n\n def update_score(self):\n # Update scoring system\n pass\n\n\nclass Snake:\n def __init__(self, start_pos: tuple, length: int, direction: str):\n self.body = [start_pos] # Initialize body with start_pos\n self.direction = direction\n\n def move(self):\n # Move snake\n pass\n\n def change_direction(self, new_direction: str):\n # Change snake direction\n pass\n\n def grow(self):\n # Grow snake\n pass\n\n\n\n## Code to be Reviewed: main.py\n```Code\n## main.py\n\nfrom game import Game\n\ndef main():\n game = Game()\n running = True\n\n while running:\n game.handle_input()\n game.update()\n game.draw()\n\n if game.handle_game_over():\n game.restart()\n\nif __name__ == \"__main__\":\n main()\n\n```\n\n## Code Review: main.py\n1. The code is implemented as per the requirements, but the game over condition and scoring system need to be clarified and implemented.\n \n## Actions\n1. Clarify and implement the game over condition and scoring system in the `Game` class.\n\n## Code Review Result\nLBTM\n\n# Instruction: rewrite code based on the Code Review and Actions\n## Rewrite Code: CodeBlock. If it still has some bugs, rewrite main.py with triple quotes. Do your utmost to optimize THIS SINGLE FILE. Return all completed codes and prohibit the return of unfinished codes.\n```Code\n## main.py\n...\n```\n": "```python\n## main.py\n\nfrom game import Game\n\ndef main():\n game = Game()\n running = True\n\n while running:\n game.handle_input()\n game.update()\n game.draw()\n\n if game.is_game_over():\n game.restart()\n\nif __name__ == \"__main__\":\n main()\n```", + "\n# System\nRole: You are a professional software engineer, and your main task is to review and revise the code. You need to ensure that the code conforms to the google-style standards, is elegantly designed and modularized, easy to read and maintain.\nLanguage: Please use the same language as the user requirement, but the title and code should be still in English. For example, if the user speaks Chinese, the specific text of your answer should also be in Chinese.\nATTENTION: Use '##' to SPLIT SECTIONS, not '#'. Output format carefully referenced \"Format example\".\n\n# Context\n## System Design\n{\"Implementation approach\":\"We will use the Pygame library to create the game and handle user input. The game logic will be implemented using Python classes and functions.\",\"File list\":[\"main.py\",\"game.py\"],\"Data structures and interfaces\":\"\\nclassDiagram\\n class Game {\\n -snake: list\\n -food: tuple\\n -score: int\\n +__init__()\\n +update()\\n +draw()\\n +handle_input()\\n +restart()\\n }\\n class Snake {\\n -body: list\\n -direction: str\\n +__init__(start_pos: tuple, length: int, direction: str)\\n +move()\\n +change_direction(new_direction: str)\\n +grow()\\n }\\n Game --> Snake\\n\",\"Program call flow\":\"\\nsequenceDiagram\\n participant M as Main\\n participant G as Game\\n M->>G: initialize game\\n G->>G: update game state\\n G->>G: draw game\\n G->>G: handle user input\\n G->>G: restart game\\n\",\"Anything UNCLEAR\":\"Clarification needed on the scoring system and how the game over condition will be handled.\"}\n\n## Tasks\n{\"Required Python packages\":[\"pygame==2.0.1\"],\"Required Other language third-party packages\":[\"No third-party dependencies required\"],\"Logic Analysis\":[[\"game.py\",\"Contains Game class and ... functions\"],[\"main.py\",\"Contains main function, from game import Game\"]],\"Task list\":[\"game.py\",\"main.py\"],\"Full API spec\":\"\",\"Shared Knowledge\":\"'game.py' contains functions shared across the project.\",\"Anything UNCLEAR\":\"Clarification needed on the scoring system and how the game over condition will be handled.\"}\n\n## Code Files\n----- game.py\n## game.py\n\nclass Game:\n def __init__(self):\n self.snake = []\n self.food = (0, 0)\n self.score = 0\n\n def update(self):\n # Update game state\n pass\n\n def draw(self):\n # Draw game\n pass\n\n def handle_input(self):\n # Handle user input\n pass\n\n def restart(self):\n # Restart game\n pass\n\n def handle_game_over(self):\n # Handle game over condition\n pass\n\n def update_score(self):\n # Update scoring system\n pass\n\n\nclass Snake:\n def __init__(self, start_pos: tuple, length: int, direction: str):\n self.body = [start_pos] # Initialize body with start_pos\n self.direction = direction\n\n def move(self):\n # Move snake\n pass\n\n def change_direction(self, new_direction: str):\n # Change snake direction\n pass\n\n def grow(self):\n # Grow snake\n pass\n\n\n\n## Code to be Reviewed: main.py\n```Code\n## main.py\n\nfrom game import Game\n\ndef main():\n game = Game()\n running = True\n\n while running:\n game.handle_input()\n game.update()\n game.draw()\n\n if game.is_game_over():\n game.restart()\n\nif __name__ == \"__main__\":\n main()\n\n```\n\n\n\n# Format example 1\n## Code Review: main.py\n1. No, we should fix the logic of class A due to ...\n2. ...\n3. ...\n4. No, function B is not implemented, ...\n5. ...\n6. ...\n\n## Actions\n1. Fix the `handle_events` method to update the game state only if a move is successful.\n ```python\n def handle_events(self):\n for event in pygame.event.get():\n if event.type == pygame.QUIT:\n return False\n if event.type == pygame.KEYDOWN:\n moved = False\n if event.key == pygame.K_UP:\n moved = self.game.move('UP')\n elif event.key == pygame.K_DOWN:\n moved = self.game.move('DOWN')\n elif event.key == pygame.K_LEFT:\n moved = self.game.move('LEFT')\n elif event.key == pygame.K_RIGHT:\n moved = self.game.move('RIGHT')\n if moved:\n # Update the game state only if a move was successful\n self.render()\n return True\n ```\n2. Implement function B\n\n## Code Review Result\nLBTM\n\n# Format example 2\n## Code Review: main.py\n1. Yes.\n2. Yes.\n3. Yes.\n4. Yes.\n5. Yes.\n6. Yes.\n\n## Actions\npass\n\n## Code Review Result\nLGTM\n\n\n\n# Instruction: Based on the actual code situation, follow one of the \"Format example\". Return only 1 file under review.\n\n## Code Review: Ordered List. Based on the \"Code to be Reviewed\", provide key, clear, concise, and specific answer. If any answer is no, explain how to fix it step by step.\n1. Is the code implemented as per the requirements? If not, how to achieve it? Analyse it step by step.\n2. Is the code logic completely correct? If there are errors, please indicate how to correct them.\n3. Does the existing code follow the \"Data structures and interfaces\"?\n4. Are all functions implemented? If there is no implementation, please indicate how to achieve it step by step.\n5. Have all necessary pre-dependencies been imported? If not, indicate which ones need to be imported\n6. Are methods from other files being reused correctly?\n\n## Actions: Ordered List. Things that should be done after CR, such as implementing class A and function B\n\n## Code Review Result: str. If the code doesn't have bugs, we don't need to rewrite it, so answer LGTM and stop. ONLY ANSWER LGTM/LBTM.\nLGTM/LBTM\n\n": "## Code Review: main.py\n1. The code is implemented as per the requirements, but the `is_game_over` method is missing. To achieve this, we need to implement the `is_game_over` method in the Game class to check for the game over condition.\n2. The code logic is correct, but the missing `is_game_over` method needs to be added to handle the game over condition.\n3. The existing code follows the \"Data structures and interfaces\" as per the provided class diagram.\n4. All functions are implemented, but the `is_game_over` method is missing in the Game class. To achieve this, we need to implement the `is_game_over` method to handle the game over condition.\n5. All necessary pre-dependencies have been imported.\n6. Methods from other files are not being reused in the current code.\n\n## Actions\n1. Implement the `is_game_over` method in the Game class to handle the game over condition.\n\n## Code Review Result\nLBTM", + "\n# System\nRole: You are a professional software engineer, and your main task is to review and revise the code. You need to ensure that the code conforms to the google-style standards, is elegantly designed and modularized, easy to read and maintain.\nLanguage: Please use the same language as the user requirement, but the title and code should be still in English. For example, if the user speaks Chinese, the specific text of your answer should also be in Chinese.\nATTENTION: Use '##' to SPLIT SECTIONS, not '#'. Output format carefully referenced \"Format example\".\n\n# Context\n## System Design\n{\"Implementation approach\":\"We will use the Pygame library to create the game and handle user input. The game logic will be implemented using Python classes and functions.\",\"File list\":[\"main.py\",\"game.py\"],\"Data structures and interfaces\":\"\\nclassDiagram\\n class Game {\\n -snake: list\\n -food: tuple\\n -score: int\\n +__init__()\\n +update()\\n +draw()\\n +handle_input()\\n +restart()\\n }\\n class Snake {\\n -body: list\\n -direction: str\\n +__init__(start_pos: tuple, length: int, direction: str)\\n +move()\\n +change_direction(new_direction: str)\\n +grow()\\n }\\n Game --> Snake\\n\",\"Program call flow\":\"\\nsequenceDiagram\\n participant M as Main\\n participant G as Game\\n M->>G: initialize game\\n G->>G: update game state\\n G->>G: draw game\\n G->>G: handle user input\\n G->>G: restart game\\n\",\"Anything UNCLEAR\":\"Clarification needed on the scoring system and how the game over condition will be handled.\"}\n\n## Tasks\n{\"Required Python packages\":[\"pygame==2.0.1\"],\"Required Other language third-party packages\":[\"No third-party dependencies required\"],\"Logic Analysis\":[[\"game.py\",\"Contains Game class and ... functions\"],[\"main.py\",\"Contains main function, from game import Game\"]],\"Task list\":[\"game.py\",\"main.py\"],\"Full API spec\":\"\",\"Shared Knowledge\":\"'game.py' contains functions shared across the project.\",\"Anything UNCLEAR\":\"Clarification needed on the scoring system and how the game over condition will be handled.\"}\n\n## Code Files\n----- game.py\n## game.py\n\nclass Game:\n def __init__(self):\n self.snake = []\n self.food = (0, 0)\n self.score = 0\n\n def update(self):\n # Update game state\n pass\n\n def draw(self):\n # Draw game\n pass\n\n def handle_input(self):\n # Handle user input\n pass\n\n def restart(self):\n # Restart game\n pass\n\n def handle_game_over(self):\n # Handle game over condition\n pass\n\n def update_score(self):\n # Update scoring system\n pass\n\n\nclass Snake:\n def __init__(self, start_pos: tuple, length: int, direction: str):\n self.body = [start_pos] # Initialize body with start_pos\n self.direction = direction\n\n def move(self):\n # Move snake\n pass\n\n def change_direction(self, new_direction: str):\n # Change snake direction\n pass\n\n def grow(self):\n # Grow snake\n pass\n\n\n\n## Code to be Reviewed: main.py\n```Code\n## main.py\n\nfrom game import Game\n\ndef main():\n game = Game()\n running = True\n\n while running:\n game.handle_input()\n game.update()\n game.draw()\n\n if game.is_game_over():\n game.restart()\n\nif __name__ == \"__main__\":\n main()\n\n```\n\n## Code Review: main.py\n1. The code is implemented as per the requirements, but the `is_game_over` method is missing. To achieve this, we need to implement the `is_game_over` method in the Game class to check for the game over condition.\n2. The code logic is correct, but the missing `is_game_over` method needs to be added to handle the game over condition.\n3. The existing code follows the \"Data structures and interfaces\" as per the provided class diagram.\n4. All functions are implemented, but the `is_game_over` method is missing in the Game class. To achieve this, we need to implement the `is_game_over` method to handle the game over condition.\n5. All necessary pre-dependencies have been imported.\n6. Methods from other files are not being reused in the current code.\n\n## Actions\n1. Implement the `is_game_over` method in the Game class to handle the game over condition.\n\n## Code Review Result\nLBTM\n\n# Instruction: rewrite code based on the Code Review and Actions\n## Rewrite Code: CodeBlock. If it still has some bugs, rewrite main.py with triple quotes. Do your utmost to optimize THIS SINGLE FILE. Return all completed codes and prohibit the return of unfinished codes.\n```Code\n## main.py\n...\n```\n": "```python\n## main.py\n\nfrom game import Game\n\ndef main():\n game = Game()\n running = True\n\n while running:\n game.handle_input()\n game.update()\n game.draw()\n\n if game.is_game_over(): # Check for game over condition\n game.restart()\n\nif __name__ == \"__main__\":\n main()\n```", + "You are a python code to Mermaid Sequence Diagram translator in function detail#SYSTEM_MSG_END#```python\n#!/usr/bin/env python\n# -*- coding: utf-8 -*-\n\"\"\"\n@Author : alexanderwu\n@File : write_review.py\n\"\"\"\nimport asyncio\nfrom typing import List, Literal\n\nfrom metagpt.actions import Action\nfrom metagpt.actions.action_node import ActionNode\n\nREVIEW = ActionNode(\n key=\"Review\",\n expected_type=List[str],\n instruction=\"Act as an experienced reviewer and critically assess the given output. Provide specific and\"\n \" constructive feedback, highlighting areas for improvement and suggesting changes.\",\n example=[\n \"The logic in the function `calculate_total` seems flawed. Shouldn't it consider the discount rate as well?\",\n \"The TODO function is not implemented yet? Should we implement it before commit?\",\n ],\n)\n\nREVIEW_RESULT = ActionNode(\n key=\"ReviewResult\",\n expected_type=Literal[\"LGTM\", \"LBTM\"],\n instruction=\"LGTM/LBTM. If the code is fully implemented, \" \"give a LGTM, otherwise provide a LBTM.\",\n example=\"LBTM\",\n)\n\nNEXT_STEPS = ActionNode(\n key=\"NextSteps\",\n expected_type=str,\n instruction=\"Based on the code review outcome, suggest actionable steps. This can include code changes, \"\n \"refactoring suggestions, or any follow-up tasks.\",\n example=\"\"\"1. Refactor the `process_data` method to improve readability and efficiency.\n2. Cover edge cases in the `validate_user` function.\n3. Implement a the TODO in the `calculate_total` function.\n4. Fix the `handle_events` method to update the game state only if a move is successful.\n ```python\n def handle_events(self):\n for event in pygame.event.get():\n if event.type == pygame.QUIT:\n return False\n if event.type == pygame.KEYDOWN:\n moved = False\n if event.key == pygame.K_UP:\n moved = self.game.move('UP')\n elif event.key == pygame.K_DOWN:\n moved = self.game.move('DOWN')\n elif event.key == pygame.K_LEFT:\n moved = self.game.move('LEFT')\n elif event.key == pygame.K_RIGHT:\n moved = self.game.move('RIGHT')\n if moved:\n # Update the game state only if a move was successful\n self.render()\n return True\n ```\n\"\"\",\n)\n\nWRITE_DRAFT = ActionNode(\n key=\"WriteDraft\",\n expected_type=str,\n instruction=\"Could you write draft code for move function in order to implement it?\",\n example=\"Draft: ...\",\n)\n\n\nWRITE_FUNCTION = ActionNode(\n key=\"WriteFunction\",\n expected_type=str,\n instruction=\"write code for the function not implemented.\",\n example=\"\"\"\n```Code\n...\n```\n\"\"\",\n)\n\n\nREWRITE_CODE = ActionNode(\n key=\"RewriteCode\",\n expected_type=str,\n instruction=\"\"\"rewrite code based on the Review and Actions\"\"\",\n example=\"\"\"\n```python\n## example.py\ndef calculate_total(price, quantity):\n total = price * quantity\n```\n\"\"\",\n)\n\n\nCODE_REVIEW_CONTEXT = \"\"\"\n# System\nRole: You are a professional software engineer, and your main task is to review and revise the code. You need to ensure that the code conforms to the google-style standards, is elegantly designed and modularized, easy to read and maintain.\nLanguage: Please use the same language as the user requirement, but the title and code should be still in English. For example, if the user speaks Chinese, the specific text of your answer should also be in Chinese.\n\n# Context\n## System Design\n{\"Implementation approach\": \"我们将使用HTML、CSS和JavaScript来实现这个单机的响应式2048游戏。为了确保游戏性能流畅和响应式设计，我们会选择使用Vue.js框架，因为它易于上手且适合构建交互式界面。我们还将使用localStorage来记录玩家的最高分。\", \"File list\": [\"index.html\", \"styles.css\", \"main.js\", \"game.js\", \"storage.js\"], \"Data structures and interfaces\": \"classDiagram\\\n class Game {\\\n -board Array\\\n -score Number\\\n -bestScore Number\\\n +constructor()\\\n +startGame()\\\n +move(direction: String)\\\n +getBoard() Array\\\n +getScore() Number\\\n +getBestScore() Number\\\n +setBestScore(score: Number)\\\n }\\\n class Storage {\\\n +getBestScore() Number\\\n +setBestScore(score: Number)\\\n }\\\n class Main {\\\n +init()\\\n +bindEvents()\\\n }\\\n Game --> Storage : uses\\\n Main --> Game : uses\", \"Program call flow\": \"sequenceDiagram\\\n participant M as Main\\\n participant G as Game\\\n participant S as Storage\\\n M->>G: init()\\\n G->>S: getBestScore()\\\n S-->>G: return bestScore\\\n M->>G: bindEvents()\\\n M->>G: startGame()\\\n loop Game Loop\\\n M->>G: move(direction)\\\n G->>S: setBestScore(score)\\\n S-->>G: return\\\n end\", \"Anything UNCLEAR\": \"目前项目要求明确，没有不清楚的地方。\"}\n\n## Tasks\n{\"Required Python packages\": [\"无需Python包\"], \"Required Other language third-party packages\": [\"vue.js\"], \"Logic Analysis\": [[\"index.html\", \"作为游戏的入口文件和主要的HTML结构\"], [\"styles.css\", \"包含所有的CSS样式，确保游戏界面美观\"], [\"main.js\", \"包含Main类，负责初始化游戏和绑定事件\"], [\"game.js\", \"包含Game类，负责游戏逻辑，如开始游戏、移动方块等\"], [\"storage.js\", \"包含Storage类，用于获取和设置玩家的最高分\"]], \"Task list\": [\"index.html\", \"styles.css\", \"storage.js\", \"game.js\", \"main.js\"], \"Full API spec\": \"\", \"Shared Knowledge\": \"\\'game.js\\' 包含游戏逻辑相关的函数，被 \\'main.js\\' 调用。\", \"Anything UNCLEAR\": \"目前项目要求明确，没有不清楚的地方。\"}\n\n## Code Files\n----- index.html\n\n\n\n \n \n 2048游戏\n \n \n\n\n
\n

2048

\n
\n
\n
分数
\n
{{ score }}
\n
\n
\n
最高分
\n
{{ bestScore }}
\n
\n
\n
\n
\n
\n {{ cell !== 0 ? cell : \\'\\' }}\n
\n
\n
\n \n
\n\n \n \n \n \n\n\n\n----- styles.css\n/* styles.css */\nbody, html {\n margin: 0;\n padding: 0;\n font-family: \\'Arial\\', sans-serif;\n}\n\n#app {\n text-align: center;\n font-size: 18px;\n color: #776e65;\n}\n\nh1 {\n color: #776e65;\n font-size: 72px;\n font-weight: bold;\n margin: 20px 0;\n}\n\n.scores-container {\n display: flex;\n justify-content: center;\n margin-bottom: 20px;\n}\n\n.score-container, .best-container {\n background: #bbada0;\n padding: 10px;\n border-radius: 5px;\n margin: 0 10px;\n min-width: 100px;\n text-align: center;\n}\n\n.score-header, .best-header {\n color: #eee4da;\n font-size: 18px;\n margin-bottom: 5px;\n}\n\n.game-container {\n max-width: 500px;\n margin: 0 auto 20px;\n background: #bbada0;\n padding: 15px;\n border-radius: 10px;\n position: relative;\n}\n\n.grid-row {\n display: flex;\n}\n\n.grid-cell {\n background: #cdc1b4;\n width: 100px;\n height: 100px;\n margin: 5px;\n display: flex;\n justify-content: center;\n align-items: center;\n font-size: 35px;\n font-weight: bold;\n color: #776e65;\n border-radius: 3px;\n}\n\n/* Dynamic classes for different number cells */\n.number-cell-2 {\n background: #eee4da;\n}\n\n.number-cell-4 {\n background: #ede0c8;\n}\n\n.number-cell-8 {\n background: #f2b179;\n color: #f9f6f2;\n}\n\n.number-cell-16 {\n background: #f59563;\n color: #f9f6f2;\n}\n\n.number-cell-32 {\n background: #f67c5f;\n color: #f9f6f2;\n}\n\n.number-cell-64 {\n background: #f65e3b;\n color: #f9f6f2;\n}\n\n.number-cell-128 {\n background: #edcf72;\n color: #f9f6f2;\n}\n\n.number-cell-256 {\n background: #edcc61;\n color: #f9f6f2;\n}\n\n.number-cell-512 {\n background: #edc850;\n color: #f9f6f2;\n}\n\n.number-cell-1024 {\n background: #edc53f;\n color: #f9f6f2;\n}\n\n.number-cell-2048 {\n background: #edc22e;\n color: #f9f6f2;\n}\n\n/* Larger numbers need smaller font sizes */\n.number-cell-1024, .number-cell-2048 {\n font-size: 30px;\n}\n\nbutton {\n background-color: #8f7a66;\n color: #f9f6f2;\n border: none;\n border-radius: 3px;\n padding: 10px 20px;\n font-size: 18px;\n cursor: pointer;\n outline: none;\n}\n\nbutton:hover {\n background-color: #9f8b76;\n}\n\n----- storage.js\n## storage.js\nclass Storage {\n // 获取最高分\n getBestScore() {\n // 尝试从localStorage中获取最高分，如果不存在则默认为0\n const bestScore = localStorage.getItem(\\'bestScore\\');\n return bestScore ? Number(bestScore) : 0;\n }\n\n // 设置最高分\n setBestScore(score) {\n // 将最高分设置到localStorage中\n localStorage.setItem(\\'bestScore\\', score.toString());\n }\n}\n\n\n\n## Code to be Reviewed: game.js\n```Code\n## game.js\nclass Game {\n constructor() {\n this.board = this.createEmptyBoard();\n this.score = 0;\n this.bestScore = 0;\n }\n\n createEmptyBoard() {\n const board = [];\n for (let i = 0; i < 4; i++) {\n board[i] = [0, 0, 0, 0];\n }\n return board;\n }\n\n startGame() {\n this.board = this.createEmptyBoard();\n this.score = 0;\n this.addRandomTile();\n this.addRandomTile();\n }\n\n addRandomTile() {\n let emptyCells = [];\n for (let r = 0; r < 4; r++) {\n for (let c = 0; c < 4; c++) {\n if (this.board[r][c] === 0) {\n emptyCells.push({ r, c });\n }\n }\n }\n if (emptyCells.length > 0) {\n let randomCell = emptyCells[Math.floor(Math.random() * emptyCells.length)];\n this.board[randomCell.r][randomCell.c] = Math.random() < 0.9 ? 2 : 4;\n }\n }\n\n move(direction) {\n // This function will handle the logic for moving tiles\n // in the specified direction and merging them\n // It will also update the score and add a new random tile if the move is successful\n // The actual implementation of this function is complex and would require\n // a significant amount of code to handle all the cases for moving and merging tiles\n // For the purposes of this example, we will not implement the full logic\n // Instead, we will just call addRandomTile to simulate a move\n this.addRandomTile();\n }\n\n getBoard() {\n return this.board;\n }\n\n getScore() {\n return this.score;\n }\n\n getBestScore() {\n return this.bestScore;\n }\n\n setBestScore(score) {\n this.bestScore = score;\n }\n}\n\n```\n\"\"\"\n\n\nCODE_REVIEW_SMALLEST_CONTEXT = \"\"\"\n## Code to be Reviewed: game.js\n```Code\n// game.js\nclass Game {\n constructor() {\n this.board = this.createEmptyBoard();\n this.score = 0;\n this.bestScore = 0;\n }\n\n createEmptyBoard() {\n const board = [];\n for (let i = 0; i < 4; i++) {\n board[i] = [0, 0, 0, 0];\n }\n return board;\n }\n\n startGame() {\n this.board = this.createEmptyBoard();\n this.score = 0;\n this.addRandomTile();\n this.addRandomTile();\n }\n\n addRandomTile() {\n let emptyCells = [];\n for (let r = 0; r < 4; r++) {\n for (let c = 0; c < 4; c++) {\n if (this.board[r][c] === 0) {\n emptyCells.push({ r, c });\n }\n }\n }\n if (emptyCells.length > 0) {\n let randomCell = emptyCells[Math.floor(Math.random() * emptyCells.length)];\n this.board[randomCell.r][randomCell.c] = Math.random() < 0.9 ? 2 : 4;\n }\n }\n\n move(direction) {\n // This function will handle the logic for moving tiles\n // in the specified direction and merging them\n // It will also update the score and add a new random tile if the move is successful\n // The actual implementation of this function is complex and would require\n // a significant amount of code to handle all the cases for moving and merging tiles\n // For the purposes of this example, we will not implement the full logic\n // Instead, we will just call addRandomTile to simulate a move\n this.addRandomTile();\n }\n\n getBoard() {\n return this.board;\n }\n\n getScore() {\n return this.score;\n }\n\n getBestScore() {\n return this.bestScore;\n }\n\n setBestScore(score) {\n this.bestScore = score;\n }\n}\n\n```\n\"\"\"\n\n\nCODE_REVIEW_SAMPLE = \"\"\"\n## Code Review: game.js\n1. The code partially implements the requirements. The `Game` class is missing the full implementation of the `move` method, which is crucial for the game\\'s functionality.\n2. The code logic is not completely correct. The `move` method is not implemented, which means the game cannot process player moves.\n3. The existing code follows the \"Data structures and interfaces\" in terms of class structure but lacks full method implementations.\n4. Not all functions are implemented. The `move` method is incomplete and does not handle the logic for moving and merging tiles.\n5. All necessary pre-dependencies seem to be imported since the code does not indicate the need for additional imports.\n6. The methods from other files (such as `Storage`) are not being used in the provided code snippet, but the class structure suggests that they will be used correctly.\n\n## Actions\n1. Implement the `move` method to handle tile movements and merging. This is a complex task that requires careful consideration of the game\\'s rules and logic. Here is a simplified version of how one might begin to implement the `move` method:\n ```javascript\n move(direction) {\n // Simplified logic for moving tiles up\n if (direction === \\'up\\') {\n for (let col = 0; col < 4; col++) {\n let tiles = this.board.map(row => row[col]).filter(val => val !== 0);\n let merged = [];\n for (let i = 0; i < tiles.length; i++) {\n if (tiles[i] === tiles[i + 1]) {\n tiles[i] *= 2;\n this.score += tiles[i];\n tiles[i + 1] = 0;\n merged.push(i);\n }\n }\n tiles = tiles.filter(val => val !== 0);\n while (tiles.length < 4) {\n tiles.push(0);\n }\n for (let row = 0; row < 4; row++) {\n this.board[row][col] = tiles[row];\n }\n }\n }\n // Additional logic needed for \\'down\\', \\'left\\', \\'right\\'\n // ...\n this.addRandomTile();\n }\n ```\n2. Integrate the `Storage` class methods to handle the best score. This means updating the `startGame` and `setBestScore` methods to use `Storage` for retrieving and setting the best score:\n ```javascript\n startGame() {\n this.board = this.createEmptyBoard();\n this.score = 0;\n this.bestScore = new Storage().getBestScore(); // Retrieve the best score from storage\n this.addRandomTile();\n this.addRandomTile();\n }\n\n setBestScore(score) {\n if (score > this.bestScore) {\n this.bestScore = score;\n new Storage().setBestScore(score); // Set the new best score in storage\n }\n }\n ```\n\n## Code Review Result\nLBTM\n\n```\n\"\"\"\n\n\nWRITE_CODE_NODE = ActionNode.from_children(\"WRITE_REVIEW_NODE\", [REVIEW, REVIEW_RESULT, NEXT_STEPS])\nWRITE_MOVE_NODE = ActionNode.from_children(\"WRITE_MOVE_NODE\", [WRITE_DRAFT, WRITE_FUNCTION])\n\n\nCR_FOR_MOVE_FUNCTION_BY_3 = \"\"\"\nThe move function implementation provided appears to be well-structured and follows a clear logic for moving and merging tiles in the specified direction. However, there are a few potential improvements that could be made to enhance the code:\n\n1. Encapsulation: The logic for moving and merging tiles could be encapsulated into smaller, reusable functions to improve readability and maintainability.\n\n2. Magic Numbers: There are some magic numbers (e.g., 4, 3) used in the loops that could be replaced with named constants for improved readability and easier maintenance.\n\n3. Comments: Adding comments to explain the logic and purpose of each section of the code can improve understanding for future developers who may need to work on or maintain the code.\n\n4. Error Handling: It's important to consider error handling for unexpected input or edge cases to ensure the function behaves as expected in all scenarios.\n\nOverall, the code could benefit from refactoring to improve readability, maintainability, and extensibility. If you would like, I can provide a refactored version of the move function that addresses these considerations.\n\"\"\"\n\n\nclass WriteCodeAN(Action):\n \"\"\"Write a code review for the context.\"\"\"\n\n async def run(self, context):\n self.llm.system_prompt = \"You are an outstanding engineer and can implement any code\"\n return await WRITE_MOVE_NODE.fill(context=context, llm=self.llm, schema=\"json\")\n\n\nasync def main():\n await WriteCodeAN().run(CODE_REVIEW_SMALLEST_CONTEXT)\n\n\nif __name__ == \"__main__\":\n asyncio.run(main())\n\n```\n\n---\nTranslate the code above into Mermaid Sequence Diagram.": "```mermaid\nsequenceDiagram\n participant User\n participant WriteCodeAN\n participant WRITE_MOVE_NODE\n\n User->>WriteCodeAN: Run\n WriteCodeAN->>WRITE_MOVE_NODE: Fill\n WRITE_MOVE_NODE->>WriteCodeAN: Filled\n WriteCodeAN->>User: Return Filled Context\n```", + "You are a function parser.#MSG_SEP#You can convert spoken words into function parameters.#SYSTEM_MSG_END#text_to_image function parameters description:\nparameter `text`: The text used for image conversion.\nparameter `size_type`: size type\n\n---\nExamples:\nIf want you to do `Draw a girl`, return `text_to_image(text=\"Draw a girl\", size_type=\"512x512\")` brief and clear.\nIf want you to do `Draw an apple`, return `text_to_image(text=\"Draw an apple\", size_type=\"512x512\")` brief and clear.\n\n---\n\nRefer to the `text_to_image` function description, and fill in the function parameters according to the example \"I want you to do xx\" in the Examples section.\nNow I want you to do `Draw an apple`, return function parameters in Examples format above, brief and clear.": "`text_to_image(text=\"Draw an apple\", size_type=\"512x512\")`", + "\nNOTICE\nRole: You are a professional engineer; the main goal is to write google-style, elegant, modular, easy to read and maintain code\nLanguage: Please use the same language as the user requirement, but the title and code should be still in English. For example, if the user speaks Chinese, the specific text of your answer should also be in Chinese.\nATTENTION: Use '##' to SPLIT SECTIONS, not '#'. Output format carefully referenced \"Format example\".\n\n# Context\n## Design\n{\"Implementation approach\": \"We will use the Pygame library to create the game interface and handle user input. The game logic will be implemented using Python classes and data structures.\", \"File list\": [\"main.py\", \"game.py\"], \"Data structures and interfaces\": \"classDiagram\\n class Game {\\n -grid: List[List[int]]\\n -score: int\\n -game_over: bool\\n +__init__()\\n +reset_game()\\n +move(direction: str)\\n +is_game_over() bool\\n +get_empty_cells() List[Tuple[int, int]]\\n +add_new_tile()\\n +get_score() int\\n }\\n class UI {\\n -game: Game\\n +__init__(game: Game)\\n +draw_grid()\\n +draw_score()\\n +draw_game_over()\\n +handle_input()\\n }\\n Game --> UI\", \"Program call flow\": \"sequenceDiagram\\n participant M as Main\\n participant G as Game\\n participant U as UI\\n M->>G: reset_game()\\n M->>U: draw_grid()\\n M->>U: draw_score()\\n M->>U: handle_input()\\n U->>G: move(direction)\\n G->>G: add_new_tile()\\n G->>U: draw_grid()\\n G->>U: draw_score()\\n G->>U: draw_game_over()\\n G->>G: is_game_over()\\n G->>G: get_empty_cells()\\n G->>G: get_score()\", \"Anything UNCLEAR\": \"...\"}\n\n## Tasks\n{\"Required Python packages\": [\"pygame==2.0.1\"], \"Required Other language third-party packages\": [\"No third-party dependencies required\"], \"Logic Analysis\": [[\"game.py\", \"Contains Game class and related functions for game logic\"], [\"main.py\", \"Contains main function, initializes the game and UI\"]], \"Task list\": [\"game.py\", \"main.py\"], \"Full API spec\": \"\", \"Shared Knowledge\": \"The game logic will be implemented using Python classes and data structures. The Pygame library will be used to create the game interface and handle user input.\", \"Anything UNCLEAR\": \"...\"}\n\n## Legacy Code\n```Code\n----- main.py\nif __name__ == \"__main__\":\nmain()\n```\n\n## Debug logs\n```text\n\n\n\n```\n\n## Bug Feedback logs\n```text\n\n```\n\n# Format example\n## Code: game.py\n```python\n## game.py\n...\n```\n\n# Instruction: Based on the context, follow \"Format example\", write code.\n\n## Code: game.py. Write code with triple quoto, based on the following attentions and context.\n1. Only One file: do your best to implement THIS ONLY ONE FILE.\n2. COMPLETE CODE: Your code will be part of the entire project, so please implement complete, reliable, reusable code snippets.\n3. Set default value: If there is any setting, ALWAYS SET A DEFAULT VALUE, ALWAYS USE STRONG TYPE AND EXPLICIT VARIABLE. AVOID circular import.\n4. Follow design: YOU MUST FOLLOW \"Data structures and interfaces\". DONT CHANGE ANY DESIGN. Do not use public member functions that do not exist in your design.\n5. CAREFULLY CHECK THAT YOU DONT MISS ANY NECESSARY CLASS/FUNCTION IN THIS FILE.\n6. Before using a external variable/module, make sure you import it first.\n7. Write out EVERY CODE DETAIL, DON'T LEAVE TODO.\n\n": "```python\n## game.py\n\nfrom typing import List, Tuple\nimport random\n\nclass Game:\n def __init__(self):\n self.grid: List[List[int]] = [[0 for _ in range(4)] for _ in range(4)]\n self.score: int = 0\n self.game_over: bool = False\n\n def reset_game(self):\n self.grid = [[0 for _ in range(4)] for _ in range(4)]\n self.score = 0\n self.game_over = False\n self.add_new_tile()\n self.add_new_tile()\n\n def move(self, direction: str):\n # Implement the logic to move the tiles in the specified direction\n pass\n\n def is_game_over(self) -> bool:\n # Implement the logic to check if the game is over\n pass\n\n def get_empty_cells(self) -> List[Tuple[int, int]]:\n # Implement the logic to get the coordinates of empty cells in the grid\n pass\n\n def add_new_tile(self):\n empty_cells = self.get_empty_cells()\n if empty_cells:\n row, col = random.choice(empty_cells)\n self.grid[row][col] = 2 if random.random() < 0.9 else 4\n\n def get_score(self) -> int:\n return self.score\n```", + "\n## context\n\n### Project Name\n\n\n### Original Requirements\n开发一个基于大语言模型与私有知识库的搜索引擎，希望可以基于大语言模型进行搜索总结\n\n### Search Information\n-\n\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Language\": \"en_us\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"Create a 2048 game\",\n \"Project Name\": \"game_2048\",\n \"Product Goals\": [\n \"Create an engaging user experience\",\n \"Improve accessibility, be responsive\",\n \"More beautiful UI\"\n ],\n \"User Stories\": [\n \"As a player, I want to be able to choose difficulty levels\",\n \"As a player, I want to see my score after each game\",\n \"As a player, I want to get restart button when I lose\",\n \"As a player, I want to see beautiful UI that make me feel good\",\n \"As a player, I want to play game via mobile phone\"\n ],\n \"Competitive Analysis\": [\n \"2048 Game A: Simple interface, lacks responsive features\",\n \"play2048.co: Beautiful and responsive UI with my best score shown\",\n \"2048game.com: Responsive UI with my best score shown, but many ads\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"Reach and engagement of campaigns\\\"\\n x-axis \\\"Low Reach\\\" --> \\\"High Reach\\\"\\n y-axis \\\"Low Engagement\\\" --> \\\"High Engagement\\\"\\n quadrant-1 \\\"We should expand\\\"\\n quadrant-2 \\\"Need to promote\\\"\\n quadrant-3 \\\"Re-evaluate\\\"\\n quadrant-4 \\\"May be improved\\\"\\n \\\"Campaign A\\\": [0.3, 0.6]\\n \\\"Campaign B\\\": [0.45, 0.23]\\n \\\"Campaign C\\\": [0.57, 0.69]\\n \\\"Campaign D\\\": [0.78, 0.34]\\n \\\"Campaign E\\\": [0.40, 0.34]\\n \\\"Campaign F\\\": [0.35, 0.78]\\n \\\"Our Target Product\\\": [0.5, 0.6]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"The main code ...\"\n ],\n [\n \"P0\",\n \"The game algorithm ...\"\n ]\n ],\n \"UI Design draft\": \"Basic function description with a simple style and layout.\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Language: # Provide the language used in the project, typically matching the user's requirement language.\n- Programming Language: # Python/JavaScript or other mainstream programming language.\n- Original Requirements: # Place the original user's requirements here.\n- Project Name: # According to the content of \"Original Requirements,\" name the project using snake case style , like 'game_2048' or 'simple_crm.\n- Product Goals: typing.List[str] # Provide up to three clear, orthogonal product goals.\n- User Stories: typing.List[str] # Provide up to 3 to 5 scenario-based user stories.\n- Competitive Analysis: typing.List[str] # Provide 5 to 7 competitive products.\n- Competitive Quadrant Chart: # Use mermaid quadrantChart syntax. Distribute scores evenly between 0 and 1\n- Requirement Analysis: # Provide a detailed analysis of the requirements.\n- Requirement Pool: typing.List[typing.List[str]] # List down the top-5 requirements with their priority (P0, P1, P2).\n- UI Design draft: # Provide a simple description of UI elements, functions, style, and layout.\n- Anything UNCLEAR: # Mention any aspects of the project that are unclear and try to clarify them.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Language\": \"zh_cn\",\n \"Programming Language\": \"Python\",\n \"Original Requirements\": \"开发一个基于大语言模型与私有知识库的搜索引擎，希望可以基于大语言模型进行搜索总结\",\n \"Project Name\": \"language_model_search\",\n \"Product Goals\": [\n \"提供基于大语言模型的高效搜索功能\",\n \"整合私有知识库，实现个性化搜索结果\",\n \"实现搜索结果的自动总结功能\"\n ],\n \"User Stories\": [\n \"作为用户，我希望能够快速找到我需要的信息\",\n \"作为用户，我希望搜索结果能够根据我的偏好进行个性化排序\",\n \"作为用户，我希望搜索结果能够自动总结，方便我快速了解内容\"\n ],\n \"Competitive Analysis\": [\n \"搜索引擎A: 提供基于大语言模型的搜索功能，但个性化程度较低\",\n \"知识库B: 整合私有知识库，但搜索速度较慢\",\n \"语言模型搜索C: 提供搜索结果自动总结功能，但搜索准确度有待提高\"\n ],\n \"Competitive Quadrant Chart\": \"quadrantChart\\n title \\\"搜索引擎竞争分析\\\"\\n x-axis \\\"低搜索准确度\\\" --> \\\"高搜索准确度\\\"\\n y-axis \\\"低个性化程度\\\" --> \\\"高个性化程度\\\"\\n quadrant-1 \\\"需改进\\\"\\n quadrant-2 \\\"需提升\\\"\\n quadrant-3 \\\"重新评估\\\"\\n quadrant-4 \\\"扩展发展\\\"\\n \\\"搜索引擎A\\\": [0.6, 0.3]\\n \\\"知识库B\\\": [0.4, 0.2]\\n \\\"语言模型搜索C\\\": [0.7, 0.5]\\n \\\"我们的目标产品\\\": [0.8, 0.7]\",\n \"Requirement Analysis\": \"\",\n \"Requirement Pool\": [\n [\n \"P0\",\n \"基于大语言模型的高效搜索功能\"\n ],\n [\n \"P1\",\n \"整合私有知识库，实现个性化搜索结果\"\n ],\n [\n \"P2\",\n \"实现搜索结果的自动总结功能\"\n ]\n ],\n \"UI Design draft\": \"搜索页面简洁明了，搜索结果清晰展示，提供个性化排序和自动总结功能。\",\n \"Anything UNCLEAR\": \"\"\n}\n[/CONTENT]", + "\n## context\nNone\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Project Name\": \"game_2048\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Project Name: # According to the content of \"Original Requirements,\" name the project using snake case style with underline, like 'game_2048' or 'simple_crm.\n\n\n## constraint\nLanguage: Please use the same language as Human INPUT.\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow instructions of nodes, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Project Name\": \"game_2048\"\n}\n[/CONTENT]", + "\n## context\nCompare the key's value of nodes_output and the corresponding requirements one by one. If a key's value that does not match the requirement is found, provide the comment content on how to modify it. No output is required for matching keys.\n\n### nodes_output\n{\"Project Name\": {\"value\": \"game snake\", \"requirement\": \"According to the content of \\\"Original Requirements,\\\" name the project using snake case style with underline, like 'game_2048' or 'simple_crm.\"}}\n\n-----\n\n## format example\n[CONTENT]\n{\n \"key1\": \"comment1\",\n \"key2\": \"comment2\",\n \"keyn\": \"commentn\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- key1: # the first key name of mismatch key\n- key2: # the second key name of mismatch key\n- keyn: # the last key name of mismatch key\n\n## constraint\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow format example's json format, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Project Name\": \"Use snake case style with underline, like 'game_2048' or 'simple_crm.'\"\n}\n[/CONTENT]", + "\n## context\nCompare the key's value of nodes_output and the corresponding requirements one by one. If a key's value that does not match the requirement is found, provide the comment content on how to modify it. No output is required for matching keys.\n\n### nodes_output\n{\"Project Name\": {\"value\": \"game_2048\", \"requirement\": \"According to the content of \\\"Original Requirements,\\\" name the project using snake case style with underline, like 'game_2048' or 'simple_crm.\"}}\n\n-----\n\n## format example\n[CONTENT]\n{\n \"key1\": \"comment1\",\n \"key2\": \"comment2\",\n \"keyn\": \"commentn\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- key1: # the first key name of mismatch key\n- key2: # the second key name of mismatch key\n- keyn: # the last key name of mismatch key\n\n## constraint\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow format example's json format, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Project Name\": \"According to the content of \\\"Original Requirements,\\\" name the project using snake case style with underline, like 'game_2048' or 'simple_crm.\"\n}\n[/CONTENT]", + "\n## context\nchange the nodes_output key's value to meet its comment and no need to add extra comment.\n\n### nodes_output\n{\"Project Name\": {\"value\": \"game snake\", \"comment\": \"Use snake case style with underline, like 'game_2048' or 'simple_crm.'\"}}\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Project Name\": \"game_2048\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Project Name: # According to the content of \"Original Requirements,\" name the project using snake case style with underline, like 'game_2048' or 'simple_crm.\n\n\n## constraint\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow format example's json format, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Project Name\": \"game_snake\"\n}\n[/CONTENT]", + "\n## context\nCompare the key's value of nodes_output and the corresponding requirements one by one. If a key's value that does not match the requirement is found, provide the comment content on how to modify it. No output is required for matching keys.\n\n### nodes_output\n{\"Project Name\": {\"value\": \"game_snake\", \"requirement\": \"According to the content of \\\"Original Requirements,\\\" name the project using snake case style with underline, like 'game_2048' or 'simple_crm.\"}}\n\n-----\n\n## format example\n[CONTENT]\n{\n \"key1\": \"comment1\",\n \"key2\": \"comment2\",\n \"keyn\": \"commentn\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- key1: # the first key name of mismatch key\n- key2: # the second key name of mismatch key\n- keyn: # the last key name of mismatch key\n\n## constraint\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow format example's json format, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Project Name\": \"The project name should be in snake case style with underscores, such as 'game_2048' or 'simple_crm'.\"\n}\n[/CONTENT]", + "\n## context\nchange the nodes_output key's value to meet its comment and no need to add extra comment.\n\n### nodes_output\n{\"Project Name\": {\"value\": \"game_snake\", \"comment\": \"The project name should be in snake case style with underscores, such as 'game_2048' or 'simple_crm'.\"}}\n\n-----\n\n## format example\n[CONTENT]\n{\n \"Project Name\": \"game_2048\"\n}\n[/CONTENT]\n\n## nodes: \": # \"\n- Project Name: # According to the content of \"Original Requirements,\" name the project using snake case style with underline, like 'game_2048' or 'simple_crm.\n\n\n## constraint\nFormat: output wrapped inside [CONTENT][/CONTENT] like format example, nothing else.\n\n## action\nFollow format example's json format, generate output and make sure it follows the format example.\n": "[CONTENT]\n{\n \"Project Name\": \"game_snake\"\n}\n[/CONTENT]", + "You are an action classifier#SYSTEM_MSG_END#If the text explicitly want you to Generate a voice file from the input text, text-to-speech, return `[SKILL]: text_to_speech` brief and clear. For instance: [SKILL]: text_to_speech\nIf the text explicitly want you to Create a drawing based on the text., return `[SKILL]: text_to_image` brief and clear. For instance: [SKILL]: text_to_image\nIf the text explicitly want you to Perform Google searches to provide real-time information., return `[SKILL]: web_search` brief and clear. For instance: [SKILL]: web_search\nOtherwise, return `[TALK]: {talk}` brief and clear. For instance: if {talk} is \"xxxx\" return [TALK]: xxxx\n\nNow what specific action is explicitly mentioned in the text: No, I do not have a poison apple. Do you have a poison apple?\n": "[TALK]: No, I do not have a poison apple. Do you have a poison apple?", + "You are an action classifier#SYSTEM_MSG_END#If the text explicitly want you to Generate a voice file from the input text, text-to-speech, return `[SKILL]: text_to_speech` brief and clear. For instance: [SKILL]: text_to_speech\nIf the text explicitly want you to Create a drawing based on the text., return `[SKILL]: text_to_image` brief and clear. For instance: [SKILL]: text_to_image\nIf the text explicitly want you to Perform Google searches to provide real-time information., return `[SKILL]: web_search` brief and clear. For instance: [SKILL]: web_search\nOtherwise, return `[TALK]: {talk}` brief and clear. For instance: if {talk} is \"xxxx\" return [TALK]: xxxx\n\nNow what specific action is explicitly mentioned in the text: Sure, I can draw you an apple. Draw me an apple.\n": "[SKILL]: text_to_image", + "You are a function parser.#MSG_SEP#You can convert spoken words into function parameters.#SYSTEM_MSG_END#text_to_image function parameters description:\nparameter `text`: The text used for image conversion.\nparameter `size_type`: size type\n\n---\nExamples:\nIf want you to do `Draw a girl`, return `text_to_image(text=\"Draw a girl\", size_type=\"512x512\")` brief and clear.\nIf want you to do `Draw an apple`, return `text_to_image(text=\"Draw an apple\", size_type=\"512x512\")` brief and clear.\n\n---\n\nRefer to the `text_to_image` function description, and fill in the function parameters according to the example \"I want you to do xx\" in the Examples section.\nNow I want you to do `Sure, I can draw you an apple. Draw me an apple.`, return function parameters in Examples format above, brief and clear.": "`text_to_image(text=\"Sure, I can draw you an apple. Draw me an apple.\", size_type=\"512x512\")`", + "You are an action classifier#SYSTEM_MSG_END#Otherwise, return `[TALK]: {talk}` brief and clear. For instance: if {talk} is \"xxxx\" return [TALK]: xxxx\n\nNow what specific action is explicitly mentioned in the text: Sure, I can draw you an apple. Draw me an apple.\n": "[DRAW]: draw an apple", + "Here is an Example for 1 input and 8 possible thoughts:\nInput: 2 8 8 14\nPossible next steps:\n2 + 8 = 10 (left: 8 10 14)\n8 / 2 = 4 (left: 4 8 14)\n14 + 2 = 16 (left: 8 8 16)\n2 * 8 = 16 (left: 8 14 16)\n8 - 2 = 6 (left: 6 8 14)\n14 - 8 = 6 (left: 2 6 8)\n14 / 2 = 7 (left: 7 8 8)\n14 - 2 = 12 (left: 8 8 12)\n\nHere is my task for 1 input and 5 possible thoughts:\nInput: 4 5 6 10\nPossible next steps:\n\n\n\n\nEach output should be strictly a list of nodes, in json format, like this:\n```json\n [\n {\n \"node_id\": str = \"unique identifier for a solution, can be an ordinal\",\n \"node_state_instruction\": \"specified sample of solution\",\n },\n ...\n ]\n```\n": "Here is the list of possible next steps for the given input in JSON format:\n\n```json\n[\n {\n \"node_id\": \"1\",\n \"node_state_instruction\": \"4 + 5 = 9 (left: 6 9 10)\"\n },\n {\n \"node_id\": \"2\",\n \"node_state_instruction\": \"5 + 6 = 11 (left: 4 11 10)\"\n },\n {\n \"node_id\": \"3\",\n \"node_state_instruction\": \"4 * 5 = 20 (left: 6 20 10)\"\n },\n {\n \"node_id\": \"4\",\n \"node_state_instruction\": \"6 - 4 = 2 (left: 2 5 10)\"\n },\n {\n \"node_id\": \"5\",\n \"node_state_instruction\": \"10 - 4 = 6 (left: 6 5 6)\"\n }\n]\n```", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n6 9 10\n": "6 * 9 - 10 = 54 - 10 = 44\n(9 - 6) * 10 = 3 * 10 = 30\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n6 20 10\n": "I'm sorry, but it seems that you haven't completed the calculation for the numbers 6, 20, and 10. If you'd like, I can help you with that.", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n4 11 10\n": "4 11 10 cannot reach 24.", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n2 5 10\n": "2 * 5 * 10 = 100\n(2 + 5) * 10 = 70\n2 5 10 are all too big\nimpossible", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n6 5 6\n": "I'm sorry, but it seems that you have not provided a complete expression for the numbers 6, 5, and 6 to reach 24. If you would like to try again, please provide a complete expression using the numbers 6, 5, and 6 to reach the target number of 24.", + "Here is an Example for 1 input and 8 possible thoughts:\nInput: 2 8 8 14\nPossible next steps:\n2 + 8 = 10 (left: 8 10 14)\n8 / 2 = 4 (left: 4 8 14)\n14 + 2 = 16 (left: 8 8 16)\n2 * 8 = 16 (left: 8 14 16)\n8 - 2 = 6 (left: 6 8 14)\n14 - 8 = 6 (left: 2 6 8)\n14 / 2 = 7 (left: 7 8 8)\n14 - 2 = 12 (left: 8 8 12)\n\nHere is my task for 1 input and 5 possible thoughts:\nInput: 4 11 10\nPossible next steps:\n\n\n\n\nEach output should be strictly a list of nodes, in json format, like this:\n```json\n [\n {\n \"node_id\": str = \"unique identifier for a solution, can be an ordinal\",\n \"node_state_instruction\": \"specified sample of solution\",\n },\n ...\n ]\n```\n": "Here are the possible next steps for the given input in the requested JSON format:\n\nInput: 4 11 10\nPossible next steps:\n```json\n[\n {\n \"node_id\": \"1\",\n \"node_state_instruction\": \"4 + 11 = 15 (left: 10 15)\"\n },\n {\n \"node_id\": \"2\",\n \"node_state_instruction\": \"11 - 4 = 7 (left: 7 10)\"\n },\n {\n \"node_id\": \"3\",\n \"node_state_instruction\": \"4 * 11 = 44 (left: 10 44)\"\n },\n {\n \"node_id\": \"4\",\n \"node_state_instruction\": \"11 / 4 = 2.75 (left: 2.75 10)\"\n },\n {\n \"node_id\": \"5\",\n \"node_state_instruction\": \"11 + 4 = 15 (left: 10 15)\"\n }\n]\n```", + "Here is an Example for 1 input and 8 possible thoughts:\nInput: 2 8 8 14\nPossible next steps:\n2 + 8 = 10 (left: 8 10 14)\n8 / 2 = 4 (left: 4 8 14)\n14 + 2 = 16 (left: 8 8 16)\n2 * 8 = 16 (left: 8 14 16)\n8 - 2 = 6 (left: 6 8 14)\n14 - 8 = 6 (left: 2 6 8)\n14 / 2 = 7 (left: 7 8 8)\n14 - 2 = 12 (left: 8 8 12)\n\nHere is my task for 1 input and 5 possible thoughts:\nInput: 6 9 10\nPossible next steps:\n\n\n\n\nEach output should be strictly a list of nodes, in json format, like this:\n```json\n [\n {\n \"node_id\": str = \"unique identifier for a solution, can be an ordinal\",\n \"node_state_instruction\": \"specified sample of solution\",\n },\n ...\n ]\n```\n": "Certainly! Here are the possible next steps for the given input in JSON format:\n\nInput: 6 9 10\nPossible next steps:\n```json\n[\n {\n \"node_id\": \"1\",\n \"node_state_instruction\": \"6 + 9 = 15 (left: 10 15)\"\n },\n {\n \"node_id\": \"2\",\n \"node_state_instruction\": \"6 * 9 = 54 (left: 10 54)\"\n },\n {\n \"node_id\": \"3\",\n \"node_state_instruction\": \"9 - 6 = 3 (left: 3 9 10)\"\n },\n {\n \"node_id\": \"4\",\n \"node_state_instruction\": \"10 - 6 = 4 (left: 4 6 9)\"\n },\n {\n \"node_id\": \"5\",\n \"node_state_instruction\": \"10 / 6 = 1 (left: 1 6 9)\"\n }\n]\n```", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n10 15\n": "I'm sorry, I cannot evaluate the numbers 10 and 15 without an operation. If you provide an operation, I can help you determine if they can reach 24.", + "Here is an Example for 1 input and 8 possible thoughts:\nInput: 2 8 8 14\nPossible next steps:\n2 + 8 = 10 (left: 8 10 14)\n8 / 2 = 4 (left: 4 8 14)\n14 + 2 = 16 (left: 8 8 16)\n2 * 8 = 16 (left: 8 14 16)\n8 - 2 = 6 (left: 6 8 14)\n14 - 8 = 6 (left: 2 6 8)\n14 / 2 = 7 (left: 7 8 8)\n14 - 2 = 12 (left: 8 8 12)\n\nHere is my task for 1 input and 5 possible thoughts:\nInput: 6 20 10\nPossible next steps:\n\n\n\n\nEach output should be strictly a list of nodes, in json format, like this:\n```json\n [\n {\n \"node_id\": str = \"unique identifier for a solution, can be an ordinal\",\n \"node_state_instruction\": \"specified sample of solution\",\n },\n ...\n ]\n```\n": "Certainly! Here is the list of nodes for the given input and possible next steps:\n\n```json\n[\n {\n \"node_id\": \"1\",\n \"node_state_instruction\": \"6 + 20 = 26 (left: 10 26)\"\n },\n {\n \"node_id\": \"2\",\n \"node_state_instruction\": \"20 - 6 = 14 (left: 14 10)\"\n },\n {\n \"node_id\": \"3\",\n \"node_state_instruction\": \"6 * 20 = 120 (left: 10 120)\"\n },\n {\n \"node_id\": \"4\",\n \"node_state_instruction\": \"20 / 6 = 3.33 (left: 3.33 10)\"\n },\n {\n \"node_id\": \"5\",\n \"node_state_instruction\": \"20 + 6 = 26 (left: 10 26)\"\n }\n]\n```", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n1 6 9\n": "1 6 9 cannot reach 24 using addition, subtraction, multiplication, or division. Therefore, it is impossible for these numbers to reach 24.", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n10 54\n": "It seems like you didn't provide the complete set of numbers for the last calculation. Could you please provide the third number so that I can help you evaluate if they can reach 24?", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n7 10\n": "I'm sorry, but it seems like you didn't provide a complete set of numbers for the last question. Could you please provide the third number so that I can evaluate if they can reach 24?", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n10 44\n": "It seems like you didn't provide the second number for the last set. Could you please provide the second number so that I can evaluate if they can reach 24?", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n10 26\n": "It seems that you have not provided the complete set of numbers for the last question. Could you please provide the third number for the set?", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n10 120\n": "It seems that you have not provided the complete set of numbers for the last question. Could you please provide the third number for the evaluation?", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n3 9 10\n": "3 * 9 * 10 = 270\n(10 - 3) * 9 = 63\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n2.75 10\n": "It seems like you have a mix of numbers and operations. Could you please provide the operations for the last set of numbers (2.75 and 10) so that I can help you evaluate if they can reach 24?", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n4 6 9\n": "4 + 6 + 9 = 19\n(9 - 4) * 6 = 5 * 6 = 30\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n14 10\n": "I'm sorry, it seems like you didn't provide the complete set of numbers for the last expression. Could you please provide the third number so that I can evaluate if they can reach 24?", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n3.33 10\n": "It seems like you have provided a list of numbers and evaluated if they can reach 24 through addition, subtraction, multiplication, and division. Here are the evaluations:\n\n10 14 - Sure\n11 12 - Impossible\n4 4 10 - Sure\n4 9 11 - Sure\n5 7 8 - Likely\n5 6 6 - Likely\n10 10 11 - Impossible\n1 3 3 - Impossible\n\nIt looks like you have evaluated the combinations accurately. As for the last set of numbers \"3.33 10,\" it seems incomplete. If you provide the intended operation or additional numbers, I can help you evaluate it.", + "Here is an Example for 1 input and 8 possible thoughts:\nInput: 2 8 8 14\nPossible next steps:\n2 + 8 = 10 (left: 8 10 14)\n8 / 2 = 4 (left: 4 8 14)\n14 + 2 = 16 (left: 8 8 16)\n2 * 8 = 16 (left: 8 14 16)\n8 - 2 = 6 (left: 6 8 14)\n14 - 8 = 6 (left: 2 6 8)\n14 / 2 = 7 (left: 7 8 8)\n14 - 2 = 12 (left: 8 8 12)\n\nHere is my task for 1 input and 5 possible thoughts:\nInput: 10 15\nPossible next steps:\n\n\n\n\nEach output should be strictly a list of nodes, in json format, like this:\n```json\n [\n {\n \"node_id\": str = \"unique identifier for a solution, can be an ordinal\",\n \"node_state_instruction\": \"specified sample of solution\",\n },\n ...\n ]\n```\n": "Here's the JSON format for the possible next steps for the given input:\n\n```json\n[\n {\n \"node_id\": \"1\",\n \"node_state_instruction\": \"10 + 15 = 25 (left: 25)\"\n },\n {\n \"node_id\": \"2\",\n \"node_state_instruction\": \"15 - 10 = 5 (left: 5)\"\n },\n {\n \"node_id\": \"3\",\n \"node_state_instruction\": \"10 * 15 = 150 (left: 150)\"\n },\n {\n \"node_id\": \"4\",\n \"node_state_instruction\": \"15 / 10 = 1.5 (left: 1.5)\"\n },\n {\n \"node_id\": \"5\",\n \"node_state_instruction\": \"15 + 10 = 25 (left: 25)\"\n }\n]\n```", + "Here is an Example for 1 input and 8 possible thoughts:\nInput: 2 8 8 14\nPossible next steps:\n2 + 8 = 10 (left: 8 10 14)\n8 / 2 = 4 (left: 4 8 14)\n14 + 2 = 16 (left: 8 8 16)\n2 * 8 = 16 (left: 8 14 16)\n8 - 2 = 6 (left: 6 8 14)\n14 - 8 = 6 (left: 2 6 8)\n14 / 2 = 7 (left: 7 8 8)\n14 - 2 = 12 (left: 8 8 12)\n\nHere is my task for 1 input and 5 possible thoughts:\nInput: 3 9 10\nPossible next steps:\n\n\n\n\nEach output should be strictly a list of nodes, in json format, like this:\n```json\n [\n {\n \"node_id\": str = \"unique identifier for a solution, can be an ordinal\",\n \"node_state_instruction\": \"specified sample of solution\",\n },\n ...\n ]\n```\n": "Here are the possible next steps for the given input in JSON format:\n\nInput: 3 9 10\nPossible next steps:\n```json\n[\n {\n \"node_id\": \"1\",\n \"node_state_instruction\": \"3 + 9 = 12 (left: 10 12)\"\n },\n {\n \"node_id\": \"2\",\n \"node_state_instruction\": \"9 - 3 = 6 (left: 6 9 10)\"\n },\n {\n \"node_id\": \"3\",\n \"node_state_instruction\": \"10 - 3 = 7 (left: 7 9 10)\"\n },\n {\n \"node_id\": \"4\",\n \"node_state_instruction\": \"3 * 9 = 27 (left: 10 27)\"\n },\n {\n \"node_id\": \"5\",\n \"node_state_instruction\": \"9 / 3 = 3 (left: 3 9 10)\"\n }\n]\n```", + "Here is an Example for 1 input and 8 possible thoughts:\nInput: 2 8 8 14\nPossible next steps:\n2 + 8 = 10 (left: 8 10 14)\n8 / 2 = 4 (left: 4 8 14)\n14 + 2 = 16 (left: 8 8 16)\n2 * 8 = 16 (left: 8 14 16)\n8 - 2 = 6 (left: 6 8 14)\n14 - 8 = 6 (left: 2 6 8)\n14 / 2 = 7 (left: 7 8 8)\n14 - 2 = 12 (left: 8 8 12)\n\nHere is my task for 1 input and 5 possible thoughts:\nInput: 4 6 9\nPossible next steps:\n\n\n\n\nEach output should be strictly a list of nodes, in json format, like this:\n```json\n [\n {\n \"node_id\": str = \"unique identifier for a solution, can be an ordinal\",\n \"node_state_instruction\": \"specified sample of solution\",\n },\n ...\n ]\n```\n": "Certainly! Here are the possible next steps for the given input in the requested JSON format:\n\nInput: 4 6 9\nPossible next steps:\n```json\n[\n {\n \"node_id\": \"1\",\n \"node_state_instruction\": \"4 + 6 = 10 (left: 10 9)\"\n },\n {\n \"node_id\": \"2\",\n \"node_state_instruction\": \"4 * 6 = 24 (left: 24 9)\"\n },\n {\n \"node_id\": \"3\",\n \"node_state_instruction\": \"4 - 6 = -2 (left: -2 9)\"\n },\n {\n \"node_id\": \"4\",\n \"node_state_instruction\": \"6 + 9 = 15 (left: 4 15)\"\n },\n {\n \"node_id\": \"5\",\n \"node_state_instruction\": \"6 * 9 = 54 (left: 4 54)\"\n }\n]\n```\nI hope this helps! If you need further assistance, feel free to ask.", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n10 27\n": "I'm sorry, but it seems like you didn't provide the third number for the last set. Could you please provide the third number so I can evaluate if the given numbers can reach 24?", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n1.5\n": "The evaluation of the given numbers is as follows:\n\n10 14: Sure\n11 12: Impossible\n4 4 10: Sure\n4 9 11: Sure\n5 7 8: Likely\n5 6 6: Likely\n10 10 11: Impossible\n1 3 3: Impossible\n1.5: I'm sorry, I cannot evaluate a single number for reaching 24.", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n4 15\n": "It seems like you didn't provide the complete set of numbers for the last question. Could you please provide the third number so that I can evaluate if they can reach 24?", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n24 9\n": "24 + 9 = 33\n24 - 9 = 15\n24 * 9 = 216\n24 / 9 = 2.67\nIt is impossible to reach 24 with the given numbers.", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n10 9\n": "It seems like you might have missed providing the operation for the last set of numbers \"10 9\". If you could provide the operation, I can help evaluate if the given numbers can reach 24.", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n25\n": "It seems that you have provided a list of numbers and evaluated whether they can reach 24 through addition, subtraction, multiplication, and division. Based on your evaluations, the combinations of numbers that can reach 24 are \"10 14\" and \"4 4 10,\" which are sure to reach 24. The combinations \"11 12\" and \"4 9 11\" are also sure to reach 24. The combination \"5 7 8\" is likely to reach 24, and the combinations \"5 6 6\" and \"10 10 11\" are unlikely to reach 24. The combination \"1 3 3\" is impossible to reach 24. If you have any more numbers to evaluate, feel free to share them with me.", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n4 54\n": "I'm sorry, but I cannot evaluate the expression \"4 54\" without an operator. Could you please provide the operator for the expression?", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n-2 9\n": "I'm sorry, but it seems like you didn't provide the third number for the last expression. Could you please provide the third number so that I can evaluate if the given numbers can reach 24?", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n10 12\n": "It seems like you have evaluated the combinations correctly. Based on the evaluations you provided, the combinations that can reach 24 are:\n\n- 10 14 (sure)\n- 4 4 10 (sure)\n- 4 9 11 (sure)\n- 5 7 8 (likely)\n- 5 6 6 (likely)\n\nThe combinations that cannot reach 24 are:\n\n- 11 12 (impossible)\n- 10 10 11 (impossible)\n- 1 3 3 (impossible)\n\nAs for the combination \"10 12,\" it seems like it was cut off. If you provide the complete expression, I can help evaluate it for you.", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n150\n": "It seems like you have evaluated the given numbers correctly. Here are the results:\n\n10 14: Sure\n11 12: Impossible\n4 4 10: Sure\n4 9 11: Sure\n5 7 8: Likely\n5 6 6: Likely\n10 10 11: Impossible\n1 3 3: Impossible\n\nAs for the number 150, it is not clear what operation or combination of numbers it is associated with. If you have a specific question or context for the number 150, please provide more details so that I can assist you further.", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n5\n": "The numbers 10 and 14 can reach 24, so that is sure. The numbers 11 and 12 cannot reach 24, so that is impossible. The numbers 4, 4, and 10 can reach 24, so that is sure. The numbers 4, 9, and 11 can reach 24, so that is sure. The numbers 5, 7, and 8 cannot reach 24, but they are within a reasonable range, so that is likely. The numbers 5, 6, and 6 cannot reach 24, but they are within a reasonable range, so that is likely. The numbers 10, 10, and 11 cannot reach 24, so that is impossible. The numbers 1, 3, and 3 cannot reach 24, so that is impossible. The number 5 by itself cannot reach 24.", + "Evaluate if given numbers can reach 24 (sure/likely/impossible)\n10 14\n10 + 14 = 24\nsure\n11 12\n11 + 12 = 23\n12 - 11 = 1\n11 * 12 = 132\n11 / 12 = 0.91\nimpossible\n4 4 10\n4 + 4 + 10 = 8 + 10 = 18\n4 * 10 - 4 = 40 - 4 = 36\n(10 - 4) * 4 = 6 * 4 = 24\nsure\n4 9 11\n9 + 11 + 4 = 20 + 4 = 24\nsure\n5 7 8\n5 + 7 + 8 = 12 + 8 = 20\n(8 - 5) * 7 = 3 * 7 = 21\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n5 6 6\n5 + 6 + 6 = 17\n(6 - 5) * 6 = 1 * 6 = 6\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely\n10 10 11\n10 + 10 + 11 = 31\n(11 - 10) * 10 = 10\n10 10 10 are all too big\nimpossible\n1 3 3\n1 * 3 * 3 = 9\n(1 + 3) * 3 = 12\n1 3 3 are all too small\nimpossible\n7 9 10\n": "7 * 9 - 10 = 63 - 10 = 53\n(9 - 7) * 10 = 2 * 10 = 20\nI cannot obtain 24 now, but numbers are within a reasonable range\nlikely", + "### Requirements\n1. Add docstrings to the given code following the google style.\n2. Replace the function body with an Ellipsis object(...) to reduce output.\n3. If the types are already annotated, there is no need to include them in the docstring.\n4. Extract only class, function or the docstrings for the module parts from the given Python code, avoiding any other text.\n\n### Input Example\n```python\ndef function_with_pep484_type_annotations(param1: int) -> bool:\n return isinstance(param1, int)\n\nclass ExampleError(Exception):\n def __init__(self, msg: str):\n self.msg = msg\n```\n\n### Output Example\n```python\ndef function_with_pep484_type_annotations(param1: int) -> bool:\n \"\"\"Example function with PEP 484 type annotations.\n\n Extended description of function.\n\n Args:\n param1: The first parameter.\n\n Returns:\n The return value. True for success, False otherwise.\n \"\"\"\n ...\n\nclass ExampleError(Exception):\n \"\"\"Exceptions are documented in the same way as classes.\n\n The __init__ method was documented in the class level docstring.\n\n Args:\n msg: Human readable string describing the exception.\n\n Attributes:\n msg: Human readable string describing the exception.\n \"\"\"\n ...\n```\n#SYSTEM_MSG_END#```python\nimport pytest\nfrom metagpt.actions.write_docstring import WriteDocstring\ncode = '\\ndef add_numbers(a: int, b: int):\\n return a + b\\n\\n\\nclass Person:\\n def __init__(self, name: str, age: int):\\n self.name = name\\n self.age = age\\n\\n def greet(self):\\n return f\"Hello, my name is {self.name} and I am {self.age} years old.\"\\n'\n\n@pytest.mark.asyncio\n@pytest.mark.parametrize(('style', 'part'), [('google', 'Args:'), ('numpy', 'Parameters'), ('sphinx', ':param name:')], ids=['google', 'numpy', 'sphinx'])\nasync def test_write_docstring(style: str, part: str, context):\n ret = await WriteDocstring(context=context).run(code, style=style)\n assert part in ret\n\n@pytest.mark.asyncio\nasync def test_write():\n code = await WriteDocstring.write_docstring(__file__)\n assert code\n```": "```python\ndef add_numbers(a: int, b: int):\n \"\"\"Adds two numbers together.\n\n Args:\n a: The first number.\n b: The second number.\n\n Returns:\n The sum of a and b.\n \"\"\"\n ...\n\nclass Person:\n \"\"\"Represents a person.\n\n Args:\n name: The name of the person.\n age: The age of the person.\n\n Attributes:\n name: The name of the person.\n age: The age of the person.\n \"\"\"\n\n def __init__(self, name: str, age: int):\n ...\n \n def greet(self):\n \"\"\"Generates a greeting message.\n\n Returns:\n A greeting string.\n \"\"\"\n ...\n```", + "\nWrite a coherent passage of 4 short paragraphs. The end sentence of each paragraph must be: It isn't difficult to do a handstand if you just stand on your hands. It caught him off guard that space smelled of seared steak. When she didn’t like a guy who was trying to pick her up, she started using sign language. Each person who knows you has a different perception of who you are.\n\nMake a plan then write. Your output should be like:\n\nPlan:\n\n\nPassage:\n\n\n\nEach output should be strictly a list of nodes, in json format, like this:\n```json\n [\n {\n \"node_id\": str = \"unique identifier for a solution, can be an ordinal\",\n \"node_state_instruction\": \"specified sample of solution\",\n },\n ...\n ]\n```\n": "Plan:\n```json\n[\n {\n \"node_id\": \"1\",\n \"node_state_instruction\": \"Introduce the concept of doing a handstand.\"\n },\n {\n \"node_id\": \"2\",\n \"node_state_instruction\": \"Describe the surprising smell of space.\"\n },\n {\n \"node_id\": \"3\",\n \"node_state_instruction\": \"Introduce a woman using sign language to deter unwanted attention.\"\n },\n {\n \"node_id\": \"4\",\n \"node_state_instruction\": \"Discuss the varied perceptions people have of an individual.\"\n }\n]\n```\n\nPassage:\n```json\n[\n {\n \"node_id\": \"1\",\n \"node_state_instruction\": \"It isn't difficult to do a handstand if you just stand on your hands.\"\n },\n {\n \"node_id\": \"2\",\n \"node_state_instruction\": \"It caught him off guard that space smelled of seared steak.\"\n },\n {\n \"node_id\": \"3\",\n \"node_state_instruction\": \"When she didn’t like a guy who was trying to pick her up, she started using sign language.\"\n },\n {\n \"node_id\": \"4\",\n \"node_state_instruction\": \"Each person who knows you has a different perception of who you are.\"\n }\n]\n```", + "Given an instruction and several choices, decide which choice is most promising. Analyze each choice in detail, then conclude in the last line \"The best choice is {s}\", where s the integer id of the choice.\nChoice 4:\nDiscuss the varied perceptions people have of an individual.\n": "Choice 4: Discuss the varied perceptions people have of an individual.\n\nAnalyzing this choice, discussing the varied perceptions people have of an individual can be promising as it allows for a deeper understanding of how different people perceive and interpret the actions, behaviors, and characteristics of an individual. This can lead to insights into the complexities of human interactions and relationships, and can provide valuable information for personal growth and self-awareness.\n\nBy exploring the reasons behind differing perceptions, it can also lead to increased empathy and understanding of others, which is beneficial for building stronger relationships and effective communication.\n\nFurthermore, discussing varied perceptions can also shed light on societal influences, biases, and stereotypes that may impact how individuals are perceived, which can be valuable for promoting inclusivity and diversity.\n\nThe best choice is 4.", + "Given an instruction and several choices, decide which choice is most promising. Analyze each choice in detail, then conclude in the last line \"The best choice is {s}\", where s the integer id of the choice.\nChoice 3:\nIntroduce a woman using sign language to deter unwanted attention.\n": "Choice 3: Introduce a woman using sign language to deter unwanted attention.\n\nAnalysis:\nIntroducing a woman using sign language can be a promising choice to deter unwanted attention. Sign language is a visual and expressive form of communication, and it can be an effective way to convey a message without drawing unnecessary attention. By using sign language, the woman can communicate with others discreetly and avoid attracting unwanted attention from individuals who may be seeking to engage her in conversation or interaction.\n\nAdditionally, using sign language can create a sense of privacy and exclusivity, as it may not be easily understood by everyone in the vicinity. This can help the woman maintain a level of personal space and discourage unwelcome advances or interactions.\n\nFurthermore, introducing sign language into the interaction can also serve as a way to promote inclusivity and awareness of different forms of communication, which can contribute to a more understanding and respectful environment.\n\nConclusion:\nThe best choice is 3.", + "Given an instruction and several choices, decide which choice is most promising. Analyze each choice in detail, then conclude in the last line \"The best choice is {s}\", where s the integer id of the choice.\nChoice 1:\nIntroduce the concept of doing a handstand.\n": "Choice 1: Introduce the concept of doing a handstand.\n\nAnalyzing the choice:\nIntroducing the concept of doing a handstand can be a fun and engaging way to promote physical activity and coordination. It can also help build strength and balance. However, it's important to ensure that proper safety measures and instructions are provided to prevent any potential injuries.\n\nConclusion:\nThe best choice is 1.", + "Given an instruction and several choices, decide which choice is most promising. Analyze each choice in detail, then conclude in the last line \"The best choice is {s}\", where s the integer id of the choice.\nChoice 2:\nDescribe the surprising smell of space.\n": "Choice 2: \n\nThe surprising smell of space has been described by astronauts as a mix of gunpowder, seared steak, hot metal, and raspberries. This unexpected combination of scents is due to the volatile organic compounds that are released from the materials in the spacecraft when they are exposed to the vacuum of space. These compounds react with the radiation from the sun, creating the unique smell that astronauts have reported. The smell of space is intriguing and unexpected, providing a sensory experience that is truly out of this world.\n\nThe best choice is 2.", + "\nWrite a coherent passage of 4 short paragraphs. The end sentence of each paragraph must be: Introduce the concept of doing a handstand.\n\nMake a plan then write. Your output should be like:\n\nPlan:\n\n\nPassage:\n\n\n\nEach output should be strictly a list of nodes, in json format, like this:\n```json\n [\n {\n \"node_id\": str = \"unique identifier for a solution, can be an ordinal\",\n \"node_state_instruction\": \"specified sample of solution\",\n },\n ...\n ]\n```\n": "Plan:\n```json\n[\n {\n \"node_id\": \"1\",\n \"node_state_instruction\": \"Decide on a suitable location to practice handstands.\"\n },\n {\n \"node_id\": \"2\",\n \"node_state_instruction\": \"Warm up your body with some stretching exercises.\"\n },\n {\n \"node_id\": \"3\",\n \"node_state_instruction\": \"Practice against a wall to build strength and balance.\"\n },\n {\n \"node_id\": \"4\",\n \"node_state_instruction\": \"Gradually work on holding the handstand position without support.\"\n }\n]\n```\n\nPassage:\n```json\n[\n {\n \"node_id\": \"1\",\n \"node_state_instruction\": \"Decide on a suitable location to practice handstands.\"\n },\n {\n \"node_id\": \"2\",\n \"node_state_instruction\": \"Before attempting a handstand, it's important to decide on a suitable location to practice. A clear, open space with a soft surface, such as a yoga mat or grass, is ideal for safety and comfort.\"\n },\n {\n \"node_id\": \"3\",\n \"node_state_instruction\": \"Warm up your body with some stretching exercises.\"\n },\n {\n \"node_id\": \"4\",\n \"node_state_instruction\": \"To prepare your body for the physical demands of a handstand, start with a series of stretching exercises. Focus on the wrists, shoulders, and core muscles to ensure they are adequately warmed up and flexible.\"\n },\n {\n \"node_id\": \"5\",\n \"node_state_instruction\": \"Practice against a wall to build strength and balance.\"\n },\n {\n \"node_id\": \"6\",\n \"node_state_instruction\": \"When learning to do a handstand, it can be helpful to practice against a wall. This provides support and allows you to focus on building strength and balance in the proper alignment. Gradually work on kicking up into the handstand position and holding it for short periods of time.\"\n },\n {\n \"node_id\": \"7\",\n \"node_state_instruction\": \"Gradually work on holding the handstand position without support.\"\n },\n {\n \"node_id\": \"8\",\n \"node_state_instruction\": \"As you gain confidence and strength, begin to practice holding the handstand position without the support of a wall. Focus on engaging your core and maintaining a straight line from your wrists to your hips. With consistent practice, you'll be able to hold a handstand with control and stability.\"\n }\n]\n```", + "Given an instruction and several choices, decide which choice is most promising. Analyze each choice in detail, then conclude in the last line \"The best choice is {s}\", where s the integer id of the choice.\nChoice 1:\nDecide on a suitable location to practice handstands.\n": "Choice 1: Find a spacious and flat area with soft ground, such as a grassy field or a padded gym floor. This will minimize the risk of injury and provide a stable surface for practicing handstands.\n\nThe best choice is 1", + "Given an instruction and several choices, decide which choice is most promising. Analyze each choice in detail, then conclude in the last line \"The best choice is {s}\", where s the integer id of the choice.\nChoice 2:\nWarm up your body with some stretching exercises.\n": "Stretching exercises are a great way to warm up the body before engaging in physical activity. They help to increase blood flow to the muscles, improve flexibility, and reduce the risk of injury. By performing stretching exercises, you can prepare your body for more strenuous activities and improve your overall performance.\n\nThe best choice is 2.", + "Given an instruction and several choices, decide which choice is most promising. Analyze each choice in detail, then conclude in the last line \"The best choice is {s}\", where s the integer id of the choice.\nChoice 3:\nPractice against a wall to build strength and balance.\n": "Choice 3: Practice against a wall to build strength and balance.\n\nAnalyzing choice 3, practicing against a wall can be a promising option for building strength and balance. The wall provides a stable surface to push against, which can help in developing strength in the muscles. It also offers support, allowing for balance practice without the risk of falling. Additionally, using a wall as a tool for practice can help in improving posture and body alignment.\n\nThe best choice is 3.", + "Given an instruction and several choices, decide which choice is most promising. Analyze each choice in detail, then conclude in the last line \"The best choice is {s}\", where s the integer id of the choice.\nChoice 4:\nGradually work on holding the handstand position without support.\n": "Choice 4: Gradually work on holding the handstand position without support.\n\nAnalyzing this choice, working on holding the handstand position without support is a promising option for improving strength, balance, and body control. This choice allows for gradual progression, which is important for building the necessary strength and stability to maintain the handstand position. It also encourages the development of proprioception and spatial awareness, which are essential for mastering this challenging skill.\n\nThe best choice is 4.", + "\nWrite a coherent passage of 4 short paragraphs. The end sentence of each paragraph must be: Decide on a suitable location to practice handstands.\n\nMake a plan then write. Your output should be like:\n\nPlan:\n\n\nPassage:\n\n\n\nEach output should be strictly a list of nodes, in json format, like this:\n```json\n [\n {\n \"node_id\": str = \"unique identifier for a solution, can be an ordinal\",\n \"node_state_instruction\": \"specified sample of solution\",\n },\n ...\n ]\n```\n": "Plan:\n```json\n[\n {\n \"node_id\": \"1\",\n \"node_state_instruction\": \"Decide on a suitable location to practice handstands.\"\n },\n {\n \"node_id\": \"2\",\n \"node_state_instruction\": \"Consider a flat and stable surface, such as a grassy field or a gym mat.\"\n },\n {\n \"node_id\": \"3\",\n \"node_state_instruction\": \"Ensure there is enough space around you to avoid any obstacles.\"\n },\n {\n \"node_id\": \"4\",\n \"node_state_instruction\": \"Practice near a wall or with a spotter for safety.\"\n }\n]\n```\n\nPassage:\n```json\n[\n {\n \"node_id\": \"1\",\n \"node_state_instruction\": \"Decide on a suitable location to practice handstands.\"\n },\n {\n \"node_id\": \"2\",\n \"node_state_instruction\": \"Consider a flat and stable surface, such as a grassy field or a gym mat.\"\n },\n {\n \"node_id\": \"3\",\n \"node_state_instruction\": \"Ensure there is enough space around you to avoid any obstacles.\"\n },\n {\n \"node_id\": \"4\",\n \"node_state_instruction\": \"Practice near a wall or with a spotter for safety.\"\n }\n]\n```", + "Given an instruction and several choices, decide which choice is most promising. Analyze each choice in detail, then conclude in the last line \"The best choice is {s}\", where s the integer id of the choice.\nChoice 3:\nEnsure there is enough space around you to avoid any obstacles.\n": "Choice 3: \n\nAnalyzing this choice, ensuring there is enough space around you to avoid any obstacles is a crucial step in maintaining safety and preventing accidents. By creating a buffer zone, you can react to unexpected obstacles or changes in the environment without risking collisions or other hazards.\n\nThe best choice is 3.", + "Given an instruction and several choices, decide which choice is most promising. Analyze each choice in detail, then conclude in the last line \"The best choice is {s}\", where s the integer id of the choice.\nChoice 2:\nConsider a flat and stable surface, such as a grassy field or a gym mat.\n": "Choice 2: Considering a flat and stable surface, such as a grassy field or a gym mat, is the most promising option. This choice provides a safe and secure environment for the activity, reducing the risk of injury. A grassy field offers a natural and soft surface, while a gym mat provides cushioning and stability. Both options allow for comfortable movement and can accommodate various physical activities.\n\nThe best choice is 2.", + "Given an instruction and several choices, decide which choice is most promising. Analyze each choice in detail, then conclude in the last line \"The best choice is {s}\", where s the integer id of the choice.\nChoice 4:\nPractice near a wall or with a spotter for safety.\n": "Choice 4: \n\nPractice near a wall or with a spotter for safety.\n\nAnalyzing this choice, practicing near a wall or with a spotter provides a safety measure to prevent accidents or injuries. When practicing a new skill or exercise, having a wall nearby can provide support and stability, reducing the risk of falling or losing balance. Similarly, having a spotter can offer assistance and guidance, ensuring that the practice is done safely and effectively.\n\nThe best choice is 4.", + "### Requirements\n1. Please summarize the latest dialogue based on the reference information (secondary) and dialogue history (primary). Do not include text that is irrelevant to the conversation.\n- The context is for reference only. If it is irrelevant to the user's search request history, please reduce its reference and usage.\n2. If there are citable links in the context, annotate them in the main text in the format [main text](citation link). If there are none in the context, do not write links.\n3. The reply should be graceful, clear, non-repetitive, smoothly written, and of moderate length, in {LANG}.\n\n### Dialogue History (For example)\nA: MLOps competitors\n\n### Current Question (For example)\nA: MLOps competitors\n\n### Current Reply (For example)\n1. Alteryx Designer: etc. if any\n2. Matlab: ditto\n3. IBM SPSS Statistics\n4. RapidMiner Studio\n5. DataRobot AI Platform\n6. Databricks Lakehouse Platform\n7. Amazon SageMaker\n8. Dataiku\n#SYSTEM_MSG_END#\n### Reference Information\nABC cleanser is preferred by many with oily skin.\nL'Oreal is a popular brand with many positive reviews.\n\n### Dialogue History\n\nuser: Which facial cleanser is good for oily skin?\n\n### Current Question\nuser: Which facial cleanser is good for oily skin?\n\n### Current Reply: Based on the information, please write the reply to the Question\n\n\n": "Based on the information provided, ABC cleanser is preferred by many with oily skin. It is a popular choice for individuals with oily skin due to its effectiveness. Additionally, L'Oreal is a well-known brand with many positive reviews, and they offer a range of products suitable for oily skin. Both of these options could be good choices for individuals with oily skin." } \ No newline at end of file diff --git a/tests/data/search_rsp_cache.json b/tests/data/search_rsp_cache.json new file mode 100644 index 000000000..822fb2069 --- /dev/null +++ b/tests/data/search_rsp_cache.json @@ -0,0 +1,879 @@ +{ + "aiohttp-get-https://serpapi.com/search-{\"params\": {\"api_key\": \"mock-serpapi-key\", \"engine\": \"google\", \"gl\": \"us\", \"google_domain\": \"google.com\", \"hl\": \"en\", \"num\": 8, \"output\": \"json\", \"q\": \"metagpt\", \"source\": \"python\"}}": { + "search_metadata": { + "id": "65a3f6595b54ef7f1dfbcdd2", + "status": "Success", + "json_endpoint": "https://serpapi.com/searches/f3454e001dacdae1/65a3f6595b54ef7f1dfbcdd2.json", + "created_at": "2024-01-14 14:57:29 UTC", + "processed_at": "2024-01-14 14:57:29 UTC", + "google_url": "https://www.google.com/search?q=metagpt&oq=metagpt&hl=en&gl=us&num=8&sourceid=chrome&ie=UTF-8", + "raw_html_file": "https://serpapi.com/searches/f3454e001dacdae1/65a3f6595b54ef7f1dfbcdd2.html", + "total_time_taken": 2.5 + }, + "search_parameters": { + "engine": "google", + "q": "metagpt", + "google_domain": "google.com", + "hl": "en", + "gl": "us", + "num": "8", + "device": "desktop" + }, + "search_information": { + "query_displayed": "metagpt", + "total_results": 91600, + "time_taken_displayed": 0.27, + "menu_items": [ + { + "position": 1, + "title": "News", + "link": "https://www.google.com/search?num=8&sca_esv=598392389&hl=en&gl=us&q=metagpt&tbm=nws&source=lnms&sa=X&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8Q0pQJegQIEBAB", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=8&q=metagpt&tbm=nws" + }, + { + "position": 2, + "title": "Images", + "link": "https://www.google.com/search?num=8&sca_esv=598392389&hl=en&gl=us&q=metagpt&tbm=isch&source=lnms&sa=X&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8Q0pQJegQIERAB", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google_images&gl=us&google_domain=google.com&hl=en&q=metagpt" + }, + { + "position": 3, + "title": "Perspectives", + "link": "https://www.google.com/search?num=8&sca_esv=598392389&hl=en&gl=us&q=metagpt&uds=AMwkrPv_BNR0fCL4lAUrdY_MslXnXP_8eZcaurn07wVclkT7zdZi70-PsAZ5cIYoShIriCGEG9cp7YID252SJZlezuQgGHVoaxAGC2P-K5BQMhuhn3rxBEI&udm=4&sa=X&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8Qs6gLegQIEhAB", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=8&q=metagpt" + }, + { + "position": 4, + "title": "Download", + "link": "https://www.google.com/search?num=8&sca_esv=598392389&hl=en&gl=us&q=MetaGPT+download&uds=AMwkrPs1tkKhl_yLs17ozqzdeOQpXginZ88vZAAruQSl2egWlmxzo18RJ2iSa2okRlGJpRvhNdkif_bMpSTk2MMlNadEZGUA9HcNBj9XUrqefB2G97SzGtM&sa=X&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8QxKsJegQIDhAB&ictx=0", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=8&q=MetaGPT+download" + }, + { + "position": 5, + "title": "Videos", + "link": "https://www.google.com/search?num=8&sca_esv=598392389&hl=en&gl=us&q=metagpt&tbm=vid&source=lnms&sa=X&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8Q0pQJegQINRAB", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google_videos&gl=us&google_domain=google.com&hl=en&num=8&q=metagpt" + }, + { + "position": 6, + "title": "Review", + "link": "https://www.google.com/search?num=8&sca_esv=598392389&hl=en&gl=us&q=MetaGPT+review&uds=AMwkrPsrb0_MXdPCtp0RJNoWQEuvuWMXOVdQk9bEznN4tlVCwT3QF14u76JluzhFRLe_8V0vj_J6GkI2lsgMS7iWf5vAS8_exlSGI2NPPyhxAtn0L9DpLP0&sa=X&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8QxKsJegQINhAB&ictx=0", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=8&q=MetaGPT+review" + }, + { + "position": 7, + "title": "Online", + "link": "https://www.google.com/search?num=8&sca_esv=598392389&hl=en&gl=us&q=MetaGPT+online&uds=AMwkrPsoRx99OfyO5-zj61oe0QMzGel38AesYPljQRlBU6r33ArXtPFSYaOzLdJPpJNVmudurhtqLwUnetN4svOtlXgjwySfgpxw9zgVeZ95Yk0B4ftC_Yw&sa=X&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8QxKsJegQINxAB&ictx=0", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=8&q=MetaGPT+online" + }, + { + "position": 8, + "title": "App", + "link": "https://www.google.com/search?num=8&sca_esv=598392389&hl=en&gl=us&q=Metagpt+app&uds=AMwkrPvM3iswphQGpo45MKxhFsVLtYmdTSGDwMjrC3YJfMStztBkIzhQ3LXUWRIS_9CLaKDV49EzlFRs65SDPWQRQ_UhZ9vnYjXCails2jTqGf73j7jxJ5g&sa=X&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8QxKsJegQIOBAB&ictx=0", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=8&q=Metagpt+app" + }, + { + "position": 9, + "title": "AI", + "link": "https://www.google.com/search?num=8&sca_esv=598392389&hl=en&gl=us&q=MetaGPT+AI&uds=AMwkrPtd3khZ7-4qbofZcpN4KpMaARLEVOHuvLVm0W3G2e-1vlpsKSHNi4ZplHhRz_p2lhtBxgOUBiCMoccC6ypD35_CMSI-u6d67n4mJNsyAnhftmvIlk8&sa=X&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8QxKsJegQIORAB&ictx=0", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=8&q=MetaGPT+AI" + } + ], + "organic_results_state": "Results for exact spelling" + }, + "inline_videos": [ + { + "position": 1, + "title": "How To Install MetaGPT - Build A Startup With One Prompt!!", + "link": "https://www.youtube.com/watch?v=uT75J_KG_aY", + "thumbnail": "https://serpapi.com/searches/65a3f6595b54ef7f1dfbcdd2/images/a0db2f9f70f02dd11e3d3d4154df9fd65b46b2fbf4804f7038c9ce99c8efea1c.jpeg", + "channel": "Matthew Berman", + "duration": "6:36", + "platform": "YouTube", + "date": "Aug 14, 2023" + }, + { + "position": 2, + "title": "MetaGPT HUGE Update: Autonomous AI Agents with ...", + "link": "https://www.youtube.com/watch?v=Xyws6iI-eH8", + "thumbnail": "https://serpapi.com/searches/65a3f6595b54ef7f1dfbcdd2/images/a0db2f9f70f02dd1d578e6031265d66299cf6aecd327454cdf67b92808f3dd86.jpeg", + "channel": "WorldofAI", + "duration": "11:38", + "platform": "YouTube", + "date": "3 weeks ago" + }, + { + "position": 3, + "title": "🚀 MetaGPT Setup: Launch a Startup with One ✍️ Prompt!", + "link": "https://www.youtube.com/watch?v=nqZlTV_L6Ao", + "thumbnail": "https://serpapi.com/searches/65a3f6595b54ef7f1dfbcdd2/images/a0db2f9f70f02dd1c5666bd22292fdc357357dac89294aabb55ebea0a40ce322.jpeg", + "channel": "Prompt Engineering", + "duration": "14:15", + "platform": "YouTube", + "date": "Sep 4, 2023", + "key_moments": [ + { + "time": "00:00", + "title": "Intro", + "link": "https://www.youtube.com/watch?v=nqZlTV_L6Ao&t=0", + "thumbnail": "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcQW-YKGXQDHplRpEDgL5Q-HlJ8HggTw_ghp_KWPh8xUcQ&s" + }, + { + "time": "00:12", + "title": "What is MetaGPT", + "link": "https://www.youtube.com/watch?v=nqZlTV_L6Ao&t=12", + "thumbnail": "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcRJ4RRAXOG6yvGPYqkuj5cMoiyYdAN6g7E3VU04SA3P7w&s" + }, + { + "time": "01:06", + "title": "Setup", + "link": "https://www.youtube.com/watch?v=nqZlTV_L6Ao&t=66", + "thumbnail": "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcTDlJBrAtfBkC8zI9wY4dOqVIaNFbjcYSZr4M1ZnD7RSw&s" + }, + { + "time": "05:23", + "title": "Changing configuration", + "link": "https://www.youtube.com/watch?v=nqZlTV_L6Ao&t=323", + "thumbnail": "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcT8MbsIRVXJy__UE4ba0FoCTMGfrykasHm3UGvSzMQAtQ&s" + }, + { + "time": "06:35", + "title": "How to Run", + "link": "https://www.youtube.com/watch?v=nqZlTV_L6Ao&t=395", + "thumbnail": "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcRuX6mOUVQVRzvnkOPYNcDpcazRC1QGeHhZh-Az9btUNA&s" + }, + { + "time": "09:02", + "title": "What outputs to expect", + "link": "https://www.youtube.com/watch?v=nqZlTV_L6Ao&t=542", + "thumbnail": "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcTFnNqvPfGrPnKJTJ1iOHGSNp6sVR5jn0Zy5N2JSGfeEQ&s" + }, + { + "time": "10:45", + "title": "Generated Design Documents", + "link": "https://www.youtube.com/watch?v=nqZlTV_L6Ao&t=645", + "thumbnail": "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcSN3I0gxudI4Mew93w_tw34HmWREz5XX8ArebReM3Y2_g&s" + }, + { + "time": "12:25", + "title": "Run the created code base", + "link": "https://www.youtube.com/watch?v=nqZlTV_L6Ao&t=745", + "thumbnail": "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcQLBx5bgKZ2Gqsu-PsIXuvtM0SBmHvBCndmKtresgqFCg&s" + } + ] + } + ], + "organic_results": [ + { + "position": 1, + "title": "geekan/MetaGPT: 🌟 The Multi-Agent Framework", + "link": "https://github.com/geekan/MetaGPT", + "redirect_link": "https://www.google.com/url?sa=t&source=web&rct=j&opi=89978449&url=https://github.com/geekan/MetaGPT&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8QFnoECBUQAQ", + "displayed_link": "https://github.com › geekan › MetaGPT", + "favicon": "https://serpapi.com/searches/65a3f6595b54ef7f1dfbcdd2/images/f37f87ccfb08b6fc2fe7e2076c022e7690f9b18357b8e5feb75a30ffbaaabfb1.png", + "snippet": "MetaGPT takes a one line requirement as input and outputs user stories / competitive analysis / requirements / data structures / APIs / documents, etc.", + "snippet_highlighted_words": [ + "MetaGPT" + ], + "sitelinks": { + "inline": [ + { + "title": "Roadmap", + "link": "https://github.com/geekan/MetaGPT/blob/main/docs/ROADMAP.md" + }, + { + "title": "README.md", + "link": "https://github.com/geekan/MetaGPT/blob/main/README.md" + }, + { + "title": "Issues 161", + "link": "https://github.com/geekan/MetaGPT/issues" + }, + { + "title": "Actions", + "link": "https://github.com/geekan/MetaGPT/actions" + } + ] + }, + "source": "GitHub" + }, + { + "position": 2, + "title": "MetaGPT: Meta Programming for A Multi-Agent ...", + "link": "https://arxiv.org/abs/2308.00352", + "redirect_link": "https://www.google.com/url?sa=t&source=web&rct=j&opi=89978449&url=https://arxiv.org/abs/2308.00352&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8QFnoECBMQAQ", + "displayed_link": "https://arxiv.org › cs", + "favicon": "https://serpapi.com/searches/65a3f6595b54ef7f1dfbcdd2/images/f37f87ccfb08b6fc2fe7e2076c022e76592372342f3f5dd76573e051b50f1bce.png", + "author": "by S Hong", + "cited_by": "Cited by 63", + "extracted_cited_by": 63, + "date": "2023", + "snippet": "Abstract:Remarkable progress has been made on automated problem solving through societies of agents based on large language models (LLMs).", + "source": "arXiv" + }, + { + "position": 3, + "title": "MetaGPT: a Multi-Agent Framework to Automate Your ...", + "link": "https://medium.datadriveninvestor.com/metagpt-a-multi-agent-framework-to-automate-your-software-company-4b6ae747cc36", + "redirect_link": "https://www.google.com/url?sa=t&source=web&rct=j&opi=89978449&url=https://medium.datadriveninvestor.com/metagpt-a-multi-agent-framework-to-automate-your-software-company-4b6ae747cc36&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8QFnoECBgQAQ", + "displayed_link": "https://medium.datadriveninvestor.com › metagpt-a-...", + "favicon": "https://serpapi.com/searches/65a3f6595b54ef7f1dfbcdd2/images/f37f87ccfb08b6fc2fe7e2076c022e76e8319069677ee18a99026fb1e05709cf.png", + "snippet": "MetaGPT is about to reach 10000 stars on Github. It's a Multi-Agent Framework that can behave as an engineer, product manager, architect, project managers.", + "snippet_highlighted_words": [ + "MetaGPT" + ], + "source": "DataDrivenInvestor" + }, + { + "position": 4, + "title": "MetaGPT - Apps on Google Play", + "link": "https://play.google.com/store/apps/details?id=com.metagpt.app&hl=en&gl=US", + "redirect_link": "https://www.google.com/url?sa=t&source=web&rct=j&opi=89978449&url=https://play.google.com/store/apps/details%3Fid%3Dcom.metagpt.app%26hl%3Den%26gl%3DUS&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8QFnoECCUQAQ", + "displayed_link": "https://play.google.com › store › apps › details › id=c...", + "favicon": "https://serpapi.com/searches/65a3f6595b54ef7f1dfbcdd2/images/f37f87ccfb08b6fc2fe7e2076c022e76334a7b2eeab09f16973a82a209ee6339.png", + "date": "Jan 1, 2024", + "snippet": "Real-time crypto monitor.Track prices, set alerts, seize opportunities instantly.", + "source": "Google Play" + }, + { + "position": 5, + "title": "MetaGPT: AI-Powered Web Development That Changes ...", + "link": "https://www.analyticsvidhya.com/blog/2024/01/meet-metagpt-the-chatgpt-powered-ai-assistant-that-turns-text-into-web-apps/", + "redirect_link": "https://www.google.com/url?sa=t&source=web&rct=j&opi=89978449&url=https://www.analyticsvidhya.com/blog/2024/01/meet-metagpt-the-chatgpt-powered-ai-assistant-that-turns-text-into-web-apps/&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8QFnoECCkQAQ", + "displayed_link": "https://www.analyticsvidhya.com › blog › 2024/01", + "favicon": "https://serpapi.com/searches/65a3f6595b54ef7f1dfbcdd2/images/f37f87ccfb08b6fc2fe7e2076c022e766a141f2bf05b1ab902f83ed00f4148a4.png", + "date": "Jan 4, 2024", + "snippet": "MetaGPT is an AI assistant that leverages the power of GPT-4, a state-of-the-art language model developed by OpenAI. ChatGPT is trained on vast ...", + "snippet_highlighted_words": [ + "MetaGPT" + ], + "source": "Analytics Vidhya" + }, + { + "position": 6, + "title": "MetaGPT | Discover AI use cases", + "link": "https://gpt3demo.com/apps/metagpt", + "redirect_link": "https://www.google.com/url?sa=t&source=web&rct=j&opi=89978449&url=https://gpt3demo.com/apps/metagpt&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8QFnoECCQQAQ", + "displayed_link": "https://gpt3demo.com › apps › metagpt", + "favicon": "https://serpapi.com/searches/65a3f6595b54ef7f1dfbcdd2/images/f37f87ccfb08b6fc2fe7e2076c022e76142721493557b5d95328dafb62b6b43a.jpeg", + "snippet": "Assign different roles to GPTs to form a collaborative software entity for complex tasks. MetaGPT takes a one-line requirement as input and outputs user ...", + "snippet_highlighted_words": [ + "MetaGPT" + ], + "source": "GPT-3 Demo" + } + ], + "related_searches": [ + { + "block_position": 1, + "query": "metagpt online", + "link": "https://www.google.com/search?num=8&sca_esv=598392389&hl=en&gl=us&q=MetaGPT+online&sa=X&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8Q1QJ6BAgnEAE", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=8&q=MetaGPT+online" + }, + { + "block_position": 1, + "query": "metagpt paper", + "link": "https://www.google.com/search?num=8&sca_esv=598392389&hl=en&gl=us&q=MetaGPT+paper&sa=X&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8Q1QJ6BAgoEAE", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=8&q=MetaGPT+paper" + }, + { + "block_position": 1, + "query": "Metagpt download", + "link": "https://www.google.com/search?num=8&sca_esv=598392389&hl=en&gl=us&q=Metagpt+download&sa=X&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8Q1QJ6BAgmEAE", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=8&q=Metagpt+download" + }, + { + "block_position": 1, + "query": "metagpt github", + "link": "https://www.google.com/search?num=8&sca_esv=598392389&hl=en&gl=us&q=Metagpt+github&sa=X&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8Q1QJ6BAgiEAE", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=8&q=Metagpt+github" + }, + { + "block_position": 1, + "query": "Metagpt review", + "link": "https://www.google.com/search?num=8&sca_esv=598392389&hl=en&gl=us&q=Metagpt+review&sa=X&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8Q1QJ6BAgjEAE", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=8&q=Metagpt+review" + }, + { + "block_position": 1, + "query": "metagpt ai", + "link": "https://www.google.com/search?num=8&sca_esv=598392389&hl=en&gl=us&q=MetaGPT+AI&sa=X&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8Q1QJ6BAggEAE", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=8&q=MetaGPT+AI" + }, + { + "block_position": 1, + "query": "metagpt huggingface", + "link": "https://www.google.com/search?num=8&sca_esv=598392389&hl=en&gl=us&q=MetaGPT+huggingface&sa=X&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8Q1QJ6BAghEAE", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=8&q=MetaGPT+huggingface" + }, + { + "block_position": 1, + "query": "metagpt openai", + "link": "https://www.google.com/search?num=8&sca_esv=598392389&hl=en&gl=us&q=Metagpt+OpenAI&sa=X&ved=2ahUKEwiZ6tvukd2DAxWuFlkFHbnFBv8Q1QJ6BAgfEAE", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=8&q=Metagpt+OpenAI" + } + ], + "pagination": { + "current": 1, + "next": "https://www.google.com/search?q=metagpt&oq=metagpt&hl=en&gl=us&num=8&start=8&sourceid=chrome&ie=UTF-8", + "other_pages": { + "2": "https://www.google.com/search?q=metagpt&oq=metagpt&hl=en&gl=us&num=8&start=8&sourceid=chrome&ie=UTF-8", + "3": "https://www.google.com/search?q=metagpt&oq=metagpt&hl=en&gl=us&num=8&start=16&sourceid=chrome&ie=UTF-8", + "4": "https://www.google.com/search?q=metagpt&oq=metagpt&hl=en&gl=us&num=8&start=24&sourceid=chrome&ie=UTF-8", + "5": "https://www.google.com/search?q=metagpt&oq=metagpt&hl=en&gl=us&num=8&start=32&sourceid=chrome&ie=UTF-8" + } + }, + "serpapi_pagination": { + "current": 1, + "next_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=8&q=metagpt&start=8", + "next": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=8&q=metagpt&start=8", + "other_pages": { + "2": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=8&q=metagpt&start=8", + "3": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=8&q=metagpt&start=16", + "4": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=8&q=metagpt&start=24", + "5": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=8&q=metagpt&start=32" + } + } + }, + "aiohttp-get-https://serpapi.com/search-{\"params\": {\"api_key\": \"mock-serpapi-key\", \"engine\": \"google\", \"gl\": \"us\", \"google_domain\": \"google.com\", \"hl\": \"en\", \"num\": 4, \"output\": \"json\", \"q\": \"metagpt\", \"source\": \"python\"}}": { + "search_metadata": { + "id": "65a3f65d8b7ed28c15233c79", + "status": "Success", + "json_endpoint": "https://serpapi.com/searches/2081c01f04a8e878/65a3f65d8b7ed28c15233c79.json", + "created_at": "2024-01-14 14:57:33 UTC", + "processed_at": "2024-01-14 14:57:33 UTC", + "google_url": "https://www.google.com/search?q=metagpt&oq=metagpt&hl=en&gl=us&num=4&sourceid=chrome&ie=UTF-8", + "raw_html_file": "https://serpapi.com/searches/2081c01f04a8e878/65a3f65d8b7ed28c15233c79.html", + "total_time_taken": 2.89 + }, + "search_parameters": { + "engine": "google", + "q": "metagpt", + "google_domain": "google.com", + "hl": "en", + "gl": "us", + "num": "4", + "device": "desktop" + }, + "search_information": { + "query_displayed": "metagpt", + "total_results": 91600, + "time_taken_displayed": 0.2, + "menu_items": [ + { + "position": 1, + "title": "News", + "link": "https://www.google.com/search?num=4&sca_esv=598392389&gl=us&hl=en&q=metagpt&tbm=nws&source=lnms&sa=X&ved=2ahUKEwigwuTwkd2DAxWyOkQIHc_uDdEQ0pQJegQIChAB", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=4&q=metagpt&tbm=nws" + }, + { + "position": 2, + "title": "Images", + "link": "https://www.google.com/search?num=4&sca_esv=598392389&gl=us&hl=en&q=metagpt&tbm=isch&source=lnms&sa=X&ved=2ahUKEwigwuTwkd2DAxWyOkQIHc_uDdEQ0pQJegQIDhAB", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google_images&gl=us&google_domain=google.com&hl=en&q=metagpt" + }, + { + "position": 3, + "title": "Perspectives", + "link": "https://www.google.com/search?num=4&sca_esv=598392389&gl=us&hl=en&q=metagpt&uds=AMwkrPv_BNR0fCL4lAUrdY_MslXnXP_8eZcaurn07wVclkT7zdZi70-PsAZ5cIYoShIriCGEG9cp7YID252SJZlezuQgGHVoaxAGC2P-K5BQMhuhn3rxBEI&udm=4&sa=X&ved=2ahUKEwigwuTwkd2DAxWyOkQIHc_uDdEQs6gLegQIDRAB", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=4&q=metagpt" + }, + { + "position": 4, + "title": "Download", + "link": "https://www.google.com/search?num=4&sca_esv=598392389&gl=us&hl=en&q=MetaGPT+download&uds=AMwkrPs1tkKhl_yLs17ozqzdeOQpXginZ88vZAAruQSl2egWlmxzo18RJ2iSa2okRlGJpRvhNdkif_bMpSTk2MMlNadEZGUA9HcNBj9XUrqefB2G97SzGtM&sa=X&ved=2ahUKEwigwuTwkd2DAxWyOkQIHc_uDdEQxKsJegQICxAB&ictx=0", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=4&q=MetaGPT+download" + }, + { + "position": 5, + "title": "Videos", + "link": "https://www.google.com/search?num=4&sca_esv=598392389&gl=us&hl=en&q=metagpt&tbm=vid&source=lnms&sa=X&ved=2ahUKEwigwuTwkd2DAxWyOkQIHc_uDdEQ0pQJegQILBAB", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google_videos&gl=us&google_domain=google.com&hl=en&num=4&q=metagpt" + }, + { + "position": 6, + "title": "Review", + "link": "https://www.google.com/search?num=4&sca_esv=598392389&gl=us&hl=en&q=MetaGPT+review&uds=AMwkrPsrb0_MXdPCtp0RJNoWQEuvuWMXOVdQk9bEznN4tlVCwT3QF14u76JluzhFRLe_8V0vj_J6GkI2lsgMS7iWf5vAS8_exlSGI2NPPyhxAtn0L9DpLP0&sa=X&ved=2ahUKEwigwuTwkd2DAxWyOkQIHc_uDdEQxKsJegQILhAB&ictx=0", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=4&q=MetaGPT+review" + }, + { + "position": 7, + "title": "Online", + "link": "https://www.google.com/search?num=4&sca_esv=598392389&gl=us&hl=en&q=MetaGPT+online&uds=AMwkrPsoRx99OfyO5-zj61oe0QMzGel38AesYPljQRlBU6r33ArXtPFSYaOzLdJPpJNVmudurhtqLwUnetN4svOtlXgjwySfgpxw9zgVeZ95Yk0B4ftC_Yw&sa=X&ved=2ahUKEwigwuTwkd2DAxWyOkQIHc_uDdEQxKsJegQILRAB&ictx=0", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=4&q=MetaGPT+online" + }, + { + "position": 8, + "title": "App", + "link": "https://www.google.com/search?num=4&sca_esv=598392389&gl=us&hl=en&q=Metagpt+app&uds=AMwkrPvM3iswphQGpo45MKxhFsVLtYmdTSGDwMjrC3YJfMStztBkIzhQ3LXUWRIS_9CLaKDV49EzlFRs65SDPWQRQ_UhZ9vnYjXCails2jTqGf73j7jxJ5g&sa=X&ved=2ahUKEwigwuTwkd2DAxWyOkQIHc_uDdEQxKsJegQILxAB&ictx=0", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=4&q=Metagpt+app" + }, + { + "position": 9, + "title": "AI", + "link": "https://www.google.com/search?num=4&sca_esv=598392389&gl=us&hl=en&q=MetaGPT+AI&uds=AMwkrPtd3khZ7-4qbofZcpN4KpMaARLEVOHuvLVm0W3G2e-1vlpsKSHNi4ZplHhRz_p2lhtBxgOUBiCMoccC6ypD35_CMSI-u6d67n4mJNsyAnhftmvIlk8&sa=X&ved=2ahUKEwigwuTwkd2DAxWyOkQIHc_uDdEQxKsJegQIMBAB&ictx=0", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=4&q=MetaGPT+AI" + } + ], + "organic_results_state": "Results for exact spelling" + }, + "inline_videos": [ + { + "position": 1, + "title": "How To Install MetaGPT - Build A Startup With One Prompt!!", + "link": "https://www.youtube.com/watch?v=uT75J_KG_aY", + "thumbnail": "https://serpapi.com/searches/65a3f65d8b7ed28c15233c79/images/bfd65a15364211be961855b9ca9c1cbfeecac1fc4f084deba696fe02f511b2b0.jpeg", + "channel": "Matthew Berman", + "duration": "6:36", + "platform": "YouTube", + "date": "Aug 14, 2023" + }, + { + "position": 2, + "title": "MetaGPT HUGE Update: Autonomous AI Agents with ...", + "link": "https://www.youtube.com/watch?v=Xyws6iI-eH8", + "thumbnail": "https://serpapi.com/searches/65a3f65d8b7ed28c15233c79/images/bfd65a15364211be43551974ef1dbd0b4a3780c1caa0ef2d1edaaee2ebc89b3c.jpeg", + "channel": "WorldofAI", + "duration": "11:38", + "platform": "YouTube", + "date": "3 weeks ago" + }, + { + "position": 3, + "title": "🚀 MetaGPT Setup: Launch a Startup with One ✍️ Prompt!", + "link": "https://www.youtube.com/watch?v=nqZlTV_L6Ao", + "thumbnail": "https://serpapi.com/searches/65a3f65d8b7ed28c15233c79/images/bfd65a15364211be779beff6d19f978b32bf888581454f54a19b9b01c5d9a6a8.jpeg", + "channel": "Prompt Engineering", + "duration": "14:15", + "platform": "YouTube", + "date": "Sep 4, 2023", + "key_moments": [ + { + "time": "00:00", + "title": "Intro", + "link": "https://www.youtube.com/watch?v=nqZlTV_L6Ao&t=0", + "thumbnail": "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcQW-YKGXQDHplRpEDgL5Q-HlJ8HggTw_ghp_KWPh8xUcQ&s" + }, + { + "time": "00:12", + "title": "What is MetaGPT", + "link": "https://www.youtube.com/watch?v=nqZlTV_L6Ao&t=12", + "thumbnail": "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcRJ4RRAXOG6yvGPYqkuj5cMoiyYdAN6g7E3VU04SA3P7w&s" + }, + { + "time": "01:06", + "title": "Setup", + "link": "https://www.youtube.com/watch?v=nqZlTV_L6Ao&t=66", + "thumbnail": "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcTDlJBrAtfBkC8zI9wY4dOqVIaNFbjcYSZr4M1ZnD7RSw&s" + }, + { + "time": "05:23", + "title": "Changing configuration", + "link": "https://www.youtube.com/watch?v=nqZlTV_L6Ao&t=323", + "thumbnail": "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcT8MbsIRVXJy__UE4ba0FoCTMGfrykasHm3UGvSzMQAtQ&s" + }, + { + "time": "06:35", + "title": "How to Run", + "link": "https://www.youtube.com/watch?v=nqZlTV_L6Ao&t=395", + "thumbnail": "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcRuX6mOUVQVRzvnkOPYNcDpcazRC1QGeHhZh-Az9btUNA&s" + }, + { + "time": "09:02", + "title": "What outputs to expect", + "link": "https://www.youtube.com/watch?v=nqZlTV_L6Ao&t=542", + "thumbnail": "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcTFnNqvPfGrPnKJTJ1iOHGSNp6sVR5jn0Zy5N2JSGfeEQ&s" + }, + { + "time": "10:45", + "title": "Generated Design Documents", + "link": "https://www.youtube.com/watch?v=nqZlTV_L6Ao&t=645", + "thumbnail": "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcSN3I0gxudI4Mew93w_tw34HmWREz5XX8ArebReM3Y2_g&s" + }, + { + "time": "12:25", + "title": "Run the created code base", + "link": "https://www.youtube.com/watch?v=nqZlTV_L6Ao&t=745", + "thumbnail": "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcQLBx5bgKZ2Gqsu-PsIXuvtM0SBmHvBCndmKtresgqFCg&s" + } + ] + } + ], + "organic_results": [ + { + "position": 1, + "title": "geekan/MetaGPT: 🌟 The Multi-Agent Framework", + "link": "https://github.com/geekan/MetaGPT", + "redirect_link": "https://www.google.com/url?sa=t&source=web&rct=j&opi=89978449&url=https://github.com/geekan/MetaGPT&ved=2ahUKEwigwuTwkd2DAxWyOkQIHc_uDdEQFnoECBcQAQ", + "displayed_link": "https://github.com › geekan › MetaGPT", + "favicon": "https://serpapi.com/searches/65a3f65d8b7ed28c15233c79/images/754322707626bed29162a2ba4a9960076a2cfb8f3558519e16fc8a6b74240174.png", + "snippet": "MetaGPT takes a one line requirement as input and outputs user stories / competitive analysis / requirements / data structures / APIs / documents, etc.", + "snippet_highlighted_words": [ + "MetaGPT" + ], + "sitelinks": { + "inline": [ + { + "title": "README.md", + "link": "https://github.com/geekan/MetaGPT/blob/main/README.md" + }, + { + "title": "Roadmap", + "link": "https://github.com/geekan/MetaGPT/blob/main/docs/ROADMAP.md" + }, + { + "title": "Issues 161", + "link": "https://github.com/geekan/MetaGPT/issues" + }, + { + "title": "Actions", + "link": "https://github.com/geekan/MetaGPT/actions" + } + ] + }, + "source": "GitHub" + }, + { + "position": 2, + "title": "MetaGPT: Meta Programming for A Multi-Agent ...", + "link": "https://arxiv.org/abs/2308.00352", + "redirect_link": "https://www.google.com/url?sa=t&source=web&rct=j&opi=89978449&url=https://arxiv.org/abs/2308.00352&ved=2ahUKEwigwuTwkd2DAxWyOkQIHc_uDdEQFnoECBUQAQ", + "displayed_link": "https://arxiv.org › cs", + "favicon": "https://serpapi.com/searches/65a3f65d8b7ed28c15233c79/images/754322707626bed29162a2ba4a996007d238ff23f244403a638b759e517db592.png", + "author": "by S Hong", + "cited_by": "Cited by 63", + "extracted_cited_by": 63, + "date": "2023", + "snippet": "Abstract:Remarkable progress has been made on automated problem solving through societies of agents based on large language models (LLMs).", + "source": "arXiv" + }, + { + "position": 3, + "title": "MetaGPT: a Multi-Agent Framework to Automate Your ...", + "link": "https://medium.datadriveninvestor.com/metagpt-a-multi-agent-framework-to-automate-your-software-company-4b6ae747cc36", + "redirect_link": "https://www.google.com/url?sa=t&source=web&rct=j&opi=89978449&url=https://medium.datadriveninvestor.com/metagpt-a-multi-agent-framework-to-automate-your-software-company-4b6ae747cc36&ved=2ahUKEwigwuTwkd2DAxWyOkQIHc_uDdEQFnoECBMQAQ", + "displayed_link": "https://medium.datadriveninvestor.com › metagpt-a-m...", + "favicon": "https://serpapi.com/searches/65a3f65d8b7ed28c15233c79/images/754322707626bed29162a2ba4a996007983965c804f215b78e84b23e3aabec98.png", + "snippet": "MetaGPT is about to reach 10000 stars on Github. It's a Multi-Agent Framework that can behave as an engineer, product manager, architect, project managers.", + "snippet_highlighted_words": [ + "MetaGPT" + ], + "source": "DataDrivenInvestor" + } + ], + "related_searches": [ + { + "block_position": 1, + "query": "metagpt online", + "link": "https://www.google.com/search?num=4&sca_esv=598392389&gl=us&hl=en&q=MetaGPT+online&sa=X&ved=2ahUKEwigwuTwkd2DAxWyOkQIHc_uDdEQ1QJ6BAgmEAE", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=4&q=MetaGPT+online" + }, + { + "block_position": 1, + "query": "metagpt paper", + "link": "https://www.google.com/search?num=4&sca_esv=598392389&gl=us&hl=en&q=MetaGPT+paper&sa=X&ved=2ahUKEwigwuTwkd2DAxWyOkQIHc_uDdEQ1QJ6BAglEAE", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=4&q=MetaGPT+paper" + }, + { + "block_position": 1, + "query": "Metagpt download", + "link": "https://www.google.com/search?num=4&sca_esv=598392389&gl=us&hl=en&q=Metagpt+download&sa=X&ved=2ahUKEwigwuTwkd2DAxWyOkQIHc_uDdEQ1QJ6BAgjEAE", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=4&q=Metagpt+download" + }, + { + "block_position": 1, + "query": "metagpt github", + "link": "https://www.google.com/search?num=4&sca_esv=598392389&gl=us&hl=en&q=Metagpt+github&sa=X&ved=2ahUKEwigwuTwkd2DAxWyOkQIHc_uDdEQ1QJ6BAgkEAE", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=4&q=Metagpt+github" + }, + { + "block_position": 1, + "query": "Metagpt review", + "link": "https://www.google.com/search?num=4&sca_esv=598392389&gl=us&hl=en&q=Metagpt+review&sa=X&ved=2ahUKEwigwuTwkd2DAxWyOkQIHc_uDdEQ1QJ6BAgiEAE", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=4&q=Metagpt+review" + }, + { + "block_position": 1, + "query": "metagpt ai", + "link": "https://www.google.com/search?num=4&sca_esv=598392389&gl=us&hl=en&q=MetaGPT+AI&sa=X&ved=2ahUKEwigwuTwkd2DAxWyOkQIHc_uDdEQ1QJ6BAgfEAE", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=4&q=MetaGPT+AI" + }, + { + "block_position": 1, + "query": "metagpt huggingface", + "link": "https://www.google.com/search?num=4&sca_esv=598392389&gl=us&hl=en&q=MetaGPT+huggingface&sa=X&ved=2ahUKEwigwuTwkd2DAxWyOkQIHc_uDdEQ1QJ6BAggEAE", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=4&q=MetaGPT+huggingface" + }, + { + "block_position": 1, + "query": "metagpt openai", + "link": "https://www.google.com/search?num=4&sca_esv=598392389&gl=us&hl=en&q=Metagpt+OpenAI&sa=X&ved=2ahUKEwigwuTwkd2DAxWyOkQIHc_uDdEQ1QJ6BAghEAE", + "serpapi_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=4&q=Metagpt+OpenAI" + } + ], + "pagination": { + "current": 1, + "next": "https://www.google.com/search?q=metagpt&oq=metagpt&hl=en&gl=us&num=4&start=4&sourceid=chrome&ie=UTF-8", + "other_pages": { + "2": "https://www.google.com/search?q=metagpt&oq=metagpt&hl=en&gl=us&num=4&start=4&sourceid=chrome&ie=UTF-8", + "3": "https://www.google.com/search?q=metagpt&oq=metagpt&hl=en&gl=us&num=4&start=8&sourceid=chrome&ie=UTF-8", + "4": "https://www.google.com/search?q=metagpt&oq=metagpt&hl=en&gl=us&num=4&start=12&sourceid=chrome&ie=UTF-8", + "5": "https://www.google.com/search?q=metagpt&oq=metagpt&hl=en&gl=us&num=4&start=16&sourceid=chrome&ie=UTF-8" + } + }, + "serpapi_pagination": { + "current": 1, + "next_link": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=4&q=metagpt&start=4", + "next": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=4&q=metagpt&start=4", + "other_pages": { + "2": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=4&q=metagpt&start=4", + "3": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=4&q=metagpt&start=8", + "4": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=4&q=metagpt&start=12", + "5": "https://serpapi.com/search.json?device=desktop&engine=google&gl=us&google_domain=google.com&hl=en&num=4&q=metagpt&start=16" + } + } + }, + "httplib2-GET-https://customsearch.googleapis.com/customsearch/v1-{\"params\": {\"q\": \"metagpt\", \"num\": \"8\", \"cx\": \"mock-google-cse\", \"key\": \"mock-google-key\", \"alt\": \"json\"}}": "{\n \"kind\": \"customsearch#search\",\n \"url\": {\n \"type\": \"application/json\",\n \"template\": \"https://www.googleapis.com/customsearch/v1?q={searchTerms}&num={count?}&start={startIndex?}&lr={language?}&safe={safe?}&cx={cx?}&sort={sort?}&filter={filter?}&gl={gl?}&cr={cr?}&googlehost={googleHost?}&c2coff={disableCnTwTranslation?}&hq={hq?}&hl={hl?}&siteSearch={siteSearch?}&siteSearchFilter={siteSearchFilter?}&exactTerms={exactTerms?}&excludeTerms={excludeTerms?}&linkSite={linkSite?}&orTerms={orTerms?}&dateRestrict={dateRestrict?}&lowRange={lowRange?}&highRange={highRange?}&searchType={searchType}&fileType={fileType?}&rights={rights?}&imgSize={imgSize?}&imgType={imgType?}&imgColorType={imgColorType?}&imgDominantColor={imgDominantColor?}&alt=json\"\n },\n \"queries\": {\n \"request\": [\n {\n \"title\": \"Google Custom Search - metagpt\",\n \"totalResults\": \"71300\",\n \"searchTerms\": \"metagpt\",\n \"count\": 8,\n \"startIndex\": 1,\n \"inputEncoding\": \"utf8\",\n \"outputEncoding\": \"utf8\",\n \"safe\": \"off\",\n \"cx\": \"mock-google-cse\"\n }\n ],\n \"nextPage\": [\n {\n \"title\": \"Google Custom Search - metagpt\",\n \"totalResults\": \"71300\",\n \"searchTerms\": \"metagpt\",\n \"count\": 8,\n \"startIndex\": 9,\n \"inputEncoding\": \"utf8\",\n \"outputEncoding\": \"utf8\",\n \"safe\": \"off\",\n \"cx\": \"mock-google-cse\"\n }\n ]\n },\n \"context\": {\n \"title\": \"metagpt1\"\n },\n \"searchInformation\": {\n \"searchTime\": 0.353952,\n \"formattedSearchTime\": \"0.35\",\n \"totalResults\": \"71300\",\n \"formattedTotalResults\": \"71,300\"\n },\n \"items\": [\n {\n \"kind\": \"customsearch#result\",\n \"title\": \"geekan/MetaGPT: The Multi-Agent Framework: Given one ... - GitHub\",\n \"htmlTitle\": \"geekan/MetaGPT: The Multi-Agent Framework: Given one ... - GitHub\",\n \"link\": \"https://github.com/geekan/MetaGPT\",\n \"displayLink\": \"github.com\",\n \"snippet\": \"The Multi-Agent Framework: Given one line Requirement, return PRD, Design, Tasks, Repo - GitHub - geekan/MetaGPT: The Multi-Agent Framework: Given one ...\",\n \"htmlSnippet\": \"The Multi-Agent Framework: Given one line Requirement, return PRD, Design, Tasks, Repo - GitHub - geekan/\\u003cb\\u003eMetaGPT\\u003c/b\\u003e: The Multi-Agent Framework: Given one ...\",\n \"cacheId\": \"gsshb0APPNgJ\",\n \"formattedUrl\": \"https://github.com/geekan/MetaGPT\",\n \"htmlFormattedUrl\": \"https://github.com/geekan/\\u003cb\\u003eMetaGPT\\u003c/b\\u003e\",\n \"pagemap\": {\n \"cse_thumbnail\": [\n {\n \"src\": \"https://encrypted-tbn1.gstatic.com/images?q=tbn:ANd9GcRuD8YUvRcltmdoxKyuIbt8UZhg3LE5mwNX7KPXDB15YIJRKdT2m5JiweuS\",\n \"width\": \"318\",\n \"height\": \"159\"\n }\n ],\n \"softwaresourcecode\": [\n {\n \"author\": \"geekan\",\n \"name\": \"MetaGPT\",\n \"text\": \"MetaGPT: The Multi-Agent Framework Assign different roles to GPTs to form a collaborative software entity for complex tasks. MetaGPT takes a one line requirement as input and outputs user stories...\"\n }\n ],\n \"metatags\": [\n {\n \"octolytics-url\": \"https://collector.github.com/github/collect\",\n \"apple-itunes-app\": \"app-id=1477376905, app-argument=https://github.com/geekan/MetaGPT\",\n \"og:image\": \"https://opengraph.githubassets.com/6178eb2aa6711c676eafc956e52345e71225c58cd0a666b54871171e847c0905/geekan/MetaGPT\",\n \"twitter:card\": \"summary_large_image\",\n \"og:image:width\": \"1200\",\n \"theme-color\": \"#1e2327\",\n \"og:site_name\": \"GitHub\",\n \"hovercard-subject-tag\": \"repository:660551251\",\n \"turbo-body-classes\": \"logged-out env-production page-responsive\",\n \"html-safe-nonce\": \"a6964edcf12c9ea83de0bf16db8105c60333287597018548250a0fa92b93d3f9\",\n \"expected-hostname\": \"github.com\",\n \"og:description\": \"🌟 The Multi-Agent Framework: Given one line Requirement, return PRD, Design, Tasks, Repo - GitHub - geekan/MetaGPT: 🌟 The Multi-Agent Framework: Given one line Requirement, return PRD, Design, Task...\",\n \"browser-errors-url\": \"https://api.github.com/_private/browser/errors\",\n \"octolytics-dimension-user_login\": \"geekan\",\n \"hostname\": \"github.com\",\n \"twitter:site\": \"@github\",\n \"browser-stats-url\": \"https://api.github.com/_private/browser/stats\",\n \"route-pattern\": \"/:user_id/:repository\",\n \"visitor-payload\": \"eyJyZWZlcnJlciI6IiIsInJlcXVlc3RfaWQiOiJBNjIxOjk2OUU6OTZERjlEQzpEM0UxM0RFOjY1QTNBQjU0IiwidmlzaXRvcl9pZCI6IjU2ODA2NDI2MDQ0ODY5MzA3NyIsInJlZ2lvbl9lZGdlIjoiaWFkIiwicmVnaW9uX3JlbmRlciI6ImlhZCJ9\",\n \"github-keyboard-shortcuts\": \"repository\",\n \"octolytics-dimension-repository_id\": \"660551251\",\n \"octolytics-dimension-repository_network_root_nwo\": \"geekan/MetaGPT\",\n \"twitter:title\": \"GitHub - geekan/MetaGPT: 🌟 The Multi-Agent Framework: Given one line Requirement, return PRD, Design, Tasks, Repo\",\n \"og:image:alt\": \"🌟 The Multi-Agent Framework: Given one line Requirement, return PRD, Design, Tasks, Repo - GitHub - geekan/MetaGPT: 🌟 The Multi-Agent Framework: Given one line Requirement, return PRD, Design, Task...\",\n \"og:type\": \"object\",\n \"optimizely-datafile\": \"{\\\"accountId\\\": \\\"16737760170\\\", \\\"projectId\\\": \\\"16737760170\\\", \\\"revision\\\": \\\"23\\\", \\\"attributes\\\": [{\\\"id\\\": \\\"16822470375\\\", \\\"key\\\": \\\"user_id\\\"}, {\\\"id\\\": \\\"17143601254\\\", \\\"key\\\": \\\"spammy\\\"}, {\\\"id\\\": \\\"18175660309\\\", \\\"key\\\": \\\"organization_plan\\\"}, {\\\"id\\\": \\\"18813001570\\\", \\\"key\\\": \\\"is_logged_in\\\"}, {\\\"id\\\": \\\"19073851829\\\", \\\"key\\\": \\\"geo\\\"}, {\\\"id\\\": \\\"20175462351\\\", \\\"key\\\": \\\"requestedCurrency\\\"}, {\\\"id\\\": \\\"20785470195\\\", \\\"key\\\": \\\"country_code\\\"}, {\\\"id\\\": \\\"21656311196\\\", \\\"key\\\": \\\"opened_downgrade_dialog\\\"}], \\\"audiences\\\": [{\\\"id\\\": \\\"$opt_dummy_audience\\\", \\\"name\\\": \\\"Optimizely-Generated Audience for Backwards Compatibility\\\", \\\"conditions\\\": \\\"[\\\\\\\"or\\\\\\\", {\\\\\\\"match\\\\\\\": \\\\\\\"exact\\\\\\\", \\\\\\\"name\\\\\\\": \\\\\\\"$opt_dummy_attribute\\\\\\\", \\\\\\\"type\\\\\\\": \\\\\\\"custom_attribute\\\\\\\", \\\\\\\"value\\\\\\\": \\\\\\\"$opt_dummy_value\\\\\\\"}]\\\"}], \\\"version\\\": \\\"4\\\", \\\"events\\\": [{\\\"id\\\": \\\"18188530140\\\", \\\"experimentIds\\\": [], \\\"key\\\": \\\"test_event\\\"}], \\\"integrations\\\": [], \\\"anonymizeIP\\\": true, \\\"botFiltering\\\": false, \\\"typedAudiences\\\": [], \\\"variables\\\": [], \\\"environmentKey\\\": \\\"production\\\", \\\"sdkKey\\\": \\\"UpVyJZaLVEGwJPQWf5pAD\\\", \\\"featureFlags\\\": [], \\\"rollouts\\\": [],\",\n \"og:title\": \"GitHub - geekan/MetaGPT: 🌟 The Multi-Agent Framework: Given one line Requirement, return PRD, Design, Tasks, Repo\",\n \"visitor-hmac\": \"471691c5f7e3204061bb09c630c440708f5c08a2824d60b0f28eea873d474cd7\",\n \"og:image:height\": \"600\",\n \"turbo-cache-control\": \"no-preview\",\n \"request-id\": \"A621:969E:96DF9DC:D3E13DE:65A3AB54\",\n \"analytics-location\": \"/\\u003cuser-name\\u003e/\\u003crepo-name\\u003e\",\n \"color-scheme\": \"light dark\",\n \"octolytics-dimension-repository_is_fork\": \"false\",\n \"go-import\": \"github.com/geekan/MetaGPT git https://github.com/geekan/MetaGPT.git\",\n \"browser-optimizely-client-errors-url\": \"https://api.github.com/_private/browser/optimizely_client/errors\",\n \"twitter:image:src\": \"https://opengraph.githubassets.com/6178eb2aa6711c676eafc956e52345e71225c58cd0a666b54871171e847c0905/geekan/MetaGPT\",\n \"octolytics-dimension-user_id\": \"2707039\",\n \"octolytics-dimension-repository_public\": \"true\",\n \"fb:app_id\": \"1401488693436528\",\n \"octolytics-dimension-repository_network_root_id\": \"660551251\",\n \"octolytics-dimension-repository_nwo\": \"geekan/MetaGPT\",\n \"viewport\": \"width=device-width\",\n \"twitter:description\": \"🌟 The Multi-Agent Framework: Given one line Requirement, return PRD, Design, Tasks, Repo - GitHub - geekan/MetaGPT: 🌟 The Multi-Agent Framework: Given one line Requirement, return PRD, Design, Task...\",\n \"current-catalog-service-hash\": \"82c569b93da5c18ed649ebd4c2c79437db4611a6a1373e805a3cb001c64130b7\",\n \"og:url\": \"https://github.com/geekan/MetaGPT\"\n }\n ],\n \"cse_image\": [\n {\n \"src\": \"https://opengraph.githubassets.com/6178eb2aa6711c676eafc956e52345e71225c58cd0a666b54871171e847c0905/geekan/MetaGPT\"\n }\n ]\n }\n },\n {\n \"kind\": \"customsearch#result\",\n \"title\": \"[2308.00352] MetaGPT: Meta Programming for A Multi-Agent ...\",\n \"htmlTitle\": \"[2308.00352] \\u003cb\\u003eMetaGPT\\u003c/b\\u003e: Meta Programming for A Multi-Agent ...\",\n \"link\": \"https://arxiv.org/abs/2308.00352\",\n \"displayLink\": \"arxiv.org\",\n \"snippet\": \"Aug 1, 2023 ... Computer Science \\u003e Artificial Intelligence · Title:MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework · Bibliographic and ...\",\n \"htmlSnippet\": \"Aug 1, 2023 \\u003cb\\u003e...\\u003c/b\\u003e Computer Science > Artificial Intelligence · Title:\\u003cb\\u003eMetaGPT\\u003c/b\\u003e: Meta Programming for A Multi-Agent Collaborative Framework · Bibliographic and ...\",\n \"cacheId\": \"8_tddNY0jEYJ\",\n \"formattedUrl\": \"https://arxiv.org/abs/2308.00352\",\n \"htmlFormattedUrl\": \"https://arxiv.org/abs/2308.00352\",\n \"pagemap\": {\n \"cse_thumbnail\": [\n {\n \"src\": \"https://encrypted-tbn3.gstatic.com/images?q=tbn:ANd9GcStsc5IszP_UC7vkymrk7PhjHGOFQhTh862xtJcQkxDem2IteJQXpob6_Vb\",\n \"width\": \"336\",\n \"height\": \"150\"\n }\n ],\n \"metatags\": [\n {\n \"og:image\": \"/static/browse/0.3.4/images/arxiv-logo-fb.png\",\n \"theme-color\": \"#ffffff\",\n \"og:image:width\": \"1200\",\n \"twitter:card\": \"summary\",\n \"citation_title\": \"MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework\",\n \"og:site_name\": \"arXiv.org\",\n \"citation_date\": \"2023/08/01\",\n \"og:description\": \"Remarkable progress has been made on automated problem solving through societies of agents based on large language models (LLMs). Existing LLM-based multi-agent systems can already solve simple dialogue tasks. Solutions to more complex tasks, however, are complicated through logic inconsistencies due to cascading hallucinations caused by naively chaining LLMs. Here we introduce MetaGPT, an innovative meta-programming framework incorporating efficient human workflows into LLM-based multi-agent collaborations. MetaGPT encodes Standardized Operating Procedures (SOPs) into prompt sequences for more streamlined workflows, thus allowing agents with human-like domain expertise to verify intermediate results and reduce errors. MetaGPT utilizes an assembly line paradigm to assign diverse roles to various agents, efficiently breaking down complex tasks into subtasks involving many agents working together. On collaborative software engineering benchmarks, MetaGPT generates more coherent solutions than previous chat-base\",\n \"og:image:secure_url\": \"/static/browse/0.3.4/images/arxiv-logo-fb.png\",\n \"twitter:image\": \"https://static.arxiv.org/icons/twitter/arxiv-logo-twitter-square.png\",\n \"citation_arxiv_id\": \"2308.00352\",\n \"citation_online_date\": \"2023/11/06\",\n \"twitter:image:alt\": \"arXiv logo\",\n \"twitter:site\": \"@arxiv\",\n \"citation_pdf_url\": \"http://arxiv.org/pdf/2308.00352.pdf\",\n \"msapplication-tilecolor\": \"#da532c\",\n \"og:type\": \"website\",\n \"og:image:alt\": \"arXiv logo\",\n \"twitter:title\": \"MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework\",\n \"og:title\": \"MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework\",\n \"citation_abstract\": \"Remarkable progress has been made on automated problem solving through societies of agents based on large language models (LLMs). Existing LLM-based multi-agent systems can already solve simple dialogue tasks. Solutions to more complex tasks, however, are complicated through logic inconsistencies due to cascading hallucinations caused by naively chaining LLMs. Here we introduce MetaGPT, an innovative meta-programming framework incorporating efficient human workflows into LLM-based multi-agent collaborations. MetaGPT encodes Standardized Operating Procedures (SOPs) into prompt sequences for more streamlined workflows, thus allowing agents with human-like domain expertise to verify intermediate results and reduce errors. MetaGPT utilizes an assembly line paradigm to assign diverse roles to various agents, efficiently breaking down complex tasks into subtasks involving many agents working together. On collaborative software engineering benchmarks, MetaGPT generates more coherent solutions than previous chat-base\",\n \"og:image:height\": \"700\",\n \"citation_author\": \"Hong, Sirui\",\n \"viewport\": \"width=device-width, initial-scale=1\",\n \"twitter:description\": \"Remarkable progress has been made on automated problem solving through societies of agents based on large language models (LLMs). Existing LLM-based multi-agent systems can already solve simple...\",\n \"og:url\": \"https://arxiv.org/abs/2308.00352v5\"\n }\n ],\n \"cse_image\": [\n {\n \"src\": \"https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-one-color-white.svg\"\n }\n ]\n }\n },\n {\n \"kind\": \"customsearch#result\",\n \"title\": \"MetaGPT: Complete Guide to the Best AI Agent Available Right Now ...\",\n \"htmlTitle\": \"\\u003cb\\u003eMetaGPT\\u003c/b\\u003e: Complete Guide to the Best AI Agent Available Right Now ...\",\n \"link\": \"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\n \"displayLink\": \"www.unite.ai\",\n \"snippet\": \"Sep 11, 2023 ... The beauty of MetaGPT lies in its structuring. It capitalizes on meta-programming techniques to manipulate, analyze, and transform code in real- ...\",\n \"htmlSnippet\": \"Sep 11, 2023 \\u003cb\\u003e...\\u003c/b\\u003e The beauty of \\u003cb\\u003eMetaGPT\\u003c/b\\u003e lies in its structuring. It capitalizes on meta-programming techniques to manipulate, analyze, and transform code in real- ...\",\n \"cacheId\": \"qkZULzxVHNAJ\",\n \"formattedUrl\": \"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-...\",\n \"htmlFormattedUrl\": \"https://www.unite.ai/\\u003cb\\u003emetagpt\\u003c/b\\u003e-complete-guide-to-the-best-ai-agent-available-...\",\n \"pagemap\": {\n \"cse_thumbnail\": [\n {\n \"src\": \"https://encrypted-tbn2.gstatic.com/images?q=tbn:ANd9GcSVwf1WWLtVpqJCZ1E_t7TrpSZ7nrwsCUWar6x9YzlOsX1aSH7EGHbkIlY\",\n \"width\": \"290\",\n \"height\": \"174\"\n }\n ],\n \"imageobject\": [\n {\n \"width\": \"1000\",\n \"url\": \"https://www.unite.ai/wp-content/uploads/2023/09/Heisenbergforlife_Center_the_scene_in_zoomed_scope_around_a_hum_7f069632-eda5-4edd-858b-cb44fec82929-1000x600.png\",\n \"height\": \"600\"\n },\n {\n \"url\": \"https://www.unite.ai/wp-content/uploads/2021/03/logoUNITE230X30BLACK-1.svg\"\n }\n ],\n \"person\": [\n {\n \"name\": \"Aayush Mittal\"\n }\n ],\n \"organization\": [\n {\n \"name\": \"Unite.AI\"\n }\n ],\n \"metatags\": [\n {\n \"og:image\": \"https://www.unite.ai/wp-content/uploads/2023/09/Heisenbergforlife_Center_the_scene_in_zoomed_scope_around_a_hum_7f069632-eda5-4edd-858b-cb44fec82929-1000x600.png\",\n \"twitter:card\": \"summary\",\n \"article:published_time\": \"2023-09-11T18:03:48+00:00\",\n \"og:image:width\": \"1121\",\n \"og:site_name\": \"Unite.AI\",\n \"twitter:url\": \"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\n \"twitter:label1\": \"Written by\",\n \"twitter:label2\": \"Est. reading time\",\n \"og:image:type\": \"image/png\",\n \"msapplication-tileimage\": \"https://www.unite.ai/wp-content/uploads/2023/09/Heisenbergforlife_Center_the_scene_in_zoomed_scope_around_a_hum_7f069632-eda5-4edd-858b-cb44fec82929.png\",\n \"og:description\": \"Discover why MetaGPT outperforms AutoGPT, BabyAgi, and other AI agents in complex coding tasks. Our in-depth article guides you through the setup process and provides illustrative examples. Build GPT-powered microapps with a single line of prompt\",\n \"twitter:creator\": \"@UniteAI\",\n \"twitter:image\": \"https://www.unite.ai/wp-content/uploads/2023/09/Heisenbergforlife_Center_the_scene_in_zoomed_scope_around_a_hum_7f069632-eda5-4edd-858b-cb44fec82929-1000x600.png\",\n \"article:publisher\": \"https://www.facebook.com/uniteai\",\n \"twitter:data1\": \"Aayush Mittal\",\n \"og:image:secure_url\": \"https://www.unite.ai/wp-content/uploads/2023/09/Heisenbergforlife_Center_the_scene_in_zoomed_scope_around_a_hum_7f069632-eda5-4edd-858b-cb44fec82929.png\",\n \"twitter:data2\": \"9 minutes\",\n \"twitter:site\": \"@UniteAI\",\n \"og:video:type\": \"video/mp4\",\n \"uri-translation\": \"on\",\n \"og:type\": \"article\",\n \"twitter:title\": \"MetaGPT: Complete Guide to the Best AI Agent Available Right Now\",\n \"og:image:alt\": \"MetaGPBassed Illustration of human and machine collaborationT\",\n \"author\": \"Aayush Mittal\",\n \"og:title\": \"MetaGPT: Complete Guide to the Best AI Agent Available Right Now\",\n \"og:image:height\": \"628\",\n \"og:updated_time\": \"2023-09-11T14:03:48-04:00\",\n \"article:tag\": \"AI AGENTS\",\n \"og:video\": \"https://www.unite.ai/wp-content/uploads/2023/09/ezgif.com-optimize-online-video-cutter.com_.mp4\",\n \"viewport\": \"width=device-width,initial-scale=1.0,user-scalable=yes\",\n \"og:locale\": \"en_US\",\n \"og:rich_attachment\": \"1\",\n \"og:url\": \"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\"\n }\n ],\n \"cse_image\": [\n {\n \"src\": \"https://www.unite.ai/wp-content/uploads/2023/09/Heisenbergforlife_Center_the_scene_in_zoomed_scope_around_a_hum_7f069632-eda5-4edd-858b-cb44fec82929-1000x600.png\"\n }\n ],\n \"blogposting\": [\n {\n \"image\": \"https://www.unite.ai/wp-content/uploads/2023/09/Heisenbergforlife_Center_the_scene_in_zoomed_scope_around_a_hum_7f069632-eda5-4edd-858b-cb44fec82929.png\",\n \"datemodified\": \"2023-09-11T18:03:48+00:00\",\n \"author\": \"Aayush Mittal\",\n \"name\": \"MetaGPT: Complete Guide to the Best AI Agent Available Right Now\",\n \"description\": \"With Large Language Models (LLMs) like ChatGPT, OpenAI has witnessed a surge in enterprise and user adoption, currently raking in around $80 million in monthly revenue. According to a recent...\",\n \"headline\": \"MetaGPT: Complete Guide to the Best AI Agent Available Right Now\",\n \"datepublished\": \"2023-09-11\"\n }\n ],\n \"newsarticle\": [\n {\n \"datemodified\": \"2023-09-11\",\n \"keywords\": \"AI AGENTSAutoGPTDockergenerative aiLLMMetaGPTnlpPROMPT ENGINEERINGpython\",\n \"headline\": \"MetaGPT: Complete Guide to the Best AI Agent Available Right Now\",\n \"datepublished\": \"2023-09-11\"\n }\n ]\n }\n },\n {\n \"kind\": \"customsearch#result\",\n \"title\": \"Thoughts on MetaGPT : r/ProductManagement\",\n \"htmlTitle\": \"Thoughts on \\u003cb\\u003eMetaGPT\\u003c/b\\u003e : r/ProductManagement\",\n \"link\": \"https://www.reddit.com/r/ProductManagement/comments/163vekc/thoughts_on_metagpt/\",\n \"displayLink\": \"www.reddit.com\",\n \"snippet\": \"Aug 28, 2023 ... Thoughts on MetaGPT. YT shorts - a quick summaryExplainer YT video - maynot be the best, but beginner friendly. ... PS: feel free to link more ...\",\n \"htmlSnippet\": \"Aug 28, 2023 \\u003cb\\u003e...\\u003c/b\\u003e Thoughts on \\u003cb\\u003eMetaGPT\\u003c/b\\u003e. YT shorts - a quick summaryExplainer YT video - maynot be the best, but beginner friendly. ... PS: feel free to link more ...\",\n \"cacheId\": \"fDkEZ_skdhcJ\",\n \"formattedUrl\": \"https://www.reddit.com/r/ProductManagement/.../thoughts_on_metagpt/\",\n \"htmlFormattedUrl\": \"https://www.reddit.com/r/ProductManagement/.../thoughts_on_\\u003cb\\u003emetagpt\\u003c/b\\u003e/\",\n \"pagemap\": {\n \"cse_thumbnail\": [\n {\n \"src\": \"https://encrypted-tbn2.gstatic.com/images?q=tbn:ANd9GcSnWudLgGG_2ao_7EWw3EW58JBUQkJ1m4LOHzyiajVHq10p0_TNAeCRlik\",\n \"width\": \"259\",\n \"height\": \"194\"\n }\n ],\n \"metatags\": [\n {\n \"og:image\": \"https://share.redd.it/preview/post/163vekc\",\n \"theme-color\": \"#000000\",\n \"og:image:width\": \"1200\",\n \"og:type\": \"website\",\n \"og:image:alt\": \"An image containing a preview of the post\",\n \"twitter:card\": \"summary_large_image\",\n \"twitter:title\": \"r/ProductManagement on Reddit: Thoughts on MetaGPT\",\n \"og:site_name\": \"Reddit\",\n \"og:title\": \"r/ProductManagement on Reddit: Thoughts on MetaGPT\",\n \"og:image:height\": \"630\",\n \"msapplication-navbutton-color\": \"#000000\",\n \"og:description\": \"Posted by u/CheraCholan - No votes and 4 comments\",\n \"twitter:image\": \"https://share.redd.it/preview/post/163vekc\",\n \"apple-mobile-web-app-status-bar-style\": \"black\",\n \"twitter:site\": \"@reddit\",\n \"viewport\": \"width=device-width, initial-scale=1, viewport-fit=cover\",\n \"apple-mobile-web-app-capable\": \"yes\",\n \"og:ttl\": \"600\",\n \"og:url\": \"https://www.reddit.com/r/ProductManagement/comments/163vekc/thoughts_on_metagpt/\"\n }\n ],\n \"cse_image\": [\n {\n \"src\": \"https://external-preview.redd.it/thoughts-on-metagpt-v0-VQP3cNl_-L2zHMe4QWMy1GTBsiLHKNj0lg-u_o_nZug.jpg?auto=webp&s=03900a2b49a801e7d769a0ae8d2ec7a05011c1fc\"\n }\n ]\n }\n },\n {\n \"kind\": \"customsearch#result\",\n \"title\": \"MetaGPT: Meta Programming for Multi-Agent Collaborative ...\",\n \"htmlTitle\": \"\\u003cb\\u003eMetaGPT\\u003c/b\\u003e: Meta Programming for Multi-Agent Collaborative ...\",\n \"link\": \"https://news.ycombinator.com/item?id=37076125\",\n \"displayLink\": \"news.ycombinator.com\",\n \"snippet\": \"You can use multiple agents, or split a lot of information across multiple requests to one agent. The result is the same. Some problems require a full ...\",\n \"htmlSnippet\": \"You can use multiple agents, or split a lot of information across multiple requests to one agent. The result is the same. Some problems require a full ...\",\n \"cacheId\": \"PvjWUfqo0GAJ\",\n \"formattedUrl\": \"https://news.ycombinator.com/item?id=37076125\",\n \"htmlFormattedUrl\": \"https://news.ycombinator.com/item?id=37076125\",\n \"pagemap\": {\n \"metatags\": [\n {\n \"referrer\": \"origin\",\n \"viewport\": \"width=device-width, initial-scale=1.0\"\n }\n ]\n }\n },\n {\n \"kind\": \"customsearch#result\",\n \"title\": \"MetaGPT: a Multi-Agent Framework to Automate Your Software ...\",\n \"htmlTitle\": \"\\u003cb\\u003eMetaGPT\\u003c/b\\u003e: a Multi-Agent Framework to Automate Your Software ...\",\n \"link\": \"https://medium.datadriveninvestor.com/metagpt-a-multi-agent-framework-to-automate-your-software-company-4b6ae747cc36\",\n \"displayLink\": \"medium.datadriveninvestor.com\",\n \"snippet\": \"MetaGPT is about to reach 10000 stars on Github. It's a Multi-Agent Framework that can behave as an engineer, product manager, architect, project managers.\",\n \"htmlSnippet\": \"\\u003cb\\u003eMetaGPT\\u003c/b\\u003e is about to reach 10000 stars on Github. It's a Multi-Agent Framework that can behave as an engineer, product manager, architect, project managers.\",\n \"cacheId\": \"qWqvRF7SoGsJ\",\n \"formattedUrl\": \"https://medium.datadriveninvestor.com/metagpt-a-multi-agent-framework-t...\",\n \"htmlFormattedUrl\": \"https://medium.datadriveninvestor.com/\\u003cb\\u003emetagpt\\u003c/b\\u003e-a-multi-agent-framework-t...\",\n \"pagemap\": {\n \"cse_thumbnail\": [\n {\n \"src\": \"https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcRKDyUf8JumEvosQ1ZmxQ1dGmOGIx1jd4bvnICexOb2jFmKZHKagMGoQ0xI\",\n \"width\": \"242\",\n \"height\": \"209\"\n }\n ],\n \"metatags\": [\n {\n \"og:image\": \"https://miro.medium.com/v2/da:true/resize:fit:1200/0*g2b2hbP8HykGIIN3\",\n \"twitter:app:url:iphone\": \"medium://p/4b6ae747cc36\",\n \"theme-color\": \"#000000\",\n \"article:published_time\": \"2023-09-05T05:20:30.732Z\",\n \"twitter:card\": \"summary_large_image\",\n \"og:site_name\": \"Medium\",\n \"al:android:package\": \"com.medium.reader\",\n \"twitter:label1\": \"Reading time\",\n \"twitter:tile:template:testing\": \"2\",\n \"twitter:app:id:iphone\": \"828256236\",\n \"title\": \"MetaGPT: a Multi-Agent Framework to Automate Your Software Company | by Peter Xing | DataDrivenInvestor\",\n \"al:ios:url\": \"medium://p/4b6ae747cc36\",\n \"og:description\": \"MetaGPT is about to reach 10,000 stars on Github. It’s a Multi-Agent Framework that can behave as an engineer, product manager, architect…\",\n \"twitter:creator\": \"@peterxing\",\n \"al:ios:app_store_id\": \"828256236\",\n \"twitter:data1\": \"2 min read\",\n \"twitter:site\": \"@DDInvestorHQ\",\n \"twitter:tile:info1:text\": \"Peter Xing\",\n \"twitter:tile:info1:icon\": \"Person\",\n \"og:type\": \"article\",\n \"twitter:title\": \"MetaGPT: a Multi-Agent Framework to Automate Your Software Company\",\n \"al:ios:app_name\": \"Medium\",\n \"twitter:cta\": \"Read on Medium\",\n \"author\": \"Peter Xing\",\n \"og:title\": \"MetaGPT: a Multi-Agent Framework to Automate Your Software Company\",\n \"al:web:url\": \"https://medium.datadriveninvestor.com/metagpt-a-multi-agent-framework-to-automate-your-software-company-4b6ae747cc36\",\n \"article:author\": \"https://medium.com/@peterxing\",\n \"twitter:tile:info2:text\": \"Sep 4, 2023\",\n \"twitter:image:src\": \"https://miro.medium.com/v2/da:true/resize:fit:1200/0*g2b2hbP8HykGIIN3\",\n \"al:android:url\": \"medium://p/4b6ae747cc36\",\n \"referrer\": \"unsafe-url\",\n \"fb:app_id\": \"542599432471018\",\n \"viewport\": \"width=device-width,minimum-scale=1,initial-scale=1,maximum-scale=1\",\n \"twitter:tile:info2:icon\": \"Calendar\",\n \"twitter:description\": \"MetaGPT is about to reach 10,000 stars on Github. It’s a Multi-Agent Framework that can behave as an engineer, product manager, architect…\",\n \"twitter:tile:image\": \"https://miro.medium.com/v2/da:true/resize:fit:1200/0*g2b2hbP8HykGIIN3\",\n \"og:url\": \"https://medium.datadriveninvestor.com/metagpt-a-multi-agent-framework-to-automate-your-software-company-4b6ae747cc36\",\n \"twitter:app:name:iphone\": \"Medium\",\n \"al:android:app_name\": \"Medium\"\n }\n ],\n \"cse_image\": [\n {\n \"src\": \"https://miro.medium.com/v2/da:true/resize:fit:1200/0*g2b2hbP8HykGIIN3\"\n }\n ]\n }\n },\n {\n \"kind\": \"customsearch#result\",\n \"title\": \"MetaGPT - ChatGPT\",\n \"htmlTitle\": \"MetaGPT - ChatGPT\",\n \"link\": \"https://chat.openai.com/g/g-gHceUPFhE-metagpt\",\n \"displayLink\": \"chat.openai.com\",\n \"snippet\": \"GPT. MetaGPT. Crafts specialized prompts for diverse GPT applications. By Ankit Pal. Sign up to chat. Requires ChatGPT Plus.\",\n \"htmlSnippet\": \"GPT. \\u003cb\\u003eMetaGPT\\u003c/b\\u003e. Crafts specialized prompts for diverse GPT applications. By Ankit Pal. Sign up to chat. Requires ChatGPT Plus.\",\n \"cacheId\": \"xhG1ItzjqPQJ\",\n \"formattedUrl\": \"https://chat.openai.com/g/g-gHceUPFhE-metagpt\",\n \"htmlFormattedUrl\": \"https://chat.openai.com/g/g-gHceUPFhE-\\u003cb\\u003emetagpt\\u003c/b\\u003e\",\n \"pagemap\": {\n \"metatags\": [\n {\n \"apple-itunes-app\": \"app-id=6448311069\",\n \"og:image\": \"https://files.oaiusercontent.com/file-pZO1spqW9XBbl6yhFEVA1xni?se=2123-10-18T23%3A51%3A44Z&sp=r&sv=2021-08-06&sr=b&rscc=max-age%3D31536000%2C%20immutable&rscd=attachment%3B%20filename%3Df77c7fb5-1ba1-487b-b610-19db286e62ab.png&sig=nPfDFuFwLLYoWGUotoX8wZ7mbXNRwg2wcIyXGpE19k0%3D\",\n \"og:type\": \"website\",\n \"og:image:width\": \"512\",\n \"og:site_name\": \"ChatGPT\",\n \"og:title\": \"ChatGPT - MetaGPT\",\n \"og:image:height\": \"512\",\n \"title\": \"ChatGPT - MetaGPT\",\n \"og:description\": \"Crafts specialized prompts for diverse GPT applications\",\n \"next-head-count\": \"21\",\n \"viewport\": \"width=device-width, initial-scale=1\",\n \"react-scroll-to-bottom:version\": \"4.2.0\",\n \"og:url\": \"/g/g-gHceUPFhE-metagpt\"\n }\n ],\n \"cse_image\": [\n {\n \"src\": \"https://files.oaiusercontent.com/file-pZO1spqW9XBbl6yhFEVA1xni?se=2123-10-18T23%3A51%3A44Z&sp=r&sv=2021-08-06&sr=b&rscc=max-age%3D31536000%2C%20immutable&rscd=attachment%3B%20filename%3Df77c7fb5-1ba1-487b-b610-19db286e62ab.png&sig=nPfDFuFwLLYoWGUotoX8wZ7mbXNRwg2wcIyXGpE19k0%3D\"\n }\n ]\n }\n },\n {\n \"kind\": \"customsearch#result\",\n \"title\": \"pip - Issue installing metagpt on Python 3.11 on Windows - Stack ...\",\n \"htmlTitle\": \"pip - Issue installing \\u003cb\\u003emetagpt\\u003c/b\\u003e on Python 3.11 on Windows - Stack ...\",\n \"link\": \"https://stackoverflow.com/questions/76871577/issue-installing-metagpt-on-python-3-11-on-windows\",\n \"displayLink\": \"stackoverflow.com\",\n \"snippet\": \"Aug 9, 2023 ... 1 Answer 1 · try to delete the file 'metagpt-0.1-py3.11.egg', and try again. · if still not work , you can use pip install -r requirements.txt ...\",\n \"htmlSnippet\": \"Aug 9, 2023 \\u003cb\\u003e...\\u003c/b\\u003e 1 Answer 1 · try to delete the file '\\u003cb\\u003emetagpt\\u003c/b\\u003e-0.1-py3.11.egg', and try again. · if still not work , you can use pip install -r requirements.txt ...\",\n \"cacheId\": \"rE7h8ENZAfsJ\",\n \"formattedUrl\": \"https://stackoverflow.com/.../issue-installing-metagpt-on-python-3-11-on-w...\",\n \"htmlFormattedUrl\": \"https://stackoverflow.com/.../issue-installing-\\u003cb\\u003emetagpt\\u003c/b\\u003e-on-python-3-11-on-w...\",\n \"pagemap\": {\n \"cse_thumbnail\": [\n {\n \"src\": \"https://encrypted-tbn2.gstatic.com/images?q=tbn:ANd9GcQYl7zuT3cw_BBRAyhdQEbQuBgqdNHXKHIYKL8S8ly8x9L_XA9sdwSmiHs\",\n \"width\": \"225\",\n \"height\": \"225\"\n }\n ],\n \"qapage\": [\n {\n \"image\": \"https://cdn.sstatic.net/Sites/stackoverflow/Img/apple-touch-icon@2.png?v=73d79a89bded\",\n \"primaryimageofpage\": \"https://cdn.sstatic.net/Sites/stackoverflow/Img/apple-touch-icon@2.png?v=73d79a89bded\",\n \"name\": \"Issue installing metagpt on Python 3.11 on Windows\",\n \"description\": \"I receives this error when trying to install metagpt packages: [WinError 32] The process cannot access the file because it is being used by another process: 'c:\\\\\\\\users\\\\\\\\anthony phan\\\\\\\\appdata\\\\\\\\local\\\\\\\\\"\n }\n ],\n \"question\": [\n {\n \"image\": \"https://cdn.sstatic.net/Sites/stackoverflow/Img/apple-touch-icon.png?v=c78bd457575a\",\n \"upvotecount\": \"1\",\n \"answercount\": \"1\",\n \"name\": \"Issue installing metagpt on Python 3.11 on Windows\",\n \"datecreated\": \"2023-08-09T22:10:59\",\n \"text\": \"I receives this error when trying to install metagpt packages: [WinError 32] The process cannot access the file because it is being used by another process: 'c:\\\\\\\\users\\\\\\\\anthony phan\\\\\\\\appdata\\\\\\\\local...\",\n \"url\": \"Share\"\n }\n ],\n \"answer\": [\n {\n \"upvotecount\": \"0\",\n \"text\": \"try to delete the file 'metagpt-0.1-py3.11.egg', and try again. if still not work , you can use pip install -r requirements.txt instead of python setup.py\",\n \"datecreated\": \"2023-08-30T02:04:27\",\n \"url\": \"Share\"\n }\n ],\n \"person\": [\n {\n \"name\": \"Anthony Q Phan\"\n },\n {\n \"name\": \"D yesfir\"\n }\n ],\n \"metatags\": [\n {\n \"og:image\": \"https://cdn.sstatic.net/Sites/stackoverflow/Img/apple-touch-icon@2.png?v=73d79a89bded\",\n \"og:type\": \"website\",\n \"twitter:card\": \"summary\",\n \"twitter:title\": \"Issue installing metagpt on Python 3.11 on Windows\",\n \"og:site_name\": \"Stack Overflow\",\n \"twitter:domain\": \"stackoverflow.com\",\n \"viewport\": \"width=device-width, height=device-height, initial-scale=1.0, minimum-scale=1.0\",\n \"twitter:description\": \"I receives this error when trying to install metagpt packages:\\n[WinError 32] The process cannot access the file because it is being used by another process: 'c:\\\\\\\\users\\\\\\\\anthony phan\\\\\\\\appdata\\\\\\\\local\\\\\\\\\",\n \"og:url\": \"https://stackoverflow.com/questions/76871577/issue-installing-metagpt-on-python-3-11-on-windows\"\n }\n ],\n \"cse_image\": [\n {\n \"src\": \"https://cdn.sstatic.net/Sites/stackoverflow/Img/apple-touch-icon@2.png?v=73d79a89bded\"\n }\n ]\n }\n }\n ]\n}\n", + "httplib2-GET-https://customsearch.googleapis.com/customsearch/v1-{\"params\": {\"q\": \"metagpt\", \"num\": \"6\", \"cx\": \"mock-google-cse\", \"key\": \"mock-google-key\", \"alt\": \"json\"}}": "{\n \"kind\": \"customsearch#search\",\n \"url\": {\n \"type\": \"application/json\",\n \"template\": \"https://www.googleapis.com/customsearch/v1?q={searchTerms}&num={count?}&start={startIndex?}&lr={language?}&safe={safe?}&cx={cx?}&sort={sort?}&filter={filter?}&gl={gl?}&cr={cr?}&googlehost={googleHost?}&c2coff={disableCnTwTranslation?}&hq={hq?}&hl={hl?}&siteSearch={siteSearch?}&siteSearchFilter={siteSearchFilter?}&exactTerms={exactTerms?}&excludeTerms={excludeTerms?}&linkSite={linkSite?}&orTerms={orTerms?}&dateRestrict={dateRestrict?}&lowRange={lowRange?}&highRange={highRange?}&searchType={searchType}&fileType={fileType?}&rights={rights?}&imgSize={imgSize?}&imgType={imgType?}&imgColorType={imgColorType?}&imgDominantColor={imgDominantColor?}&alt=json\"\n },\n \"queries\": {\n \"request\": [\n {\n \"title\": \"Google Custom Search - metagpt\",\n \"totalResults\": \"85300\",\n \"searchTerms\": \"metagpt\",\n \"count\": 6,\n \"startIndex\": 1,\n \"inputEncoding\": \"utf8\",\n \"outputEncoding\": \"utf8\",\n \"safe\": \"off\",\n \"cx\": \"mock-google-cse\"\n }\n ],\n \"nextPage\": [\n {\n \"title\": \"Google Custom Search - metagpt\",\n \"totalResults\": \"85300\",\n \"searchTerms\": \"metagpt\",\n \"count\": 6,\n \"startIndex\": 7,\n \"inputEncoding\": \"utf8\",\n \"outputEncoding\": \"utf8\",\n \"safe\": \"off\",\n \"cx\": \"mock-google-cse\"\n }\n ]\n },\n \"context\": {\n \"title\": \"metagpt1\"\n },\n \"searchInformation\": {\n \"searchTime\": 0.193417,\n \"formattedSearchTime\": \"0.19\",\n \"totalResults\": \"85300\",\n \"formattedTotalResults\": \"85,300\"\n },\n \"items\": [\n {\n \"kind\": \"customsearch#result\",\n \"title\": \"geekan/MetaGPT: The Multi-Agent Framework: Given one ... - GitHub\",\n \"htmlTitle\": \"geekan/MetaGPT: The Multi-Agent Framework: Given one ... - GitHub\",\n \"link\": \"https://github.com/geekan/MetaGPT\",\n \"displayLink\": \"github.com\",\n \"snippet\": \"The Multi-Agent Framework: Given one line Requirement, return PRD, Design, Tasks, Repo - GitHub - geekan/MetaGPT: The Multi-Agent Framework: Given one ...\",\n \"htmlSnippet\": \"The Multi-Agent Framework: Given one line Requirement, return PRD, Design, Tasks, Repo - GitHub - geekan/\\u003cb\\u003eMetaGPT\\u003c/b\\u003e: The Multi-Agent Framework: Given one ...\",\n \"cacheId\": \"gsshb0APPNgJ\",\n \"formattedUrl\": \"https://github.com/geekan/MetaGPT\",\n \"htmlFormattedUrl\": \"https://github.com/geekan/\\u003cb\\u003eMetaGPT\\u003c/b\\u003e\",\n \"pagemap\": {\n \"cse_thumbnail\": [\n {\n \"src\": \"https://encrypted-tbn1.gstatic.com/images?q=tbn:ANd9GcRuD8YUvRcltmdoxKyuIbt8UZhg3LE5mwNX7KPXDB15YIJRKdT2m5JiweuS\",\n \"width\": \"318\",\n \"height\": \"159\"\n }\n ],\n \"softwaresourcecode\": [\n {\n \"author\": \"geekan\",\n \"name\": \"MetaGPT\",\n \"text\": \"MetaGPT: The Multi-Agent Framework Assign different roles to GPTs to form a collaborative software entity for complex tasks. MetaGPT takes a one line requirement as input and outputs user stories...\"\n }\n ],\n \"metatags\": [\n {\n \"octolytics-url\": \"https://collector.github.com/github/collect\",\n \"apple-itunes-app\": \"app-id=1477376905, app-argument=https://github.com/geekan/MetaGPT\",\n \"og:image\": \"https://opengraph.githubassets.com/6178eb2aa6711c676eafc956e52345e71225c58cd0a666b54871171e847c0905/geekan/MetaGPT\",\n \"twitter:card\": \"summary_large_image\",\n \"og:image:width\": \"1200\",\n \"theme-color\": \"#1e2327\",\n \"og:site_name\": \"GitHub\",\n \"hovercard-subject-tag\": \"repository:660551251\",\n \"turbo-body-classes\": \"logged-out env-production page-responsive\",\n \"html-safe-nonce\": \"a6964edcf12c9ea83de0bf16db8105c60333287597018548250a0fa92b93d3f9\",\n \"expected-hostname\": \"github.com\",\n \"og:description\": \"🌟 The Multi-Agent Framework: Given one line Requirement, return PRD, Design, Tasks, Repo - GitHub - geekan/MetaGPT: 🌟 The Multi-Agent Framework: Given one line Requirement, return PRD, Design, Task...\",\n \"browser-errors-url\": \"https://api.github.com/_private/browser/errors\",\n \"octolytics-dimension-user_login\": \"geekan\",\n \"hostname\": \"github.com\",\n \"twitter:site\": \"@github\",\n \"browser-stats-url\": \"https://api.github.com/_private/browser/stats\",\n \"route-pattern\": \"/:user_id/:repository\",\n \"visitor-payload\": \"eyJyZWZlcnJlciI6IiIsInJlcXVlc3RfaWQiOiJBNjIxOjk2OUU6OTZERjlEQzpEM0UxM0RFOjY1QTNBQjU0IiwidmlzaXRvcl9pZCI6IjU2ODA2NDI2MDQ0ODY5MzA3NyIsInJlZ2lvbl9lZGdlIjoiaWFkIiwicmVnaW9uX3JlbmRlciI6ImlhZCJ9\",\n \"github-keyboard-shortcuts\": \"repository\",\n \"octolytics-dimension-repository_id\": \"660551251\",\n \"octolytics-dimension-repository_network_root_nwo\": \"geekan/MetaGPT\",\n \"twitter:title\": \"GitHub - geekan/MetaGPT: 🌟 The Multi-Agent Framework: Given one line Requirement, return PRD, Design, Tasks, Repo\",\n \"og:image:alt\": \"🌟 The Multi-Agent Framework: Given one line Requirement, return PRD, Design, Tasks, Repo - GitHub - geekan/MetaGPT: 🌟 The Multi-Agent Framework: Given one line Requirement, return PRD, Design, Task...\",\n \"og:type\": \"object\",\n \"optimizely-datafile\": \"{\\\"accountId\\\": \\\"16737760170\\\", \\\"projectId\\\": \\\"16737760170\\\", \\\"revision\\\": \\\"23\\\", \\\"attributes\\\": [{\\\"id\\\": \\\"16822470375\\\", \\\"key\\\": \\\"user_id\\\"}, {\\\"id\\\": \\\"17143601254\\\", \\\"key\\\": \\\"spammy\\\"}, {\\\"id\\\": \\\"18175660309\\\", \\\"key\\\": \\\"organization_plan\\\"}, {\\\"id\\\": \\\"18813001570\\\", \\\"key\\\": \\\"is_logged_in\\\"}, {\\\"id\\\": \\\"19073851829\\\", \\\"key\\\": \\\"geo\\\"}, {\\\"id\\\": \\\"20175462351\\\", \\\"key\\\": \\\"requestedCurrency\\\"}, {\\\"id\\\": \\\"20785470195\\\", \\\"key\\\": \\\"country_code\\\"}, {\\\"id\\\": \\\"21656311196\\\", \\\"key\\\": \\\"opened_downgrade_dialog\\\"}], \\\"audiences\\\": [{\\\"id\\\": \\\"$opt_dummy_audience\\\", \\\"name\\\": \\\"Optimizely-Generated Audience for Backwards Compatibility\\\", \\\"conditions\\\": \\\"[\\\\\\\"or\\\\\\\", {\\\\\\\"match\\\\\\\": \\\\\\\"exact\\\\\\\", \\\\\\\"name\\\\\\\": \\\\\\\"$opt_dummy_attribute\\\\\\\", \\\\\\\"type\\\\\\\": \\\\\\\"custom_attribute\\\\\\\", \\\\\\\"value\\\\\\\": \\\\\\\"$opt_dummy_value\\\\\\\"}]\\\"}], \\\"version\\\": \\\"4\\\", \\\"events\\\": [{\\\"id\\\": \\\"18188530140\\\", \\\"experimentIds\\\": [], \\\"key\\\": \\\"test_event\\\"}], \\\"integrations\\\": [], \\\"anonymizeIP\\\": true, \\\"botFiltering\\\": false, \\\"typedAudiences\\\": [], \\\"variables\\\": [], \\\"environmentKey\\\": \\\"production\\\", \\\"sdkKey\\\": \\\"UpVyJZaLVEGwJPQWf5pAD\\\", \\\"featureFlags\\\": [], \\\"rollouts\\\": [],\",\n \"og:title\": \"GitHub - geekan/MetaGPT: 🌟 The Multi-Agent Framework: Given one line Requirement, return PRD, Design, Tasks, Repo\",\n \"visitor-hmac\": \"471691c5f7e3204061bb09c630c440708f5c08a2824d60b0f28eea873d474cd7\",\n \"og:image:height\": \"600\",\n \"turbo-cache-control\": \"no-preview\",\n \"request-id\": \"A621:969E:96DF9DC:D3E13DE:65A3AB54\",\n \"analytics-location\": \"/\\u003cuser-name\\u003e/\\u003crepo-name\\u003e\",\n \"color-scheme\": \"light dark\",\n \"octolytics-dimension-repository_is_fork\": \"false\",\n \"go-import\": \"github.com/geekan/MetaGPT git https://github.com/geekan/MetaGPT.git\",\n \"browser-optimizely-client-errors-url\": \"https://api.github.com/_private/browser/optimizely_client/errors\",\n \"twitter:image:src\": \"https://opengraph.githubassets.com/6178eb2aa6711c676eafc956e52345e71225c58cd0a666b54871171e847c0905/geekan/MetaGPT\",\n \"octolytics-dimension-user_id\": \"2707039\",\n \"octolytics-dimension-repository_public\": \"true\",\n \"fb:app_id\": \"1401488693436528\",\n \"octolytics-dimension-repository_network_root_id\": \"660551251\",\n \"octolytics-dimension-repository_nwo\": \"geekan/MetaGPT\",\n \"viewport\": \"width=device-width\",\n \"twitter:description\": \"🌟 The Multi-Agent Framework: Given one line Requirement, return PRD, Design, Tasks, Repo - GitHub - geekan/MetaGPT: 🌟 The Multi-Agent Framework: Given one line Requirement, return PRD, Design, Task...\",\n \"current-catalog-service-hash\": \"82c569b93da5c18ed649ebd4c2c79437db4611a6a1373e805a3cb001c64130b7\",\n \"og:url\": \"https://github.com/geekan/MetaGPT\"\n }\n ],\n \"cse_image\": [\n {\n \"src\": \"https://opengraph.githubassets.com/6178eb2aa6711c676eafc956e52345e71225c58cd0a666b54871171e847c0905/geekan/MetaGPT\"\n }\n ]\n }\n },\n {\n \"kind\": \"customsearch#result\",\n \"title\": \"[2308.00352] MetaGPT: Meta Programming for A Multi-Agent ...\",\n \"htmlTitle\": \"[2308.00352] \\u003cb\\u003eMetaGPT\\u003c/b\\u003e: Meta Programming for A Multi-Agent ...\",\n \"link\": \"https://arxiv.org/abs/2308.00352\",\n \"displayLink\": \"arxiv.org\",\n \"snippet\": \"Aug 1, 2023 ... Computer Science \\u003e Artificial Intelligence · Title:MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework · Bibliographic and ...\",\n \"htmlSnippet\": \"Aug 1, 2023 \\u003cb\\u003e...\\u003c/b\\u003e Computer Science > Artificial Intelligence · Title:\\u003cb\\u003eMetaGPT\\u003c/b\\u003e: Meta Programming for A Multi-Agent Collaborative Framework · Bibliographic and ...\",\n \"cacheId\": \"8_tddNY0jEYJ\",\n \"formattedUrl\": \"https://arxiv.org/abs/2308.00352\",\n \"htmlFormattedUrl\": \"https://arxiv.org/abs/2308.00352\",\n \"pagemap\": {\n \"cse_thumbnail\": [\n {\n \"src\": \"https://encrypted-tbn3.gstatic.com/images?q=tbn:ANd9GcStsc5IszP_UC7vkymrk7PhjHGOFQhTh862xtJcQkxDem2IteJQXpob6_Vb\",\n \"width\": \"336\",\n \"height\": \"150\"\n }\n ],\n \"metatags\": [\n {\n \"og:image\": \"/static/browse/0.3.4/images/arxiv-logo-fb.png\",\n \"theme-color\": \"#ffffff\",\n \"og:image:width\": \"1200\",\n \"twitter:card\": \"summary\",\n \"citation_title\": \"MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework\",\n \"og:site_name\": \"arXiv.org\",\n \"citation_date\": \"2023/08/01\",\n \"og:description\": \"Remarkable progress has been made on automated problem solving through societies of agents based on large language models (LLMs). Existing LLM-based multi-agent systems can already solve simple dialogue tasks. Solutions to more complex tasks, however, are complicated through logic inconsistencies due to cascading hallucinations caused by naively chaining LLMs. Here we introduce MetaGPT, an innovative meta-programming framework incorporating efficient human workflows into LLM-based multi-agent collaborations. MetaGPT encodes Standardized Operating Procedures (SOPs) into prompt sequences for more streamlined workflows, thus allowing agents with human-like domain expertise to verify intermediate results and reduce errors. MetaGPT utilizes an assembly line paradigm to assign diverse roles to various agents, efficiently breaking down complex tasks into subtasks involving many agents working together. On collaborative software engineering benchmarks, MetaGPT generates more coherent solutions than previous chat-base\",\n \"og:image:secure_url\": \"/static/browse/0.3.4/images/arxiv-logo-fb.png\",\n \"twitter:image\": \"https://static.arxiv.org/icons/twitter/arxiv-logo-twitter-square.png\",\n \"citation_arxiv_id\": \"2308.00352\",\n \"citation_online_date\": \"2023/11/06\",\n \"twitter:image:alt\": \"arXiv logo\",\n \"twitter:site\": \"@arxiv\",\n \"citation_pdf_url\": \"http://arxiv.org/pdf/2308.00352.pdf\",\n \"msapplication-tilecolor\": \"#da532c\",\n \"og:type\": \"website\",\n \"og:image:alt\": \"arXiv logo\",\n \"twitter:title\": \"MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework\",\n \"og:title\": \"MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework\",\n \"citation_abstract\": \"Remarkable progress has been made on automated problem solving through societies of agents based on large language models (LLMs). Existing LLM-based multi-agent systems can already solve simple dialogue tasks. Solutions to more complex tasks, however, are complicated through logic inconsistencies due to cascading hallucinations caused by naively chaining LLMs. Here we introduce MetaGPT, an innovative meta-programming framework incorporating efficient human workflows into LLM-based multi-agent collaborations. MetaGPT encodes Standardized Operating Procedures (SOPs) into prompt sequences for more streamlined workflows, thus allowing agents with human-like domain expertise to verify intermediate results and reduce errors. MetaGPT utilizes an assembly line paradigm to assign diverse roles to various agents, efficiently breaking down complex tasks into subtasks involving many agents working together. On collaborative software engineering benchmarks, MetaGPT generates more coherent solutions than previous chat-base\",\n \"og:image:height\": \"700\",\n \"citation_author\": \"Hong, Sirui\",\n \"viewport\": \"width=device-width, initial-scale=1\",\n \"twitter:description\": \"Remarkable progress has been made on automated problem solving through societies of agents based on large language models (LLMs). Existing LLM-based multi-agent systems can already solve simple...\",\n \"og:url\": \"https://arxiv.org/abs/2308.00352v5\"\n }\n ],\n \"cse_image\": [\n {\n \"src\": \"https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-one-color-white.svg\"\n }\n ]\n }\n },\n {\n \"kind\": \"customsearch#result\",\n \"title\": \"MetaGPT: Complete Guide to the Best AI Agent Available Right Now ...\",\n \"htmlTitle\": \"\\u003cb\\u003eMetaGPT\\u003c/b\\u003e: Complete Guide to the Best AI Agent Available Right Now ...\",\n \"link\": \"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\n \"displayLink\": \"www.unite.ai\",\n \"snippet\": \"Sep 11, 2023 ... The beauty of MetaGPT lies in its structuring. It capitalizes on meta-programming techniques to manipulate, analyze, and transform code in real- ...\",\n \"htmlSnippet\": \"Sep 11, 2023 \\u003cb\\u003e...\\u003c/b\\u003e The beauty of \\u003cb\\u003eMetaGPT\\u003c/b\\u003e lies in its structuring. It capitalizes on meta-programming techniques to manipulate, analyze, and transform code in real- ...\",\n \"cacheId\": \"qkZULzxVHNAJ\",\n \"formattedUrl\": \"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-...\",\n \"htmlFormattedUrl\": \"https://www.unite.ai/\\u003cb\\u003emetagpt\\u003c/b\\u003e-complete-guide-to-the-best-ai-agent-available-...\",\n \"pagemap\": {\n \"cse_thumbnail\": [\n {\n \"src\": \"https://encrypted-tbn2.gstatic.com/images?q=tbn:ANd9GcSVwf1WWLtVpqJCZ1E_t7TrpSZ7nrwsCUWar6x9YzlOsX1aSH7EGHbkIlY\",\n \"width\": \"290\",\n \"height\": \"174\"\n }\n ],\n \"imageobject\": [\n {\n \"width\": \"1000\",\n \"url\": \"https://www.unite.ai/wp-content/uploads/2023/09/Heisenbergforlife_Center_the_scene_in_zoomed_scope_around_a_hum_7f069632-eda5-4edd-858b-cb44fec82929-1000x600.png\",\n \"height\": \"600\"\n },\n {\n \"url\": \"https://www.unite.ai/wp-content/uploads/2021/03/logoUNITE230X30BLACK-1.svg\"\n }\n ],\n \"person\": [\n {\n \"name\": \"Aayush Mittal\"\n }\n ],\n \"organization\": [\n {\n \"name\": \"Unite.AI\"\n }\n ],\n \"metatags\": [\n {\n \"og:image\": \"https://www.unite.ai/wp-content/uploads/2023/09/Heisenbergforlife_Center_the_scene_in_zoomed_scope_around_a_hum_7f069632-eda5-4edd-858b-cb44fec82929-1000x600.png\",\n \"twitter:card\": \"summary\",\n \"article:published_time\": \"2023-09-11T18:03:48+00:00\",\n \"og:image:width\": \"1121\",\n \"og:site_name\": \"Unite.AI\",\n \"twitter:url\": \"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\n \"twitter:label1\": \"Written by\",\n \"twitter:label2\": \"Est. reading time\",\n \"og:image:type\": \"image/png\",\n \"msapplication-tileimage\": \"https://www.unite.ai/wp-content/uploads/2023/09/Heisenbergforlife_Center_the_scene_in_zoomed_scope_around_a_hum_7f069632-eda5-4edd-858b-cb44fec82929.png\",\n \"og:description\": \"Discover why MetaGPT outperforms AutoGPT, BabyAgi, and other AI agents in complex coding tasks. Our in-depth article guides you through the setup process and provides illustrative examples. Build GPT-powered microapps with a single line of prompt\",\n \"twitter:creator\": \"@UniteAI\",\n \"twitter:image\": \"https://www.unite.ai/wp-content/uploads/2023/09/Heisenbergforlife_Center_the_scene_in_zoomed_scope_around_a_hum_7f069632-eda5-4edd-858b-cb44fec82929-1000x600.png\",\n \"article:publisher\": \"https://www.facebook.com/uniteai\",\n \"twitter:data1\": \"Aayush Mittal\",\n \"og:image:secure_url\": \"https://www.unite.ai/wp-content/uploads/2023/09/Heisenbergforlife_Center_the_scene_in_zoomed_scope_around_a_hum_7f069632-eda5-4edd-858b-cb44fec82929.png\",\n \"twitter:data2\": \"9 minutes\",\n \"twitter:site\": \"@UniteAI\",\n \"og:video:type\": \"video/mp4\",\n \"uri-translation\": \"on\",\n \"og:type\": \"article\",\n \"twitter:title\": \"MetaGPT: Complete Guide to the Best AI Agent Available Right Now\",\n \"og:image:alt\": \"MetaGPBassed Illustration of human and machine collaborationT\",\n \"author\": \"Aayush Mittal\",\n \"og:title\": \"MetaGPT: Complete Guide to the Best AI Agent Available Right Now\",\n \"og:image:height\": \"628\",\n \"og:updated_time\": \"2023-09-11T14:03:48-04:00\",\n \"article:tag\": \"AI AGENTS\",\n \"og:video\": \"https://www.unite.ai/wp-content/uploads/2023/09/ezgif.com-optimize-online-video-cutter.com_.mp4\",\n \"viewport\": \"width=device-width,initial-scale=1.0,user-scalable=yes\",\n \"og:locale\": \"en_US\",\n \"og:rich_attachment\": \"1\",\n \"og:url\": \"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\"\n }\n ],\n \"cse_image\": [\n {\n \"src\": \"https://www.unite.ai/wp-content/uploads/2023/09/Heisenbergforlife_Center_the_scene_in_zoomed_scope_around_a_hum_7f069632-eda5-4edd-858b-cb44fec82929-1000x600.png\"\n }\n ],\n \"blogposting\": [\n {\n \"image\": \"https://www.unite.ai/wp-content/uploads/2023/09/Heisenbergforlife_Center_the_scene_in_zoomed_scope_around_a_hum_7f069632-eda5-4edd-858b-cb44fec82929.png\",\n \"datemodified\": \"2023-09-11T18:03:48+00:00\",\n \"author\": \"Aayush Mittal\",\n \"name\": \"MetaGPT: Complete Guide to the Best AI Agent Available Right Now\",\n \"description\": \"With Large Language Models (LLMs) like ChatGPT, OpenAI has witnessed a surge in enterprise and user adoption, currently raking in around $80 million in monthly revenue. According to a recent...\",\n \"headline\": \"MetaGPT: Complete Guide to the Best AI Agent Available Right Now\",\n \"datepublished\": \"2023-09-11\"\n }\n ],\n \"newsarticle\": [\n {\n \"datemodified\": \"2023-09-11\",\n \"keywords\": \"AI AGENTSAutoGPTDockergenerative aiLLMMetaGPTnlpPROMPT ENGINEERINGpython\",\n \"headline\": \"MetaGPT: Complete Guide to the Best AI Agent Available Right Now\",\n \"datepublished\": \"2023-09-11\"\n }\n ]\n }\n },\n {\n \"kind\": \"customsearch#result\",\n \"title\": \"Thoughts on MetaGPT : r/ProductManagement\",\n \"htmlTitle\": \"Thoughts on \\u003cb\\u003eMetaGPT\\u003c/b\\u003e : r/ProductManagement\",\n \"link\": \"https://www.reddit.com/r/ProductManagement/comments/163vekc/thoughts_on_metagpt/\",\n \"displayLink\": \"www.reddit.com\",\n \"snippet\": \"Aug 28, 2023 ... Thoughts on MetaGPT. YT shorts - a quick summaryExplainer YT video - maynot be the best, but beginner friendly. ... PS: feel free to link more ...\",\n \"htmlSnippet\": \"Aug 28, 2023 \\u003cb\\u003e...\\u003c/b\\u003e Thoughts on \\u003cb\\u003eMetaGPT\\u003c/b\\u003e. YT shorts - a quick summaryExplainer YT video - maynot be the best, but beginner friendly. ... PS: feel free to link more ...\",\n \"cacheId\": \"fDkEZ_skdhcJ\",\n \"formattedUrl\": \"https://www.reddit.com/r/ProductManagement/.../thoughts_on_metagpt/\",\n \"htmlFormattedUrl\": \"https://www.reddit.com/r/ProductManagement/.../thoughts_on_\\u003cb\\u003emetagpt\\u003c/b\\u003e/\",\n \"pagemap\": {\n \"cse_thumbnail\": [\n {\n \"src\": \"https://encrypted-tbn2.gstatic.com/images?q=tbn:ANd9GcSnWudLgGG_2ao_7EWw3EW58JBUQkJ1m4LOHzyiajVHq10p0_TNAeCRlik\",\n \"width\": \"259\",\n \"height\": \"194\"\n }\n ],\n \"metatags\": [\n {\n \"og:image\": \"https://share.redd.it/preview/post/163vekc\",\n \"theme-color\": \"#000000\",\n \"og:image:width\": \"1200\",\n \"og:type\": \"website\",\n \"og:image:alt\": \"An image containing a preview of the post\",\n \"twitter:card\": \"summary_large_image\",\n \"twitter:title\": \"r/ProductManagement on Reddit: Thoughts on MetaGPT\",\n \"og:site_name\": \"Reddit\",\n \"og:title\": \"r/ProductManagement on Reddit: Thoughts on MetaGPT\",\n \"og:image:height\": \"630\",\n \"msapplication-navbutton-color\": \"#000000\",\n \"og:description\": \"Posted by u/CheraCholan - No votes and 4 comments\",\n \"twitter:image\": \"https://share.redd.it/preview/post/163vekc\",\n \"apple-mobile-web-app-status-bar-style\": \"black\",\n \"twitter:site\": \"@reddit\",\n \"viewport\": \"width=device-width, initial-scale=1, viewport-fit=cover\",\n \"apple-mobile-web-app-capable\": \"yes\",\n \"og:ttl\": \"600\",\n \"og:url\": \"https://www.reddit.com/r/ProductManagement/comments/163vekc/thoughts_on_metagpt/\"\n }\n ],\n \"cse_image\": [\n {\n \"src\": \"https://external-preview.redd.it/thoughts-on-metagpt-v0-VQP3cNl_-L2zHMe4QWMy1GTBsiLHKNj0lg-u_o_nZug.jpg?auto=webp&s=03900a2b49a801e7d769a0ae8d2ec7a05011c1fc\"\n }\n ]\n }\n },\n {\n \"kind\": \"customsearch#result\",\n \"title\": \"MetaGPT: Meta Programming for Multi-Agent Collaborative ...\",\n \"htmlTitle\": \"\\u003cb\\u003eMetaGPT\\u003c/b\\u003e: Meta Programming for Multi-Agent Collaborative ...\",\n \"link\": \"https://news.ycombinator.com/item?id=37076125\",\n \"displayLink\": \"news.ycombinator.com\",\n \"snippet\": \"You can use multiple agents, or split a lot of information across multiple requests to one agent. The result is the same. Some problems require a full ...\",\n \"htmlSnippet\": \"You can use multiple agents, or split a lot of information across multiple requests to one agent. The result is the same. Some problems require a full ...\",\n \"cacheId\": \"PvjWUfqo0GAJ\",\n \"formattedUrl\": \"https://news.ycombinator.com/item?id=37076125\",\n \"htmlFormattedUrl\": \"https://news.ycombinator.com/item?id=37076125\",\n \"pagemap\": {\n \"metatags\": [\n {\n \"referrer\": \"origin\",\n \"viewport\": \"width=device-width, initial-scale=1.0\"\n }\n ]\n }\n },\n {\n \"kind\": \"customsearch#result\",\n \"title\": \"MetaGPT: a Multi-Agent Framework to Automate Your Software ...\",\n \"htmlTitle\": \"\\u003cb\\u003eMetaGPT\\u003c/b\\u003e: a Multi-Agent Framework to Automate Your Software ...\",\n \"link\": \"https://medium.datadriveninvestor.com/metagpt-a-multi-agent-framework-to-automate-your-software-company-4b6ae747cc36\",\n \"displayLink\": \"medium.datadriveninvestor.com\",\n \"snippet\": \"MetaGPT is about to reach 10000 stars on Github. It's a Multi-Agent Framework that can behave as an engineer, product manager, architect, project managers.\",\n \"htmlSnippet\": \"\\u003cb\\u003eMetaGPT\\u003c/b\\u003e is about to reach 10000 stars on Github. It's a Multi-Agent Framework that can behave as an engineer, product manager, architect, project managers.\",\n \"cacheId\": \"qWqvRF7SoGsJ\",\n \"formattedUrl\": \"https://medium.datadriveninvestor.com/metagpt-a-multi-agent-framework-t...\",\n \"htmlFormattedUrl\": \"https://medium.datadriveninvestor.com/\\u003cb\\u003emetagpt\\u003c/b\\u003e-a-multi-agent-framework-t...\",\n \"pagemap\": {\n \"cse_thumbnail\": [\n {\n \"src\": \"https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcRKDyUf8JumEvosQ1ZmxQ1dGmOGIx1jd4bvnICexOb2jFmKZHKagMGoQ0xI\",\n \"width\": \"242\",\n \"height\": \"209\"\n }\n ],\n \"metatags\": [\n {\n \"og:image\": \"https://miro.medium.com/v2/da:true/resize:fit:1200/0*g2b2hbP8HykGIIN3\",\n \"twitter:app:url:iphone\": \"medium://p/4b6ae747cc36\",\n \"theme-color\": \"#000000\",\n \"article:published_time\": \"2023-09-05T05:20:30.732Z\",\n \"twitter:card\": \"summary_large_image\",\n \"og:site_name\": \"Medium\",\n \"al:android:package\": \"com.medium.reader\",\n \"twitter:label1\": \"Reading time\",\n \"twitter:tile:template:testing\": \"2\",\n \"twitter:app:id:iphone\": \"828256236\",\n \"title\": \"MetaGPT: a Multi-Agent Framework to Automate Your Software Company | by Peter Xing | DataDrivenInvestor\",\n \"al:ios:url\": \"medium://p/4b6ae747cc36\",\n \"og:description\": \"MetaGPT is about to reach 10,000 stars on Github. It’s a Multi-Agent Framework that can behave as an engineer, product manager, architect…\",\n \"twitter:creator\": \"@peterxing\",\n \"al:ios:app_store_id\": \"828256236\",\n \"twitter:data1\": \"2 min read\",\n \"twitter:site\": \"@DDInvestorHQ\",\n \"twitter:tile:info1:text\": \"Peter Xing\",\n \"twitter:tile:info1:icon\": \"Person\",\n \"og:type\": \"article\",\n \"twitter:title\": \"MetaGPT: a Multi-Agent Framework to Automate Your Software Company\",\n \"al:ios:app_name\": \"Medium\",\n \"twitter:cta\": \"Read on Medium\",\n \"author\": \"Peter Xing\",\n \"og:title\": \"MetaGPT: a Multi-Agent Framework to Automate Your Software Company\",\n \"al:web:url\": \"https://medium.datadriveninvestor.com/metagpt-a-multi-agent-framework-to-automate-your-software-company-4b6ae747cc36\",\n \"article:author\": \"https://medium.com/@peterxing\",\n \"twitter:tile:info2:text\": \"Sep 4, 2023\",\n \"twitter:image:src\": \"https://miro.medium.com/v2/da:true/resize:fit:1200/0*g2b2hbP8HykGIIN3\",\n \"al:android:url\": \"medium://p/4b6ae747cc36\",\n \"referrer\": \"unsafe-url\",\n \"fb:app_id\": \"542599432471018\",\n \"viewport\": \"width=device-width,minimum-scale=1,initial-scale=1,maximum-scale=1\",\n \"twitter:tile:info2:icon\": \"Calendar\",\n \"twitter:description\": \"MetaGPT is about to reach 10,000 stars on Github. It’s a Multi-Agent Framework that can behave as an engineer, product manager, architect…\",\n \"twitter:tile:image\": \"https://miro.medium.com/v2/da:true/resize:fit:1200/0*g2b2hbP8HykGIIN3\",\n \"og:url\": \"https://medium.datadriveninvestor.com/metagpt-a-multi-agent-framework-to-automate-your-software-company-4b6ae747cc36\",\n \"twitter:app:name:iphone\": \"Medium\",\n \"al:android:app_name\": \"Medium\"\n }\n ],\n \"cse_image\": [\n {\n \"src\": \"https://miro.medium.com/v2/da:true/resize:fit:1200/0*g2b2hbP8HykGIIN3\"\n }\n ]\n }\n }\n ]\n}\n", + "aiohttp-post-https://google.serper.dev/search-{\"data\": \"[{\\\"num\\\": 8, \\\"page\\\": 1, \\\"q\\\": \\\"metagpt\\\"}]\", \"headers\": {\"Content-Type\": \"application/json\", \"X-API-KEY\": \"mock-serper-key\"}}": [ + { + "searchParameters": { + "q": "metagpt", + "num": 8, + "page": 1, + "type": "search", + "engine": "google" + }, + "organic": [ + { + "title": "geekan/MetaGPT: The Multi-Agent Framework: Given one ... - GitHub", + "link": "https://github.com/geekan/MetaGPT", + "snippet": "MetaGPT takes a one line requirement as input and outputs user stories / competitive analysis / requirements / data structures / APIs / documents, etc.", + "sitelinks": [ + { + "title": "Roadmap", + "link": "https://github.com/geekan/MetaGPT/blob/main/docs/ROADMAP.md" + }, + { + "title": "README.md", + "link": "https://github.com/geekan/MetaGPT/blob/main/README.md" + }, + { + "title": "Issues 161", + "link": "https://github.com/geekan/MetaGPT/issues" + }, + { + "title": "Actions", + "link": "https://github.com/geekan/MetaGPT/actions" + } + ], + "position": 1 + }, + { + "title": "[2308.00352] MetaGPT: Meta Programming for A Multi-Agent ... - arXiv", + "link": "https://arxiv.org/abs/2308.00352", + "snippet": "Abstract:Remarkable progress has been made on automated problem solving through societies of agents based on large language models (LLMs).", + "date": "Aug 1, 2023", + "position": 2 + }, + { + "title": "MetaGPT: a Multi-Agent Framework to Automate Your Software ...", + "link": "https://medium.datadriveninvestor.com/metagpt-a-multi-agent-framework-to-automate-your-software-company-4b6ae747cc36", + "snippet": "MetaGPT is about to reach 10000 stars on Github. It's a Multi-Agent Framework that can behave as an engineer, product manager, architect, project managers.", + "position": 3 + }, + { + "title": "MetaGPT HUGE Update: Autonomous AI Agents with Incremental ...", + "link": "https://www.youtube.com/watch?v=Xyws6iI-eH8", + "snippet": "In this video, we unravel the magic at the core of MetaGPT, exploring its multi-agent framework ...", + "date": "Dec 23, 2023", + "attributes": { + "Duration": "11:38", + "Posted": "Dec 23, 2023" + }, + "imageUrl": "https://i.ytimg.com/vi/Xyws6iI-eH8/default.jpg?sqp=-oaymwEECHgQQw&rs=AMzJL3k9VHKSi-z-si4PJd1tNv8Itm4h5g", + "position": 4 + }, + { + "title": "MetaGPT - Apps on Google Play", + "link": "https://play.google.com/store/apps/details?id=com.metagpt.app&hl=en&gl=US", + "snippet": "Real-time crypto monitor.Track prices, set alerts, seize opportunities instantly.", + "date": "Jan 1, 2024", + "position": 5 + }, + { + "title": "MetaGPT | Discover AI use cases - GPT-3 Demo", + "link": "https://gpt3demo.com/apps/metagpt", + "snippet": "Assign different roles to GPTs to form a collaborative software entity for complex tasks. MetaGPT takes a one-line requirement as input and outputs user ...", + "position": 6 + }, + { + "title": "MetaGPT: AI-Powered Web Development That Changes the Game", + "link": "https://www.analyticsvidhya.com/blog/2024/01/meet-metagpt-the-chatgpt-powered-ai-assistant-that-turns-text-into-web-apps/", + "snippet": "MetaGPT is an AI assistant that leverages the power of GPT-4, a state-of-the-art language model developed by OpenAI. ChatGPT is trained on vast ...", + "date": "Jan 4, 2024", + "position": 7 + }, + { + "title": "MetaGPT: Complete Guide to the Best AI Agent Available Right Now", + "link": "https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/", + "snippet": "Discover why MetaGPT outperforms AutoGPT, BabyAgi, and other AI agents in complex coding tasks. Our in-depth article guides you through the ...", + "date": "Sep 11, 2023", + "position": 8 + } + ], + "relatedSearches": [ + { + "query": "MetaGPT online" + }, + { + "query": "Metagpt download" + }, + { + "query": "MetaGPT paper" + }, + { + "query": "Metagpt app" + }, + { + "query": "Metagpt github" + }, + { + "query": "MetaGPT huggingface" + }, + { + "query": "MetaGPT review" + }, + { + "query": "MetaGPT AI" + } + ] + } + ], + "aiohttp-post-https://google.serper.dev/search-{\"data\": \"[{\\\"num\\\": 6, \\\"page\\\": 1, \\\"q\\\": \\\"metagpt\\\"}]\", \"headers\": {\"Content-Type\": \"application/json\", \"X-API-KEY\": \"mock-serper-key\"}}": [ + { + "searchParameters": { + "q": "metagpt", + "num": 6, + "page": 1, + "type": "search", + "engine": "google" + }, + "organic": [ + { + "title": "geekan/MetaGPT: The Multi-Agent Framework: Given one ... - GitHub", + "link": "https://github.com/geekan/MetaGPT", + "snippet": "MetaGPT takes a one line requirement as input and outputs user stories / competitive analysis / requirements / data structures / APIs / documents, etc.", + "sitelinks": [ + { + "title": "Roadmap", + "link": "https://github.com/geekan/MetaGPT/blob/main/docs/ROADMAP.md" + }, + { + "title": "README.md", + "link": "https://github.com/geekan/MetaGPT/blob/main/README.md" + }, + { + "title": "Issues 161", + "link": "https://github.com/geekan/MetaGPT/issues" + }, + { + "title": "Actions", + "link": "https://github.com/geekan/MetaGPT/actions" + } + ], + "position": 1 + }, + { + "title": "[2308.00352] MetaGPT: Meta Programming for A Multi-Agent ... - arXiv", + "link": "https://arxiv.org/abs/2308.00352", + "snippet": "Abstract:Remarkable progress has been made on automated problem solving through societies of agents based on large language models (LLMs).", + "date": "Aug 1, 2023", + "position": 2 + }, + { + "title": "MetaGPT: a Multi-Agent Framework to Automate Your Software ...", + "link": "https://medium.datadriveninvestor.com/metagpt-a-multi-agent-framework-to-automate-your-software-company-4b6ae747cc36", + "snippet": "MetaGPT is about to reach 10000 stars on Github. It's a Multi-Agent Framework that can behave as an engineer, product manager, architect, project managers.", + "position": 3 + }, + { + "title": "MetaGPT HUGE Update: Autonomous AI Agents with Incremental ...", + "link": "https://www.youtube.com/watch?v=Xyws6iI-eH8", + "snippet": "In this video, we unravel the magic at the core of MetaGPT, exploring its multi-agent framework ...", + "date": "Dec 23, 2023", + "attributes": { + "Duration": "11:38", + "Posted": "Dec 23, 2023" + }, + "imageUrl": "https://i.ytimg.com/vi/Xyws6iI-eH8/default.jpg?sqp=-oaymwEECHgQQw&rs=AMzJL3k9VHKSi-z-si4PJd1tNv8Itm4h5g", + "position": 4 + }, + { + "title": "MetaGPT - Apps on Google Play", + "link": "https://play.google.com/store/apps/details?id=com.metagpt.app&hl=en&gl=US", + "snippet": "Real-time crypto monitor.Track prices, set alerts, seize opportunities instantly.", + "date": "Jan 1, 2024", + "position": 5 + }, + { + "title": "MetaGPT: AI-Powered Web Development That Changes the Game", + "link": "https://www.analyticsvidhya.com/blog/2024/01/meet-metagpt-the-chatgpt-powered-ai-assistant-that-turns-text-into-web-apps/", + "snippet": "MetaGPT is an AI assistant that leverages the power of GPT-4, a state-of-the-art language model developed by OpenAI. ChatGPT is trained on vast ...", + "date": "Jan 4, 2024", + "position": 6 + } + ], + "relatedSearches": [ + { + "query": "MetaGPT online" + }, + { + "query": "MetaGPT paper" + }, + { + "query": "Metagpt review" + }, + { + "query": "Metagpt download" + }, + { + "query": "Metagpt github" + }, + { + "query": "MetaGPT AI" + }, + { + "query": "MetaGPT huggingface" + }, + { + "query": "Metagpt OpenAI" + } + ] + } + ], + "curl-cffi-POST-https://duckduckgo.com-{\"data\": {\"q\": \"metagpt\"}}": "metagpt at DuckDuckGo
", + "curl-cffi-GET-https://links.duckduckgo.com/d.js-{\"params\": {\"bing_market\": \"wt-WT\", \"df\": null, \"ex\": \"-1\", \"kl\": \"wt-wt\", \"l\": \"wt-wt\", \"q\": \"metagpt\", \"s\": \"0\", \"sp\": \"0\", \"vqd\": \"4-118631859838297093459588814466521506726\"}}": "if (DDG.deep && DDG.deep.setUpstream) DDG.deep.setUpstream(\"bingv7aa\");DDG.deep.bn={'ivc':1};if (DDG.pageLayout) DDG.pageLayout.load('a',[], {\"page_load_url\":\"https://duckduckgo.com/y.js?iurl=%7B2%7DIG%3DFD8EFA3AD04A446FA24ACF32036DB0FF%26CID%3D3E2A18C583DE6B6F14A00CC382616A60%26Type%3DEvent.CPT%26DATA%3D0\"});DDG.deep.signalSummary = \"retail:h\";DDG.inject('DDG.Data.languages.resultLanguages', {\"en\":[\"https://github.com/geekan/MetaGPT\",\"https://docs.deepwisdom.ai/\",\"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\"https://docs.deepwisdom.ai/main/en/guide/get_started/introduction.html\",\"https://arxiv.org/abs/2308.00352\",\"https://github.com/geekan/MetaGPT/blob/main/README.md\",\"https://interestingengineering.com/innovation/metagpt-create-apps-text-prompts\",\"https://aibusiness.com/nlp/metagpt-text-to-app-ai-simplifies-web-dev\",\"https://github.com/PlaiD3/MetaGPT/blob/main/README.md\",\"https://www.youtube.com/watch?v=wpgC5fmtU70\",\"https://www.tomsguide.com/news/ai-tool-uses-chatgpt-to-build-you-a-website-in-30-minutes-and-we-tried-it\",\"https://www.kdnuggets.com/meet-metagpt-the-chatgptpowered-ai-assistant-that-turns-text-into-web-apps\",\"https://www.straight.com/guides/software/a-complete-guide-to-metagpt-the-best-ai-agent-available-now/\",\"https://www.marktechpost.com/2023/08/09/meet-metagpt-the-open-source-ai-framework-that-transforms-gpts-into-engineers-architects-and-managers/\",\"https://medium.com/gta-generative-tech-advances/metagpt-an-interesting-approach-to-multi-agent-collaboration-5ace263c4fd8\",\"https://www.youtube.com/watch?v=nqZlTV_L6Ao\",\"https://medium.com/aimonks/metagpt-a-framework-for-multi-agent-meta-programming-6c79f2eafb8e\",\"https://medium.com/@smraiyyan/metagpt-unleashed-crafting-your-virtual-software-company-from-scratch-6ea60cd70da1\",\"https://www.almabetter.com/bytes/articles/metagpt\",\"https://geekflare.com/metagpt-multi-agent-framework/\",\"https://analyticsindiamag.com/metagpt-realising-the-gpt-4-dream/\",\"https://medium.com/technology-hits/autogpt-langchain-deep-lake-metagpt-a-revolutionary-framework-for-building-advanced-ai-e2c579d86494\",\"https://generativeai.pub/analyzing-an-exciting-generative-ai-research-called-metagpt-2106385312db\",\"https://www.msn.com/en-us/news/technology/generative-ai-apis-and-chatgpt-alternatives-for-developers-to-consider/ar-AA1ltwXb\",\"https://www.theguardian.com/technology/2024/jan/08/ai-tools-chatgpt-copyrighted-material-openai\",\"https://www.forbes.com/sites/katiejennings/2024/01/05/health-ai-startup-nabla-was-built-on-gpt-4-now-its-abandoning-openai-for-open-source/\",\"https://www.technologyreview.com/2024/01/04/1086046/whats-next-for-ai-in-2024/\"],\"zh-CN\":[\"https://blog.csdn.net/Attitude93/article/details/135550499\",\"https://zhuanlan.zhihu.com/p/677608276\"]});DDG.deep.pageLayoutSummary = \"w29v1r1\";DDG.inject('DDG.Data.languages.adLanguages', {});if (DDG.pageLayout) DDG.pageLayout.load('d',[{\"a\":\"geekan/MetaGPT. This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository. About. \\ud83c\\udf1f The Multi-Agent Framework: Given one line Requirement, return PRD, Design, Tasks, Repo deepwisdom.ai/ Topics. agent multi-agent gpt hacktoberfest llm metagpt Resources. Readme\",\"ae\":null,\"b\":\"gh\\tGitHub\\tgithub.com\",\"c\":\"https://github.com/geekan/MetaGPT\",\"d\":\"github.com/geekan/MetaGPT\",\"da\":\"\",\"h\":0,\"i\":\"github.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: The Multi-Agent Framework - GitHub\",\"u\":\"https://github.com/geekan/MetaGPT\"},{\"a\":\"MetaGPT. The Multi-Agent Framework. Assign different roles to GPTs to form a collaborative software entity for complex tasks. Get Started. View on Github. Agents. Explore agent creation, configuration, and management, including algorithms and techniques. Demos.\",\"ae\":null,\"c\":\"https://docs.deepwisdom.ai/\",\"d\":\"docs.deepwisdom.ai\",\"da\":\"\",\"h\":0,\"i\":\"docs.deepwisdom.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT | MetaGPT\",\"u\":\"https://docs.deepwisdom.ai/\"},{\"a\":\"MetaGPT is a Multi-agent system that utilizes Large Language models and Standardized Operating Procedures to generate code in real-time. It outperforms other AI agents in code generation, collaboration, and code review. Learn how to install and use MetaGPT with examples and benchmarks.\",\"ae\":null,\"c\":\"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\"d\":\"www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\"da\":\"\",\"e\":\"2023-09-11T00:00:00.0000000\",\"h\":0,\"i\":\"www.unite.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Complete Guide to the Best AI Agent Available Right Now\",\"u\":\"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\"},{\"a\":\"Internally, MetaGPT includes product managers / architects / project managers / engineers. It provides the entire process of a software company along with carefully orchestrated SOPs. Code = SOP (Team) is the core philosophy. We materialize SOP and apply it to teams composed of LLMs. Software Company Multi-Role Schematic.\",\"ae\":null,\"c\":\"https://docs.deepwisdom.ai/main/en/guide/get_started/introduction.html\",\"d\":\"docs.deepwisdom.ai/main/en/guide/get_started/introduction.html\",\"da\":\"\",\"h\":0,\"i\":\"docs.deepwisdom.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: The Multi-Agent Framework | MetaGPT\",\"u\":\"https://docs.deepwisdom.ai/main/en/guide/get_started/introduction.html\"},{\"a\":\"MetaGPT is a novel method that uses human workflows to improve the performance of LLM-based multi-agent systems. It encodes SOPs into prompt sequences and assigns roles to agents to break down complex tasks into subtasks.\",\"ae\":null,\"b\":\"arx\\tarXiv.org\\tarxiv.org\",\"c\":\"https://arxiv.org/abs/2308.00352\",\"d\":\"arxiv.org/abs/2308.00352\",\"da\":\"translations\",\"e\":\"2023-08-01T00:00:00.0000000\",\"h\":0,\"i\":\"arxiv.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework\",\"u\":\"https://arxiv.org/abs/2308.00352\"},{\"a\":\"MetaGPT takes a one line requirement as input and outputs user stories / competitive analysis / requirements / data structures / APIs / documents, etc. \n Internally, MetaGPT includes product managers / architects / project managers / engineers.\",\"ae\":null,\"b\":\"gh\\tGitHub\\tgithub.com\",\"c\":\"https://github.com/geekan/MetaGPT/blob/main/README.md\",\"d\":\"github.com/geekan/MetaGPT/blob/main/README.md\",\"da\":\"\",\"h\":0,\"i\":\"github.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: The Multi-Agent Framework - GitHub\",\"u\":\"https://github.com/geekan/MetaGPT/blob/main/README.md\"},{\"a\":\"MetaGPT is a tool that lets you build websites, apps, and more using only text-based prompts powered by ChatGPT, an AI chatbot that can write code and improve it. You can create one app for free or subscribe for unlimited apps with MetaGPT.\",\"ae\":null,\"c\":\"https://interestingengineering.com/innovation/metagpt-create-apps-text-prompts\",\"d\":\"interestingengineering.com/innovation/metagpt-create-apps-text-prompts\",\"da\":\"\",\"e\":\"2023-05-08T12:57:00.0000000\",\"h\":0,\"i\":\"interestingengineering.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Create web-based apps with only text prompts\",\"u\":\"https://interestingengineering.com/innovation/metagpt-create-apps-text-prompts\"},{\"a\":\"MetaGPT is a new text-to-app generator that can create web apps from text descriptions. It uses ChatGPT's API and is free to use for commercial purposes. You can build microapps for various tasks or platforms, such as Facebook Messenger, Trello, or Microsoft Word.\",\"ae\":null,\"c\":\"https://aibusiness.com/nlp/metagpt-text-to-app-ai-simplifies-web-dev\",\"d\":\"aibusiness.com/nlp/metagpt-text-to-app-ai-simplifies-web-dev\",\"da\":\"\",\"e\":\"2023-08-07T00:00:00.0000000\",\"h\":0,\"i\":\"aibusiness.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Text-To-App AI Simplifies Web Dev\",\"u\":\"https://aibusiness.com/nlp/metagpt-text-to-app-ai-simplifies-web-dev\"},{\"a\":\"MetaGPT takes a one line requirement as input and outputs user stories / competitive analysis / requirements / data structures / APIs / documents, etc. \n; Internally, MetaGPT includes product managers / architects / project managers / engineers. It provides the entire process of a software company along with carefully orchestrated SOPs.\n \n\",\"ae\":null,\"b\":\"gh\\tGitHub\\tgithub.com\",\"c\":\"https://github.com/PlaiD3/MetaGPT/blob/main/README.md\",\"d\":\"github.com/PlaiD3/MetaGPT/blob/main/README.md\",\"da\":\"\",\"h\":0,\"i\":\"github.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Multi-Agent Meta Programming Framework - GitHub\",\"u\":\"https://github.com/PlaiD3/MetaGPT/blob/main/README.md\"},{\"a\":\"MetaGPT is a powerful no-code solution for app building that allows users to create web apps without any prerequisites of coding or technical experience. It ...\",\"ae\":null,\"b\":\"yt\\tYouTube\\twww.youtube.com\",\"c\":\"https://www.youtube.com/watch?v=wpgC5fmtU70\",\"d\":\"www.youtube.com/watch?v=wpgC5fmtU70\",\"da\":\"mlb_games,nba_games,ncaafb_games,ncaamb_games,nfl_games,nhl_games,soccer_games,translations,videos,wheretowatch\",\"h\":0,\"i\":\"www.youtube.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Meet MetaGPT A GPT-4-powered Application That Can Create ... - YouTube\",\"u\":\"https://www.youtube.com/watch?v=wpgC5fmtU70\"},{\"a\":\"Developed by New York-based company WhimsyWorks, MetaGPT offers users a no-code solution to turn their idea into a website or online app using AI. If you've recently used an AI chatbot, the ...\",\"ae\":null,\"c\":\"https://www.tomsguide.com/news/ai-tool-uses-chatgpt-to-build-you-a-website-in-30-minutes-and-we-tried-it\",\"d\":\"www.tomsguide.com/news/ai-tool-uses-chatgpt-to-build-you-a-website-in-30-minutes-and-we-tried-it\",\"da\":\"translations\",\"e\":\"2023-05-10T00:00:00.0000000\",\"h\":0,\"i\":\"www.tomsguide.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"AI tool uses ChatGPT to build you an app in 30 minutes - Tom's Guide\",\"u\":\"https://www.tomsguide.com/news/ai-tool-uses-chatgpt-to-build-you-a-website-in-30-minutes-and-we-tried-it\"},{\"a\":\"MetaGPT is a tool that lets you create no-code web applications using natural language. You can type in a text prompt and get a functional web app in seconds, using the power of GPT-4 and the Code Interpreter. Learn how to use MetaGPT for data science, analytics, and more.\",\"ae\":null,\"c\":\"https://www.kdnuggets.com/meet-metagpt-the-chatgptpowered-ai-assistant-that-turns-text-into-web-apps\",\"d\":\"www.kdnuggets.com/meet-metagpt-the-chatgptpowered-ai-assistant-that-turns-text-into-web-apps\",\"da\":\"\",\"e\":\"2023-09-08T00:00:00.0000000\",\"h\":0,\"i\":\"www.kdnuggets.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Meet MetaGPT: The ChatGPT-Powered AI Assistant That Turns Text Into Web ...\",\"u\":\"https://www.kdnuggets.com/meet-metagpt-the-chatgptpowered-ai-assistant-that-turns-text-into-web-apps\"},{\"a\":\"MetaGPT is a multi-agent system that uses large language models to perform complex tasks. It can understand, generate, and interact with natural language input and output. Learn about its features, capabilities, applications, and advantages in this complete guide.\",\"ae\":null,\"c\":\"https://www.straight.com/guides/software/a-complete-guide-to-metagpt-the-best-ai-agent-available-now/\",\"d\":\"www.straight.com/guides/software/a-complete-guide-to-metagpt-the-best-ai-agent-available-now/\",\"da\":\"\",\"e\":\"2023-12-13T00:00:00.0000000\",\"h\":0,\"i\":\"www.straight.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"A Complete Guide to MetaGPT: The Best AI Agent Available Now\",\"u\":\"https://www.straight.com/guides/software/a-complete-guide-to-metagpt-the-best-ai-agent-available-now/\"},{\"a\":\"MetaGPT is an open-source AI framework that transforms GPTs into engineers, architects, and managers by using role-based action specifications and SOPs. It can generate high-quality code, design, and documentation for software engineering, data analysis, and game development tasks.\",\"ae\":null,\"c\":\"https://www.marktechpost.com/2023/08/09/meet-metagpt-the-open-source-ai-framework-that-transforms-gpts-into-engineers-architects-and-managers/\",\"d\":\"www.marktechpost.com/2023/08/09/meet-metagpt-the-open-source-ai-framework-that-transforms-gpts-into-engineers-architects-and-managers/\",\"da\":\"translations\",\"e\":\"2023-08-09T00:00:00.0000000\",\"h\":0,\"i\":\"www.marktechpost.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Meet MetaGPT: The Open-Source AI Framework That Transforms GPTs into ...\",\"u\":\"https://www.marktechpost.com/2023/08/09/meet-metagpt-the-open-source-ai-framework-that-transforms-gpts-into-engineers-architects-and-managers/\"},{\"a\":\"MetaGPT is the maestro who brings harmony to this chaos. By encoding Standardized Operating Procedures (SOPs) into prompts, MetaGPT ensures structured collaboration akin to a well-rehearsed ...\",\"ae\":null,\"c\":\"https://medium.com/gta-generative-tech-advances/metagpt-an-interesting-approach-to-multi-agent-collaboration-5ace263c4fd8\",\"d\":\"medium.com/gta-generative-tech-advances/metagpt-an-interesting-approach-to-multi-agent-collaboration-5ace263c4fd8\",\"da\":\"\",\"e\":\"2023-08-15T00:00:00.0000000\",\"h\":0,\"i\":\"medium.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: An Interesting Approach to Multi-Agent Collaboration\",\"u\":\"https://medium.com/gta-generative-tech-advances/metagpt-an-interesting-approach-to-multi-agent-collaboration-5ace263c4fd8\"},{\"a\":\"Welcome to our video review! \\ud83c\\udfa5 Dive into the world of MetaGPT, a revolutionary project that's redefining the boundaries of AI. \\ud83e\\udd16 Imagine having an entire e...\",\"ae\":null,\"b\":\"yt\\tYouTube\\twww.youtube.com\",\"c\":\"https://www.youtube.com/watch?v=nqZlTV_L6Ao\",\"d\":\"www.youtube.com/watch?v=nqZlTV_L6Ao\",\"da\":\"mlb_games,nba_games,ncaafb_games,ncaamb_games,nfl_games,nhl_games,soccer_games,videos,wheretowatch\",\"e\":\"2023-09-04T00:00:00.0000000\",\"h\":0,\"i\":\"www.youtube.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT Setup: Launch a Startup with One \\ufe0f Prompt! - YouTube\",\"u\":\"https://www.youtube.com/watch?v=nqZlTV_L6Ao\"},{\"a\":\"MetaGPT is a model that uses the power of natural language to create and execute meta programs for multi-agent collaboration. Meta programs are programs that can generate or modify other programs ...\",\"ae\":null,\"c\":\"https://medium.com/aimonks/metagpt-a-framework-for-multi-agent-meta-programming-6c79f2eafb8e\",\"d\":\"medium.com/aimonks/metagpt-a-framework-for-multi-agent-meta-programming-6c79f2eafb8e\",\"da\":\"\",\"e\":\"2023-08-03T00:00:00.0000000\",\"h\":0,\"i\":\"medium.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: A Framework for Multi-Agent Meta Programming\",\"u\":\"https://medium.com/aimonks/metagpt-a-framework-for-multi-agent-meta-programming-6c79f2eafb8e\"},{\"a\":\"MetaGPT, available on Github (crossed 13,000 stars), aims to change the way we make software.This exciting tool can take a single line of what you want to do and turn it into many things like user ...\",\"ae\":null,\"c\":\"https://medium.com/@smraiyyan/metagpt-unleashed-crafting-your-virtual-software-company-from-scratch-6ea60cd70da1\",\"d\":\"medium.com/@smraiyyan/metagpt-unleashed-crafting-your-virtual-software-company-from-scratch-6ea60cd70da1\",\"da\":\"translations\",\"e\":\"2023-08-07T00:00:00.0000000\",\"h\":0,\"i\":\"medium.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT Lets You Create Your Own Virtual Software Company from ... - Medium\",\"u\":\"https://medium.com/@smraiyyan/metagpt-unleashed-crafting-your-virtual-software-company-from-scratch-6ea60cd70da1\"},{\"a\":\"Discover the revolutionary advancements of MetaGPT and its potential impact on the future of AI-powered solutions. Artificial Intelligence has experienced remarkable progress in recent years, with one term in particular capturing the attention of the digital landscape: MetaGPT online. It can also be referred to as one of the ChatGPT alternatives.In an increasingly competitive environment ...\",\"ae\":null,\"c\":\"https://www.almabetter.com/bytes/articles/metagpt\",\"d\":\"www.almabetter.com/bytes/articles/metagpt\",\"da\":\"\",\"e\":\"2023-08-28T00:00:00.0000000\",\"h\":0,\"i\":\"www.almabetter.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: The Future of Multi-Agent Collaboration in AI\",\"u\":\"https://www.almabetter.com/bytes/articles/metagpt\"},{\"a\":\"MetaGPT is a framework that uses different GPTs to generate APIs, user stories, data structures, and more. It can automate software development tasks, enhance existing programs, and collaborate with other agents. Learn how to get started, use cases, advantages, and alternatives of MetaGPT.\",\"ae\":null,\"c\":\"https://geekflare.com/metagpt-multi-agent-framework/\",\"d\":\"geekflare.com/metagpt-multi-agent-framework/\",\"da\":\"\",\"e\":\"2023-09-18T00:00:00.0000000\",\"h\":0,\"i\":\"geekflare.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Is This the Best Multi-Agent Framework Yet? - Geekflare\",\"u\":\"https://geekflare.com/metagpt-multi-agent-framework/\"},{\"a\":\"MetaGPT is a web app that allows users to build web applications using natural language prompts and ChatGPT, a multimodal language model. The service has been used to create dashboards, code-based visualisations, and even a marriage proposal, showing the potential of GPT-4 and its plugins.\",\"ae\":null,\"c\":\"https://analyticsindiamag.com/metagpt-realising-the-gpt-4-dream/\",\"d\":\"analyticsindiamag.com/metagpt-realising-the-gpt-4-dream/\",\"da\":\"\",\"e\":\"2023-04-26T00:00:00.0000000\",\"h\":0,\"i\":\"analyticsindiamag.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT \\u2014 Realising the GPT-4 Dream - Analytics India Magazine\",\"u\":\"https://analyticsindiamag.com/metagpt-realising-the-gpt-4-dream/\"},{\"a\":\"MetaGPT, or multimodal Generative Pretrained Transformers, represents a significant leap in the evolution of artificial intelligence. This new generation of AI models is capable of understanding ...\",\"ae\":null,\"c\":\"https://medium.com/technology-hits/autogpt-langchain-deep-lake-metagpt-a-revolutionary-framework-for-building-advanced-ai-e2c579d86494\",\"d\":\"medium.com/technology-hits/autogpt-langchain-deep-lake-metagpt-a-revolutionary-framework-for-building-advanced-ai-e2c579d86494\",\"da\":\"\",\"e\":\"2023-08-28T00:00:00.0000000\",\"h\":0,\"i\":\"medium.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"AutoGPT \\u2014 LangChain \\u2014 Deep Lake \\u2014 MetaGPT: A ... - Medium\",\"u\":\"https://medium.com/technology-hits/autogpt-langchain-deep-lake-metagpt-a-revolutionary-framework-for-building-advanced-ai-e2c579d86494\"},{\"a\":\"Overview of the MetaGPT framework. Presented is a two-layer architectural design: i) the Foundational Components Layer, which is essential for agent operations and system-wide communication, and ii) the Collaboration Layer, which facilitates agent coordination through key mechanisms such as knowledge sharing and workflow encapsulation.\",\"ae\":null,\"c\":\"https://generativeai.pub/analyzing-an-exciting-generative-ai-research-called-metagpt-2106385312db\",\"d\":\"generativeai.pub/analyzing-an-exciting-generative-ai-research-called-metagpt-2106385312db\",\"da\":\"translations\",\"e\":\"2023-08-14T00:00:00.0000000\",\"h\":0,\"i\":\"generativeai.pub\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Analyzing an exciting Generative AI research called MetaGPT.\",\"u\":\"https://generativeai.pub/analyzing-an-exciting-generative-ai-research-called-metagpt-2106385312db\"},{\"a\":\"MetaGPT\\u5c06\\u4f1a\\u6309\\u7167\\u4e0b\\u8ff0\\u4f18\\u5148\\u7ea7\\u6765\\u8bfb\\u53d6\\u4f60\\u7684\\u914d\\u7f6e\\uff1aconfig/key.yaml > config/config.yaml > environment variable. \\u6211\\u8fd9\\u91cc\\u4f7f\\u7528\\u73af\\u5883\\u53d8\\u91cf\\u7684\\u65b9\\u5f0f\\u3002. \\uff081\\uff09\\u521b\\u5efa\\u4e00\\u4e2a\\u5de5\\u7a0b\\u76ee\\u5f55 MyMetaGPT\\uff0c\\u7528VSCode\\u6253\\u5f00. \\uff082\\uff09\\u65b0\\u5efa\\u4e00\\u4e2a.env\\u6587\\u4ef6\\uff0c\\u5c06\\u4ee5\\u4e0a\\u914d\\u7f6e\\u586b\\u52a0\\u5230\\u8be5\\u6587\\u4ef6\\u4e2d. \\u5728Python\\u6587\\u4ef6\\uff08MetaGPT_test.py\\uff09\\u4e2d\\u5c06\\u8be5.env\\u6587\\u4ef6 ...\",\"ae\":null,\"c\":\"https://blog.csdn.net/Attitude93/article/details/135550499\",\"d\":\"blog.csdn.net/Attitude93/article/details/135550499\",\"da\":\"translations\",\"e\":\"2024-01-13T00:00:00.0000000\",\"h\":0,\"i\":\"blog.csdn.net\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"AI Agent\\u7cfb\\u5217\\u3011\\u3010MetaGPT\\u30110. \\u4f60\\u7684\\u7b2c\\u4e00\\u4e2aMetaGPT\\u7a0b\\u5e8f - CSDN\\u535a\\u5ba2\",\"u\":\"https://blog.csdn.net/Attitude93/article/details/135550499\"},{\"a\":\"Here are nine of the best ChatGPT alternatives and generative AI APIs for developers that are worth checking out. 1. Meta: Llama2. do is download and install Llama 2 locally. Related video: How AI ...\",\"ae\":null,\"c\":\"https://www.msn.com/en-us/news/technology/generative-ai-apis-and-chatgpt-alternatives-for-developers-to-consider/ar-AA1ltwXb\",\"d\":\"www.msn.com/en-us/news/technology/generative-ai-apis-and-chatgpt-alternatives-for-developers-to-consider/ar-AA1ltwXb\",\"da\":\"news\",\"e\":\"2023-12-13T00:00:00.0000000\",\"h\":0,\"i\":\"www.msn.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Generative AI APIs and ChatGPT Alternatives for Developers to ... - MSN\",\"u\":\"https://www.msn.com/en-us/news/technology/generative-ai-apis-and-chatgpt-alternatives-for-developers-to-consider/ar-AA1ltwXb\"},{\"a\":\"Pressure grows on artificial intelligence firms over the content used to train their products\",\"ae\":null,\"c\":\"https://www.theguardian.com/technology/2024/jan/08/ai-tools-chatgpt-copyrighted-material-openai\",\"d\":\"www.theguardian.com/technology/2024/jan/08/ai-tools-chatgpt-copyrighted-material-openai\",\"da\":\"news,translations\",\"e\":\"2024-01-08T07:15:00.0000000\",\"h\":0,\"i\":\"www.theguardian.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"'Impossible' to create AI tools like ChatGPT without copyrighted ...\",\"u\":\"https://www.theguardian.com/technology/2024/jan/08/ai-tools-chatgpt-copyrighted-material-openai\"},{\"a\":\"The Paris-based startup has raised $24 million at a $180 million valuation to shift its doctor note-taking software towards the open source AI models championed by Meta AI chief Yann Lecun, one of ...\",\"ae\":null,\"c\":\"https://www.forbes.com/sites/katiejennings/2024/01/05/health-ai-startup-nabla-was-built-on-gpt-4-now-its-abandoning-openai-for-open-source/\",\"d\":\"www.forbes.com/sites/katiejennings/2024/01/05/health-ai-startup-nabla-was-built-on-gpt-4-now-its-abandoning-openai-for-open-source/\",\"da\":\"news,translations\",\"e\":\"2024-01-05T11:00:00.0000000\",\"h\":0,\"i\":\"www.forbes.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Health AI Startup Nabla Was Built On GPT-4. Now, It's ... - Forbes\",\"u\":\"https://www.forbes.com/sites/katiejennings/2024/01/05/health-ai-startup-nabla-was-built-on-gpt-4-now-its-abandoning-openai-for-open-source/\"},{\"a\":\"\\u57282023\\u5e7412\\u670819\\u65e5\\u65f6\\uff0c\\u542c\\u4e86\\u6797\\u4e49\\u7ae0\\u8001\\u5e08\\u5173\\u4e8e"\\u57fa\\u4e8eMetaGPT\\u8fdb\\u884c\\u667a\\u80fd\\u4f53\\u5f00\\u53d1"\\u7684\\u8bb2\\u5ea7\\uff1a \\u89c9\\u5f97\\u65b0\\u5947\\u6709\\u8da3\\uff0c\\u5982\\u679c\\u80fd\\u8fd9\\u6837\\u5728\\u5de5\\u4f5c\\u751f\\u6d3b\\u4e2d\\u5b8c\\u6210\\u81ea\\u5df1\\u7684\\u4efb\\u52a1\\uff0c\\u90a3\\u7b80\\u76f4\\u662f\\u4e8b\\u534a\\u529f\\u500d\\u3002\\u4e8e\\u662f\\u8fd9\\u4e24\\u5929\\u53c8\\u5b66\\u4e60\\u4e86\\u300aMetaGPT\\u667a\\u80fd\\u4f53\\u5f00\\u53d1\\u5165\\u95e8\\u300b\\u6559\\u2026\",\"ae\":null,\"c\":\"https://zhuanlan.zhihu.com/p/677608276\",\"d\":\"zhuanlan.zhihu.com/p/677608276\",\"da\":\"translations\",\"e\":\"2024-01-12T00:00:00.0000000\",\"h\":0,\"i\":\"zhuanlan.zhihu.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"\\u5b66\\u4e60\\u7b14\\u8bb0-\\u300aMetaGPT\\u667a\\u80fd\\u4f53\\u5f00\\u53d1\\u5165\\u95e8\\u300b\\u6559\\u7a0b - \\u77e5\\u4e4e\",\"u\":\"https://zhuanlan.zhihu.com/p/677608276\"},{\"a\":\"In 2024, generative AI might actually become useful for the regular, non-tech person, and we are going to see more people tinkering with a million little AI models. State-of-the-art AI models ...\",\"ae\":null,\"c\":\"https://www.technologyreview.com/2024/01/04/1086046/whats-next-for-ai-in-2024/\",\"d\":\"www.technologyreview.com/2024/01/04/1086046/whats-next-for-ai-in-2024/\",\"da\":\"translations\",\"e\":\"2024-01-04T09:14:17.0000000\",\"h\":0,\"i\":\"www.technologyreview.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"What's next for AI in 2024 | MIT Technology Review\",\"u\":\"https://www.technologyreview.com/2024/01/04/1086046/whats-next-for-ai-in-2024/\"},{\"n\":\"/d.js?q=metagpt&kl=wt-wt&l=wt-wt&p=&s=29&ex=-1&ct=US&sp=0&vqd=4-118631859838297093459588814466521506726\"}]);DDG.duckbar.load('images');DDG.duckbar.load('news');DDG.duckbar.load('videos', {\"ads\":[],\"query\":\"metagpt\",\"queryEncoded\":\"metagpt\",\"response_type\":\"places\",\"results\":[{\"content\":\"https://www.youtube.com/watch?v=uT75J_KG_aY\",\"description\":\"In this video, we review MetaGPT, a new project that aims to recreate an entire engineering organization using AI. MetaGPT is a CEO, Product Manager, Architect, Project Manager, Engineering, and QA. Write a simple prompt, and you get everything from the requirements to the PRDs to the code and tests. How To Find Me: Become a Patron \\ud83d\\udd25 - https ...\",\"duration\":\"6:36\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/uT75J_KG_aY?autoplay=1\",\"image_token\":\"57974159b78b309485721c0bce280219d9927e071e542a34777864767d6cb8d4\",\"images\":{\"large\":\"https://tse3.mm.bing.net/th?id=OVP.BbSKV8N1vyYv-3m8vyuCoQEsDh&pid=Api\",\"medium\":\"https://tse3.mm.bing.net/th?id=OVP.BbSKV8N1vyYv-3m8vyuCoQEsDh&pid=Api\",\"motion\":\"https://tse3.mm.bing.net/th?id=OM1.bsXxoMoJ9ZWQBw&pid=Api\",\"small\":\"https://tse3.mm.bing.net/th?id=OVP.BbSKV8N1vyYv-3m8vyuCoQEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-08-14T14:09:10.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":75408},\"title\":\"How To Install MetaGPT - Build A Startup With One Prompt!!\",\"uploader\":\"Matthew Berman\"},{\"content\":\"https://www.youtube.com/watch?v=pJwR5pv0_gs\",\"description\":\"Multi agent framework tutorial of MetaGPT & chatDev; Check the Hubspot x Jasper research of Using Generative AI to Scale Your Content Operations: https://offers.hubspot.com/generative-ai-for-content-operations?utm_source=youtube&utm_medium=social&utm_campaign=CR0087Sep2023_AIJason/partner_youtube \\ud83d\\udd17 Links - Follow me on twitter: https ...\",\"duration\":\"13:41\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/pJwR5pv0_gs?autoplay=1\",\"image_token\":\"18ac54a8e5144c74f2010219781c47c295099a6eed7479645733832910d19aec\",\"images\":{\"large\":\"https://tse4.mm.bing.net/th?id=OVP.LJ0SK8DLWjCcwVVh-PEcOwHgFo&pid=Api\",\"medium\":\"https://tse4.mm.bing.net/th?id=OVP.LJ0SK8DLWjCcwVVh-PEcOwHgFo&pid=Api\",\"motion\":\"https://tse4.mm.bing.net/th?id=OM2.PxMMOsse4Yi_FQ&pid=Api\",\"small\":\"https://tse4.mm.bing.net/th?id=OVP.LJ0SK8DLWjCcwVVh-PEcOwHgFo&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-09-08T11:36:03.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":167793},\"title\":\"Build AI agent workforce - Multi agent framework with MetaGPT & chatDev\",\"uploader\":\"AI Jason\"},{\"content\":\"https://www.youtube.com/watch?v=q16Gi9pTG_M\",\"description\":\"In this captivating video, we explore the core concept of MetaGPT, which centers on task distribution and coordination among individual GPT agents. Each agent is bestowed with specific roles that capitalize on their unique strengths and expertise. Imagine one GPT excelling in natural language understanding, while another showcases prowess in ...\",\"duration\":\"14:56\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/q16Gi9pTG_M?autoplay=1\",\"image_token\":\"bee3657ef83c9da2bc4ccfea770244e18958f5789a39d0136c3a049cc22a0e54\",\"images\":{\"large\":\"https://tse4.mm.bing.net/th?id=OVP.eiPUmQWRU1sE-01-x5Kn7gEsDh&pid=Api\",\"medium\":\"https://tse4.mm.bing.net/th?id=OVP.eiPUmQWRU1sE-01-x5Kn7gEsDh&pid=Api\",\"motion\":\"https://tse4.mm.bing.net/th?id=OM2.eWDmjf8nvrSrhw&pid=Api\",\"small\":\"https://tse4.mm.bing.net/th?id=OVP.eiPUmQWRU1sE-01-x5Kn7gEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-07-25T00:37:40.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":14365},\"title\":\"MetaGPT: Deploy POWERFUL Autonomous Ai Agents BETTER Than SUPERAGI! (Installation Tutorial)\",\"uploader\":\"WorldofAI\"},{\"content\":\"https://www.youtube.com/watch?v=nqZlTV_L6Ao\",\"description\":\"Welcome to our video review! \\ud83c\\udfa5 Dive into the world of MetaGPT, a revolutionary project that's redefining the boundaries of AI. \\ud83e\\udd16 Imagine having an entire engineering team - from CEO to QA - compacted into one AI system. Just input a prompt, and voila! You're handed everything from requirements, PRDs, to the actual code and tests. Let ...\",\"duration\":\"14:15\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/nqZlTV_L6Ao?autoplay=1\",\"image_token\":\"9d13b27084400da23ef8d8567bd6b5c8a3758d4129f2b28c3619c0e2e1ba8276\",\"images\":{\"large\":\"https://tse4.mm.bing.net/th?id=OVP.VBEy5DF-0BQshjEkqA9T0wHgFo&pid=Api\",\"medium\":\"https://tse4.mm.bing.net/th?id=OVP.VBEy5DF-0BQshjEkqA9T0wHgFo&pid=Api\",\"motion\":\"https://tse4.mm.bing.net/th?id=OM2.N7S3-wAngkj7VA&pid=Api\",\"small\":\"https://tse4.mm.bing.net/th?id=OVP.VBEy5DF-0BQshjEkqA9T0wHgFo&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-09-04T11:45:06.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":23248},\"title\":\"\\ud83d\\ude80 MetaGPT Setup: Launch a Startup with One \\u270d\\ufe0f Prompt!\",\"uploader\":\"Prompt Engineering\"},{\"content\":\"https://www.youtube.com/watch?v=VxhPcnsA7KA\",\"description\":\"Meet MetaGPT, MetaGPT is a complete Software Engineering organization at your disposal. MetaGPT employees autonomous AI agents specializing in the roles found in real Software Development companies. Deploying Autonomous GPT AI Product Managers, Architects, Project Managers and Engineers your software is developed for you and Documented! This ...\",\"duration\":\"8:49\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/VxhPcnsA7KA?autoplay=1\",\"image_token\":\"c56ab50565d7135c0d45f37ea4b70f565eced03024b608392498704c54b0fe66\",\"images\":{\"large\":\"https://tse1.mm.bing.net/th?id=OVP.RTXFEZ-JNqeIG3Bfi8B3UQEsDh&pid=Api\",\"medium\":\"https://tse1.mm.bing.net/th?id=OVP.RTXFEZ-JNqeIG3Bfi8B3UQEsDh&pid=Api\",\"motion\":\"https://tse1.mm.bing.net/th?id=OM2.219b44Lsywj5Bg&pid=Api\",\"small\":\"https://tse1.mm.bing.net/th?id=OVP.RTXFEZ-JNqeIG3Bfi8B3UQEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-08-21T00:25:20.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":6190},\"title\":\"How To Install MetaGPT - Your own AI Software Company, Create Programs With a single Prompt!\",\"uploader\":\"StuffAboutStuff\"},{\"content\":\"https://www.youtube.com/watch?v=Xyws6iI-eH8\",\"description\":\"In this video, we unravel the magic at the core of MetaGPT, exploring its multi-agent framework and the groundbreaking December 15 update (v0.5.0) that introduced incremental development. Join us on this journey of innovation and efficiency in AI! \\ud83d\\udd25 Become a Patron (Private Discord): https://patreon.com/WorldofAi \\u2615 To help and Support me ...\",\"duration\":\"11:38\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/Xyws6iI-eH8?autoplay=1\",\"image_token\":\"db0651076b86c15566c0f032ab3e035fa65863cdf0b3bf46a18d10201bad1bab\",\"images\":{\"large\":\"https://tse3.mm.bing.net/th?id=OVP.X0OdKOGTJgwUw3Op_rmcewEsDh&pid=Api\",\"medium\":\"https://tse3.mm.bing.net/th?id=OVP.X0OdKOGTJgwUw3Op_rmcewEsDh&pid=Api\",\"motion\":\"https://tse3.mm.bing.net/th?id=OM2.-Hw5pO2PnG7h1g&pid=Api\",\"small\":\"https://tse3.mm.bing.net/th?id=OVP.X0OdKOGTJgwUw3Op_rmcewEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-12-23T21:22:36.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":7003},\"title\":\"MetaGPT HUGE Update: Autonomous AI Agents with Incremental Memory!\",\"uploader\":\"WorldofAI\"},{\"content\":\"https://www.youtube.com/watch?v=T_wBUpzxxPY\",\"description\":\"In this video i talk about this awesome project called MetaGPT in my video. Now, MetaGPT is like an all-in-one AI powerhouse. It can do everything from being a CEO to a QA tester for an engineering organization. And the cool thing is, you just give it a simple prompt, and it spits out everything you need - requirements, PRDs, code, and tests ...\",\"duration\":\"4:00\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/T_wBUpzxxPY?autoplay=1\",\"image_token\":\"ef14791d7faff848cb15177567e9f4f9c04ccae4fafc7ef7386e69df3a012010\",\"images\":{\"large\":\"https://tse1.mm.bing.net/th?id=OVP.EWCOFStB_tQza4SLrUA0AAEsDh&pid=Api\",\"medium\":\"https://tse1.mm.bing.net/th?id=OVP.EWCOFStB_tQza4SLrUA0AAEsDh&pid=Api\",\"motion\":\"https://tse1.mm.bing.net/th?id=OM1.itG5pHJg6MKYzg_1696190983&pid=Api\",\"small\":\"https://tse1.mm.bing.net/th?id=OVP.EWCOFStB_tQza4SLrUA0AAEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-09-11T10:41:22.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":368},\"title\":\"MetaGPT Installation Guide: From Setup to Startup With One Prompt!\",\"uploader\":\"Py Man\"},{\"content\":\"https://www.youtube.com/watch?v=EgipcKPhqME\",\"description\":\"In this video I provide a great demo and overview of a project called MetaGPT. Have you ever wondered if each person in a development project (such as the project manager, developers, architects, QA testers, etc.) were all AI's and how they'd behave? MetaGPT is doing just that. Not only are all the docs, designs, and tasks delivered, but also a ...\",\"duration\":\"7:35\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/EgipcKPhqME?autoplay=1\",\"image_token\":\"624d4ccdb6d1605da1e388e85c9124957bcba9c70a11a575e751ba6fc09bc5f8\",\"images\":{\"large\":\"https://tse4.mm.bing.net/th?id=OVP.hG0c3nw7X-uz0gzUjnOVNwEsDh&pid=Api\",\"medium\":\"https://tse4.mm.bing.net/th?id=OVP.hG0c3nw7X-uz0gzUjnOVNwEsDh&pid=Api\",\"motion\":\"https://tse4.mm.bing.net/th?id=OM1.8F2lEMy1JlCKsQ_1698986522&pid=Api\",\"small\":\"https://tse4.mm.bing.net/th?id=OVP.hG0c3nw7X-uz0gzUjnOVNwEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-09-24T08:00:11.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":1587},\"title\":\"MetaGPT Tutorial | It builds an entire project (with working source code) with just one prompt!!\",\"uploader\":\"CraceCasts\"},{\"content\":\"https://www.youtube.com/watch?v=YtxMderNrzU\",\"description\":\"Subscribe to my Newsletter (My AI updates and news clearly explained): https://louisbouchard.substack.com/ References: Read the full article: https://www.louisbouchard.ai/metagpt/ Hong et al., 2023: MetaGPT, https://arxiv.org/pdf/2308.00352.pdf Code: https://github.com/geekan/MetaGPT/blob/main/README.md Twitter: https://twitter.com/Whats_AI ...\",\"duration\":\"7:38\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/YtxMderNrzU?autoplay=1\",\"image_token\":\"2e0774ace2e34bbe23ece04e80b7bb2ee976fd8ef7f53001e8f8b137763561dc\",\"images\":{\"large\":\"https://tse2.mm.bing.net/th?id=OVP.HP81CZ34ap22GZZG2l024QHgFo&pid=Api\",\"medium\":\"https://tse2.mm.bing.net/th?id=OVP.HP81CZ34ap22GZZG2l024QHgFo&pid=Api\",\"motion\":\"https://tse2.mm.bing.net/th?id=OM2.xArTjo5bOxSBhg&pid=Api\",\"small\":\"https://tse2.mm.bing.net/th?id=OVP.HP81CZ34ap22GZZG2l024QHgFo&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-08-27T15:05:12.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":9594},\"title\":\"MetaGPT: Redefining Multi-Agent Collaboration for Complex Tasks\",\"uploader\":\"What's AI by Louis Bouchard\"},{\"content\":\"https://www.youtube.com/watch?v=geLX30qax8Q\",\"description\":\"Dive into the world of autonomous agent swarms with our comprehensive Autonomous Agent Swarms Totorial! \\ud83c\\udf1f Whether you're a beginner or an AI enthusiast, this video will guide you through the fascinating process of creating and managing intelligent agent swarms using LangChain. Learn how to harness the power of collaborative AI agents for ...\",\"duration\":\"47:02\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/geLX30qax8Q?autoplay=1\",\"image_token\":\"9e9f6de14802f66e1364f3ef0c9a7973fcfab471dff810a91595a2ea60242256\",\"images\":{\"large\":\"https://tse2.mm.bing.net/th?id=OVP.0DNi9RS5yLCZHu9MXx5lQAEsDh&pid=Api\",\"medium\":\"https://tse2.mm.bing.net/th?id=OVP.0DNi9RS5yLCZHu9MXx5lQAEsDh&pid=Api\",\"motion\":\"https://tse2.mm.bing.net/th?id=OM.u64mpOw5ZNaPbg_1704713419&pid=Api\",\"small\":\"https://tse2.mm.bing.net/th?id=OVP.0DNi9RS5yLCZHu9MXx5lQAEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-11-12T00:29:27.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":12461},\"title\":\"Autonomous AI Agent Swarms | COMPLETE Tutorial\",\"uploader\":\"AspnAI\"}],\"vqd\":{\"metagpt\":\"4-118631859838297093459588814466521506726\"}});DDG.duckbar.loadModule('related_searches', {\"ads\":[],\"query\":\"metagpt\",\"queryEncoded\":\"metagpt\",\"response_type\":\"places\",\"results\":[{\"display_text\":\"metagpt sign in\",\"text\":\"metagpt sign in\",\"web_search_url\":\"?q=metagpt%20sign%20in\"},{\"display_text\":\"metagpt download\",\"text\":\"metagpt download\",\"web_search_url\":\"?q=metagpt%20download\"},{\"display_text\":\"metagpt vs autogpt\",\"text\":\"metagpt vs autogpt\",\"web_search_url\":\"?q=metagpt%20vs%20autogpt\"},{\"display_text\":\"metagpt examples\",\"text\":\"metagpt examples\",\"web_search_url\":\"?q=metagpt%20examples\"},{\"display_text\":\"metagpt windows\",\"text\":\"metagpt windows\",\"web_search_url\":\"?q=metagpt%20windows\"},{\"display_text\":\"metagpt arxiv\",\"text\":\"metagpt arxiv\",\"web_search_url\":\"?q=metagpt%20arxiv\"},{\"display_text\":\"tell me what is metagpt\",\"text\":\"tell me what is metagpt\",\"web_search_url\":\"?q=tell%20me%20what%20is%20metagpt\"},{\"display_text\":\"metagpt pdf\",\"text\":\"metagpt pdf\",\"web_search_url\":\"?q=metagpt%20pdf\"}],\"vqd\":{\"metagpt\":\"4-118631859838297093459588814466521506726\"}});if (DDG.pageLayout) DDG.pageLayout.initialize({\"mainline\":{\"items\":[[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"videos\"],[\"related_searches\"]]}}, { start: 0 });DDG.deep.emit(\"load:completed\");", + "curl-cffi-POST-https://duckduckgo.com-{\"data\": {\"q\": \"llm\"}}": "llm at DuckDuckGo
", + "curl-cffi-GET-https://links.duckduckgo.com/d.js-{\"params\": {\"bing_market\": \"wt-WT\", \"df\": null, \"ex\": \"-1\", \"kl\": \"wt-wt\", \"l\": \"wt-wt\", \"q\": \"llm\", \"s\": \"0\", \"sp\": \"0\", \"vqd\": \"4-36212277936736004277629252433802891730\"}}": "if (DDG.deep && DDG.deep.setUpstream) DDG.deep.setUpstream(\"bingv7aa\");DDG.deep.bn={'ivc':1};if (DDG.pageLayout) DDG.pageLayout.load('a',[{\"a\":\"\\u6d77\\u5916\\u30c8\\u30c3\\u30d7\\u6821\\u307810,000\\u4eba\\u4ee5\\u4e0a\\u306e\\u5408\\u683c\\u5b9f\\u7e3e!\\u307e\\u305a\\u306f\\u7121\\u6599\\u500b\\u5225\\u76f8\\u8ac7\\u3001\\u7121\\u6599\\u30a4\\u30d9\\u30f3\\u30c8\\u3078. \\u9078\\u3079\\u308b\\u5b66\\u7fd2\\u5f62\\u614b\\u3001\\u53d7\\u8b1b\\u671f\\u95933\\u5e74\\u3001\\u518d\\u53d7\\u8b1b\\u7121\\u6599\\u306e\\u30a2\\u30b4\\u30b9\\u3067\\u30b0\\u30ed\\u30fc\\u30d0\\u30eb\\u30ec\\u30d9\\u30eb\\u306e\\u30ad\\u30e3\\u30ea\\u30a2\\u3092\\u76ee\\u6307\\u305b!\",\"adext\":{\"callout\":{\"t\":\"\\u304d\\u3081\\u306e\\u7d30\\u304b\\u3044\\u6307\\u5c0e \\u00b7 \\u7d4c\\u9a13\\u3068\\u60c5\\u71b1\\u306b\\u3042\\u3075\\u308c\\u308b\\u8b1b\\u5e2b \\u00b7 \\u77ed\\u671f\\u9593\\u30b9\\u30b3\\u30a2UP\\u6cd5\\u3092\\u63d0\\u4f9b \\u00b7 \\u5c11\\u4eba\\u6570\\u306e\\u5b9f\\u8df5\\u6f14\\u7fd2\",\"tid\":\"4\"},\"filterlinks\":{\"l\":[],\"tid\":\"\"},\"sitelinks\":{\"l\":[{\"targetUrl\":\"https://duckduckgo.com/y.js?ad_domain=agos.co.jp&ad_provider=bingv7aa&ad_type=txad&eddgt=obP7zXz7zpZHDybCoDxesg%3D%3D&rut=0775aea54a76bdc651a07b5b6d9bb0b5a3312b830116a0a5705f920eadce0a8d&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De8YtC6eTayvxICgO_74gaNxDVUCUx3MSqhDzr%2DQGWoWU46I_SK7hUYitxw1bDkwu51N4R%2Dy2n%2DF_Z3diiXwWhVMScvGQLYOHkpzJlogFXTiR4vkRTEI6CXepxoBdTo7ZRbEHQEhLvaxEQc7HHcGBfvrIhV5UmO4EI5CnyEEmhFkhykgShvntmZi4sK2RdNtojVZjcZh_jLwFJhGr6O4xBoHNjWmJk%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuYWdvcy5jby5qcCUyZmluZm9ybWF0aW9uJTJmc291ZGFuLmh0bWwlM2Z1dG1fc291cmNlJTNkYmluZyVjMiVhMCUyNnV0bV9tZWRpdW0lM2RjcGMlMjZ1dG1fY2FtcGFpZ24lM2Rub3Rfc3l1eW91JTI2bXNjbGtpZCUzZDkxNjNmMTQ2ZjQwNTFlNjEwNzdkOTRlNTA5ZTljNTQyJTI2dXRtX3Rlcm0lM2RMTE0lMjZ1dG1fY29udGVudCUzZEtXRF9BTExfQUxfKExMTSk%26rlid%3D9163f146f4051e61077d94e509e9c542&vqd=4-301837748834523973528319863104123825131&iurl=%7B1%7DIG%3D6793BB4EFA564D10A45BB0C3C75BF04C%26CID%3D342CD9B7486D6452145CCDB1496D6555%26ID%3DDevEx%2C5064.1\",\"text\":\"\\u7121\\u6599\\u500b\\u5225\\u76f8\\u8ac7\"},{\"targetUrl\":\"https://duckduckgo.com/y.js?ad_domain=agos.co.jp&ad_provider=bingv7aa&ad_type=txad&eddgt=obP7zXz7zpZHDybCoDxesg%3D%3D&rut=bd3f96c640d64876df3ee14fe038b68503cd63d36a1fd092aa33d3e738d4ed8c&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De8yHdgbe16wB0x%2DcAPgvHZdDVUCUx70tji1zX3o7kwOuVJH76ae5KWFMlI2xXK47X50C5H7blNCrYkTCjolxW9T95HCiAsMjqqwkdd71fw%2Di12GNjj6a_x8xmsYgUv1SoUaHMS9mJF40HnjMi2osart3afpkZ1yVburNNhgqcOwg%2DXMgDWa%2DLVR%2Dt%2DwG_ZxijIuc87rPnYDGWI6M3y7Us1OIVOMRY%26u%3DaHR0cCUzYSUyZiUyZnd3dy5hZ29zLmNvLmpwJTJmcHJvZ3JhbSUyZiUzZnV0bV9zb3VyY2UlM2RiaW5nJWMyJWEwJTI2dXRtX21lZGl1bSUzZGNwYyUyNnV0bV9jYW1wYWlnbiUzZG5vdF9zeXV5b3UlMjZtc2Nsa2lkJTNkYWJjY2UzN2M4N2RmMTZjNjNiNTE4NmRmN2EzM2RiYzUlMjZ1dG1fdGVybSUzZExMTSUyNnV0bV9jb250ZW50JTNkS1dEX0FMTF9BTF8oTExNKQ%26rlid%3Dabcce37c87df16c63b5186df7a33dbc5&vqd=4-135094424682227864277222089879108223323&iurl=%7B1%7DIG%3D6793BB4EFA564D10A45BB0C3C75BF04C%26CID%3D342CD9B7486D6452145CCDB1496D6555%26ID%3DDevEx%2C5066.1\",\"text\":\"\\u30d7\\u30ed\\u30b0\\u30e9\\u30e0\\u7d39\\u4ecb\"},{\"targetUrl\":\"https://duckduckgo.com/y.js?ad_domain=agos.co.jp&ad_provider=bingv7aa&ad_type=txad&eddgt=obP7zXz7zpZHDybCoDxesg%3D%3D&rut=2eb357c941584c2d37f902ddd1c08c859ed01f062ab3fb05a001faf0f72e6ba5&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De8C54ez8bikjRXEwXnJwi9YDVUCUy8nH9%2DJAueBxXkb0K7ZCNC0iUGDL4ho5%2DSoNtD5viVDKN7ZH22nltAGg05iJ0ZuWgdIts%2DGgXJTkql6SPae7Qmp04T63VKixa%2DGPkNHV0IE2WaD1BTEMUqr91ygKbGPrQddZylBRsNDKI6Ohg5xJwTIHBIZLIc%2D2VktajIGg3mr7TywC9C8C404NhOGDS3izs%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuYWdvcy5jby5qcCUyZm9ubGluZXNlcnZpY2VzJTJmbW9kdWxlcyUyZmFnZW5kYXglMmZpbmRleC5waHAlM2ZvcCUzZGNhbCUyNnV0bV9zb3VyY2UlM2RiaW5nJWMyJWEwJTI2dXRtX21lZGl1bSUzZGNwYyUyNnV0bV9jYW1wYWlnbiUzZG5vdF9zeXV5b3UlMjZtc2Nsa2lkJTNkZjY5ZDIyYjcyNmE1MWRhZDQ0MTg0NzE0ZGI0OTk3MWUlMjZ1dG1fdGVybSUzZExMTSUyNnV0bV9jb250ZW50JTNkS1dEX0FMTF9BTF8oTExNKQ%26rlid%3Df69d22b726a51dad44184714db49971e&vqd=4-128396212371085491387335181864088230840&iurl=%7B1%7DIG%3D6793BB4EFA564D10A45BB0C3C75BF04C%26CID%3D342CD9B7486D6452145CCDB1496D6555%26ID%3DDevEx%2C5068.1\",\"text\":\"\\u7121\\u6599\\u30a4\\u30d9\\u30f3\\u30c8\"},{\"targetUrl\":\"https://duckduckgo.com/y.js?ad_domain=agos.co.jp&ad_provider=bingv7aa&ad_type=txad&eddgt=obP7zXz7zpZHDybCoDxesg%3D%3D&rut=479b8c8e06853252a860df6f4f7b0021979d4a7d81cfe31b1fa267ba9bfa146b&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De8NjJGPM715u8VUlRmw8zHUDVUCUw7uCpgewYW%2DimZqp8QeEMcJSAl25ZE2QP%2De2LLi0zNlmdSsmFlczgatCZw3wHznKksYZxIlU8DyVauuq_BlQ7Y_XHQN5FZwp1JVZ62qIViejaQ8jOUUWe5WlcQ_RlJzztAhEBmbtWJsPAdjkkG2_jriD6uq5jjLVwdj7zJvT%2DPdQaxmcV5d1uLOKlfTWWSqcM%26u%3DaHR0cCUzYSUyZiUyZnd3dy5hZ29zLmNvLmpwJTJmdXNlZnVsJTJmJTNmdXRtX3NvdXJjZSUzZGJpbmclYzIlYTAlMjZ1dG1fbWVkaXVtJTNkY3BjJTI2dXRtX2NhbXBhaWduJTNkbm90X3N5dXlvdSUyNm1zY2xraWQlM2Q0NWViYjRlZWE2M2YxZjk0ZTI1YWQxNGQ1YmQzOTFkNSUyNnV0bV90ZXJtJTNkTExNJTI2dXRtX2NvbnRlbnQlM2RLV0RfQUxMX0FMXyhMTE0p%26rlid%3D45ebb4eea63f1f94e25ad14d5bd391d5&vqd=4-297887522221861120640855135832921374674&iurl=%7B1%7DIG%3D6793BB4EFA564D10A45BB0C3C75BF04C%26CID%3D342CD9B7486D6452145CCDB1496D6555%26ID%3DDevEx%2C5070.1\",\"text\":\"\\u7559\\u5b66\\u304a\\u5f79\\u7acb\\u3061\\u60c5\\u5831\"}],\"tid\":\"9\\t11[10]\\t13[12]\\t15[14]\\t17[16]\",\"type\":\"SiteLink\"},\"smart\":{\"t\":\"\\u30b3\\u30fc\\u30b9: TOEFL(R)TEST\\u5bfe\\u7b56\\u30b3\\u30fc\\u30b9, IELTS\\u8a66\\u9a13\\u5bfe\\u7b56\\u30b3\\u30fc\\u30b9, GMAT(R)\\u8a66\\u9a13\\u5bfe\\u7b56\\u30b3\\u30fc\\u30b9\",\"tid\":\"8\"},\"tid\":\"1\"},\"ae\":{\"callout\":[\"\\u304d\\u3081\\u306e\\u7d30\\u304b\\u3044\\u6307\\u5c0e \\u00b7 \\u7d4c\\u9a13\\u3068\\u60c5\\u71b1\\u306b\\u3042\\u3075\\u308c\\u308b\\u8b1b\\u5e2b \\u00b7 \\u77ed\\u671f\\u9593\\u30b9\\u30b3\\u30a2UP\\u6cd5\\u3092\\u63d0\\u4f9b \\u00b7 \\u5c11\\u4eba\\u6570\\u306e\\u5b9f\\u8df5\\u6f14\\u7fd2\"]},\"c\":\"https://duckduckgo.com/y.js?ad_domain=agos.co.jp&ad_provider=bingv7aa&ad_type=txad&eddgt=obP7zXz7zpZHDybCoDxesg%3D%3D&rut=b63afcc493f34d7a7c7d3518e392551bed85e9ae7571c18b4dc955aaf8259616&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De8rhXDjLYuOBw8iJkGNRsNfzVUCUwxbNLujGJigHwGI9U5xO6%2DITqhX%2DRxoJEeElFXLF7C9j%2DxBG6M752LwJ8JIWQMG9aHf9eRSn8J307_mnj%2DzyVlkx3nyY0oZxNIfHP8d_eF8Bl_Gv8mmnjESk_mCDLz9CtFNkGvFdusnGnhhSX20uHcFptCvdD5h78HZ7eC9J8%2DwA%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuYWdvcy5jby5qcCUyZmxhbmQlMmZsbG0lMmYlM2Z1dG1fc291cmNlJTNkYmluZyVjMiVhMCUyNnV0bV9tZWRpdW0lM2RjcGMlMjZ1dG1fY2FtcGFpZ24lM2Rub3Rfc3l1eW91JTI2bXNjbGtpZCUzZDA5ODZlN2Y3OTRiZTE4ZThhNWJjODI1NDY5NTJkZmI1JTI2dXRtX3Rlcm0lM2RMTE0lMjZ1dG1fY29udGVudCUzZEtXRF9BTExfQUxfKExMTSk%26rlid%3D0986e7f794be18e8a5bc82546952dfb5&vqd=4-174227029600136465336090119074482095864&iurl=%7B1%7DIG%3D6793BB4EFA564D10A45BB0C3C75BF04C%26CID%3D342CD9B7486D6452145CCDB1496D6555%26ID%3DDevEx%2C5058.1\",\"d\":\"agos.co.jp\",\"h\":0,\"i\":\"\",\"k\":0,\"m\":0,\"o\":\"\",\"p\":1,\"relevancy\":{\"abstract\":\"%E6%B5%B7%E5%A4%96%E3%83%88%E3%83%83%E3%83%97%E6%A0%A1%E3%81%B810%2C000%E4%BA%BA%E4%BB%A5%E4%B8%8A%E3%81%AE%E5%90%88%E6%A0%BC%E5%AE%9F%E7%B8%BE!%E3%81%BE%E3%81%9A%E3%81%AF%E7%84%A1%E6%96%99%E5%80%8B%E5%88%A5%E7%9B%B8%E8%AB%87%E3%80%81%E7%84%A1%E6%96%99%E3%82%A4%E3%83%99%E3%83%B3%E3%83%88%E3%81%B8.%20%E9%81%B8%E3%81%B9%E3%82%8B%E5%AD%A6%E7%BF%92%E5%BD%A2%E6%85%8B%E3%80%81%E5%8F%97%E8%AC%9B%E6%9C%9F%E9%96%933%E5%B9%B4%E3%80%81%E5%86%8D%E5%8F%97%E8%AC%9B%E7%84%A1%E6%96%99%E3%81%AE%E3%82%A2%E3%82%B4%E3%82%B9%E3%81%A7%E3%82%B0%E3%83%AD%E3%83%BC%E3%83%90%E3%83%AB%E3%83%AC%E3%83%99%E3%83%AB%E3%81%AE%E3%82%AD%E3%83%A3%E3%83%AA%E3%82%A2%E3%82%92%E7%9B%AE%E6%8C%87%E3%81%9B!\",\"adx_name\":\"none\",\"cq_retail\":\"high\",\"is_good_v10\":0,\"q\":\"llm\",\"q_words\":1,\"q_words_fuzzy\":0,\"q_words_in_ad\":\"0\",\"root_domain\":\"agos.co.jp\",\"start\":\"0\",\"title\":\"LLM%E3%81%AE%E3%81%9F%E3%82%81%E3%81%AE%E8%A9%A6%E9%A8%93%E5%AF%BE%E7%AD%96%E3%81%AA%E3%82%89%20%2D%20%E5%85%A8%E5%9B%BD%E3%83%88%E3%83%83%E3%83%97%E3%82%AF%E3%83%A9%E3%82%B9%E3%81%AE%E6%B5%B7%E5%A4%96%E7%95%99%E5%AD%A6%E5%AF%BE%E7%AD%96\"},\"s\":\"bingv7aa\",\"t\":\"LLM\\u306e\\u305f\\u3081\\u306e\\u8a66\\u9a13\\u5bfe\\u7b56\\u306a\\u3089 - \\u5168\\u56fd\\u30c8\\u30c3\\u30d7\\u30af\\u30e9\\u30b9\\u306e\\u6d77\\u5916\\u7559\\u5b66\\u5bfe\\u7b56\",\"tid\":\"1,4,8,9,11[10],13[12],15[14],17[16]\",\"u\":\"https://duckduckgo.com/y.js?ad_domain=agos.co.jp&ad_provider=bingv7aa&ad_type=txad&eddgt=obP7zXz7zpZHDybCoDxesg%3D%3D&rut=b63afcc493f34d7a7c7d3518e392551bed85e9ae7571c18b4dc955aaf8259616&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De8rhXDjLYuOBw8iJkGNRsNfzVUCUwxbNLujGJigHwGI9U5xO6%2DITqhX%2DRxoJEeElFXLF7C9j%2DxBG6M752LwJ8JIWQMG9aHf9eRSn8J307_mnj%2DzyVlkx3nyY0oZxNIfHP8d_eF8Bl_Gv8mmnjESk_mCDLz9CtFNkGvFdusnGnhhSX20uHcFptCvdD5h78HZ7eC9J8%2DwA%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuYWdvcy5jby5qcCUyZmxhbmQlMmZsbG0lMmYlM2Z1dG1fc291cmNlJTNkYmluZyVjMiVhMCUyNnV0bV9tZWRpdW0lM2RjcGMlMjZ1dG1fY2FtcGFpZ24lM2Rub3Rfc3l1eW91JTI2bXNjbGtpZCUzZDA5ODZlN2Y3OTRiZTE4ZThhNWJjODI1NDY5NTJkZmI1JTI2dXRtX3Rlcm0lM2RMTE0lMjZ1dG1fY29udGVudCUzZEtXRF9BTExfQUxfKExMTSk%26rlid%3D0986e7f794be18e8a5bc82546952dfb5&vqd=4-174227029600136465336090119074482095864&iurl=%7B1%7DIG%3D6793BB4EFA564D10A45BB0C3C75BF04C%26CID%3D342CD9B7486D6452145CCDB1496D6555%26ID%3DDevEx%2C5058.1\"}], {\"page_load_url\":\"https://duckduckgo.com/y.js?ifu=%7B3%7Dappid%3D055AAD1BA669BEB8B048128DC89A107C678B527B%26rguid%3D413d05f2a7f54c8f8ad4685aa4a1d7d9&iurl=%7B2%7DIG%3D6793BB4EFA564D10A45BB0C3C75BF04C%26CID%3D342CD9B7486D6452145CCDB1496D6555%26Type%3DEvent.CPT%26DATA%3D0\",\"visibility_url\":\"https://duckduckgo.com/y.js?ivu=%7B4%7Dtype%3Dmv%26reqver%3D1.0%26rg%3D413d05f2a7f54c8f8ad4685aa4a1d7d9\"});DDG.duckbar.future_signal_tab({signal:'medium',from:'deep_answer'});DDG.duckbar.add({\"data\":{\"Abstract\":\"A large language model is a language model notable for its ability to achieve general-purpose language understanding and generation. LLMs acquire these abilities by learning statistical relationships from text documents during a computationally intensive self-supervised and semi-supervised training process. LLMs are artificial neural networks following a transformer architecture. They can be used for text generation by taking an input text and repeatedly predicting the next token or word. Up to 2020, fine tuning was the only way a model could be adapted to be able to accomplish specific tasks. Larger sized models, such as GPT-3, however, can be prompt-engineered to achieve similar results. They are thought to acquire knowledge about syntax, semantics and \\\"ontology\\\" inherent in human language corpora, but also inaccuracies and biases present in the corpora.\",\"AbstractSource\":\"Wikipedia\",\"AbstractText\":\"A large language model is a language model notable for its ability to achieve general-purpose language understanding and generation. LLMs acquire these abilities by learning statistical relationships from text documents during a computationally intensive self-supervised and semi-supervised training process. LLMs are artificial neural networks following a transformer architecture. They can be used for text generation by taking an input text and repeatedly predicting the next token or word. Up to 2020, fine tuning was the only way a model could be adapted to be able to accomplish specific tasks. Larger sized models, such as GPT-3, however, can be prompt-engineered to achieve similar results. They are thought to acquire knowledge about syntax, semantics and \\\"ontology\\\" inherent in human language corpora, but also inaccuracies and biases present in the corpora.\",\"AbstractURL\":\"https://en.wikipedia.org/wiki/Large_language_model\",\"Answer\":\"\",\"AnswerType\":\"\",\"Definition\":\"\",\"DefinitionSource\":\"\",\"DefinitionURL\":\"\",\"Entity\":\"\",\"Heading\":\"Large language model\",\"Image\":\"\",\"ImageHeight\":0,\"ImageIsLogo\":0,\"ImageWidth\":0,\"Infobox\":\"\",\"Redirect\":\"\",\"RelatedTopics\":[{\"FirstURL\":\"https://duckduckgo.com/Foundation_models\",\"Icon\":{\"Height\":\"\",\"URL\":\"\",\"Width\":\"\"},\"Result\":\"Foundation models - A foundation model is an AI model that is trained on broad data such that it can be applied across a wide range of use cases.\",\"Text\":\"Foundation models - A foundation model is an AI model that is trained on broad data such that it can be applied across a wide range of use cases.\"},{\"FirstURL\":\"https://duckduckgo.com/Generative_artificial_intelligence\",\"Icon\":{\"Height\":\"\",\"URL\":\"\",\"Width\":\"\"},\"Result\":\"Generative AI - Generative artificial intelligence is artificial intelligence capable of generating text, images, or other media, using generative models. Generative AI models learn the patterns and structure of their input training data and then generate new data that has similar characteristics.\",\"Text\":\"Generative AI - Generative artificial intelligence is artificial intelligence capable of generating text, images, or other media, using generative models. Generative AI models learn the patterns and structure of their input training data and then generate new data that has similar characteristics.\"},{\"FirstURL\":\"https://duckduckgo.com/c/Deep_learning\",\"Icon\":{\"Height\":\"\",\"URL\":\"\",\"Width\":\"\"},\"Result\":\"Deep learning\",\"Text\":\"Deep learning\"},{\"FirstURL\":\"https://duckduckgo.com/c/Natural_language_processing\",\"Icon\":{\"Height\":\"\",\"URL\":\"\",\"Width\":\"\"},\"Result\":\"Natural language processing\",\"Text\":\"Natural language processing\"}],\"Results\":[],\"Type\":\"A\",\"meta\":{\"attribution\":null,\"blockgroup\":null,\"created_date\":null,\"description\":\"Wikipedia\",\"designer\":null,\"dev_date\":null,\"dev_milestone\":\"live\",\"developer\":[{\"name\":\"DDG Team\",\"type\":\"ddg\",\"url\":\"http://www.duckduckhack.com\"}],\"example_query\":\"nikola tesla\",\"id\":\"wikipedia_fathead\",\"is_stackexchange\":null,\"js_callback_name\":\"wikipedia\",\"live_date\":null,\"maintainer\":{\"github\":\"duckduckgo\"},\"name\":\"Wikipedia\",\"perl_module\":\"DDG::Fathead::Wikipedia\",\"producer\":null,\"production_state\":\"online\",\"repo\":\"fathead\",\"signal_from\":\"wikipedia_fathead\",\"src_domain\":\"en.wikipedia.org\",\"src_id\":1,\"src_name\":\"Wikipedia\",\"src_options\":{\"directory\":\"\",\"is_fanon\":0,\"is_mediawiki\":1,\"is_wikipedia\":1,\"language\":\"en\",\"min_abstract_length\":\"20\",\"skip_abstract\":0,\"skip_abstract_paren\":0,\"skip_end\":\"0\",\"skip_icon\":0,\"skip_image_name\":0,\"skip_qr\":\"\",\"source_skip\":\"\",\"src_info\":\"\"},\"src_url\":null,\"status\":\"live\",\"tab\":\"About\",\"topic\":[\"productivity\"],\"unsafe\":0}},\"duckbar_topic\":\"About\",\"from\":\"deep_answer\",\"meta\":{\"attribution\":null,\"blockgroup\":null,\"created_date\":null,\"description\":\"Wikipedia\",\"designer\":null,\"dev_date\":null,\"dev_milestone\":\"live\",\"developer\":[{\"name\":\"DDG Team\",\"type\":\"ddg\",\"url\":\"http://www.duckduckhack.com\"}],\"example_query\":\"nikola tesla\",\"id\":\"wikipedia_fathead\",\"is_stackexchange\":null,\"js_callback_name\":\"wikipedia\",\"live_date\":null,\"maintainer\":{\"github\":\"duckduckgo\"},\"name\":\"Wikipedia\",\"perl_module\":\"DDG::Fathead::Wikipedia\",\"producer\":null,\"production_state\":\"online\",\"repo\":\"fathead\",\"signal_from\":\"wikipedia_fathead\",\"src_domain\":\"en.wikipedia.org\",\"src_id\":1,\"src_name\":\"Wikipedia\",\"src_options\":{\"directory\":\"\",\"is_fanon\":0,\"is_mediawiki\":1,\"is_wikipedia\":1,\"language\":\"en\",\"min_abstract_length\":\"20\",\"skip_abstract\":0,\"skip_abstract_paren\":0,\"skip_end\":\"0\",\"skip_icon\":0,\"skip_image_name\":0,\"skip_qr\":\"\",\"source_skip\":\"\",\"src_info\":\"\"},\"src_url\":null,\"status\":\"live\",\"tab\":\"About\",\"topic\":[\"productivity\"],\"unsafe\":0},\"model\":\"FatheadArticle\",\"pixel_id\":\"wikipedia_fathead_deep\",\"signal\":\"medium\",\"templates\":{\"detail\":\"info_detail\"}});DDG.deep.signalSummary = \"about:m,retail:h\";DDG.inject('DDG.Data.languages.resultLanguages', {\"en\":[\"https://en.wikipedia.org/wiki/Large_language_model\",\"https://en.wikipedia.org/wiki/Master_of_Laws\",\"https://www.lsac.org/discover-law/types-law-programs/llm-degree-programs\",\"https://llm-guide.com/what-is-an-llm\",\"https://hls.harvard.edu/graduate-program/ll-m-program/\",\"http://llm.lsac.org/\",\"https://hls.harvard.edu/graduate-program/graduate-program-admissions-and-financial-aid/ll-m-admissions/\",\"https://www.usnews.com/education/articles/getting-an-llm-degree-what-to-know\",\"https://law.stanford.edu/office-of-student-affairs/the-master-of-laws-llm-degree/\",\"https://gould.usc.edu/academics/degrees/online-llm/\",\"https://www.lawyeredu.org/LLM-degree/\",\"https://www.law.nyu.edu/llmjsd/master-of-laws\",\"https://gould.usc.edu/academics/degrees/llm/\",\"https://www.techtarget.com/whatis/definition/large-language-model-LLM\",\"https://aws.amazon.com/what-is/large-language-model/\",\"https://www.computerworld.com/article/3697649/what-are-large-language-models-and-how-are-they-used-in-generative-ai.html\",\"https://graduate.northeastern.edu/program/master-of-laws-llm-online-17868/\",\"https://www.law.northwestern.edu/academics/degree-programs/llms/\",\"https://www.gartner.com/en/information-technology/glossary/large-language-models-llm\",\"https://en.wikipedia.org/wiki/Wikipedia:Large_language_models\",\"https://www.elastic.co/what-is/large-language-models\",\"https://www.geeksforgeeks.org/large-language-model-llm/\",\"https://developers.google.com/machine-learning/resources/intro-llms\"]});DDG.deep.pageLayoutSummary = \"a1w5dic1w18r1,e1\";DDG.inject('DDG.Data.languages.adLanguages', {});if (DDG.pageLayout) DDG.pageLayout.load('d',[{\"a\":\"A large language model (LLM) is a language model notable for its ability to achieve general-purpose language understanding and generation. LLMs acquire these abilities by learning statistical relationships from text documents during a computationally intensive self-supervised and semi-supervised training process. LLMs are artificial neural networks following a transformer architecture.\",\"ae\":null,\"b\":\"w\\tWikipedia\\ten.wikipedia.org\",\"c\":\"https://en.wikipedia.org/wiki/Large_language_model\",\"d\":\"en.wikipedia.org/wiki/Large_language_model\",\"da\":\"en_wikipedia_queries,nlp_fathead,nlp_wiki\",\"h\":0,\"i\":\"en.wikipedia.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Large language model - Wikipedia\",\"u\":\"https://en.wikipedia.org/wiki/Large_language_model\"},{\"a\":\"A Master of Laws (M.L. or LL.M.; Latin: Magister Legum or Legum Magister) is an advanced postgraduate academic degree, pursued by those either holding an undergraduate academic law degree, a professional law degree, or an undergraduate degree in a related subject.In most jurisdictions, the LL.M. is the advanced professional degree for those usually already admitted into legal practice.\",\"ae\":null,\"b\":\"w\\tWikipedia\\ten.wikipedia.org\",\"c\":\"https://en.wikipedia.org/wiki/Master_of_Laws\",\"d\":\"en.wikipedia.org/wiki/Master_of_Laws\",\"da\":\"en_wikipedia_queries,nlp_fathead,nlp_wiki\",\"h\":0,\"i\":\"en.wikipedia.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Master of Laws - Wikipedia\",\"u\":\"https://en.wikipedia.org/wiki/Master_of_Laws\"},{\"a\":\"An LLM, or Master of Laws, is a graduate qualification in the field of law. The LLM was created for lawyers to expand their knowledge, study a specialized area of law, and gain international qualifications if they have earned a law degree outside the U.S. or Canada. If you're looking to advance your legal career or take the next step in your ...\",\"ae\":null,\"c\":\"https://www.lsac.org/discover-law/types-law-programs/llm-degree-programs\",\"d\":\"www.lsac.org/discover-law/types-law-programs/llm-degree-programs\",\"da\":\"\",\"h\":0,\"i\":\"www.lsac.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"LLM Degree | Masters of Laws | The Law School Admission Council\",\"u\":\"https://www.lsac.org/discover-law/types-law-programs/llm-degree-programs\"},{\"a\":\"The LLM - short for Master of Laws - is an internationally recognized postgraduate law degree that is usually completed in one year of full-time studies. It's different from a JD or an LLB, which are first law degrees and are generally required to practice law. Specialized LLMs can be found in tax law, business law, and other subjects.\",\"ae\":null,\"c\":\"https://llm-guide.com/what-is-an-llm\",\"d\":\"llm-guide.com/what-is-an-llm\",\"da\":\"\",\"h\":0,\"i\":\"llm-guide.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"What is an LL.M.? | LLM GUIDE\",\"u\":\"https://llm-guide.com/what-is-an-llm\"},{\"a\":\"Learn about the LL.M. (Master of Laws) program at Harvard Law School, a one-year degree program for students from various legal systems and backgrounds. Find out the degree requirements, academic resources, and class profile of the LL.M. students.\",\"ae\":null,\"c\":\"https://hls.harvard.edu/graduate-program/ll-m-program/\",\"d\":\"hls.harvard.edu/graduate-program/ll-m-program/\",\"da\":\"\",\"h\":0,\"i\":\"hls.harvard.edu\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"LL.M. Program - Harvard Law School | Harvard Law School\",\"u\":\"https://hls.harvard.edu/graduate-program/ll-m-program/\"},{\"a\":\"LSAC offers services for various types of law programs offered by ABA-approved law schools, such as LLM, MCL, MLS, JM, MSLS, JSD, SJD, and DCL. Sign up now to search, apply, and credential assemble for over 130 law programs.\",\"ae\":null,\"c\":\"http://llm.lsac.org/\",\"d\":\"llm.lsac.org\",\"da\":\"\",\"h\":0,\"i\":\"llm.lsac.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Welcome to LLM & Other Law Programs | Law School Admission Council\",\"u\":\"http://llm.lsac.org/\"},{\"a\":\"Learn about the eligibility, criteria, and application process for the one-year LL.M. (Master of Laws) program at Harvard Law School, which typically includes 180 students from some 70 countries. Find out the tuition and financial aid options, and see sample applications and FAQs.\",\"ae\":null,\"c\":\"https://hls.harvard.edu/graduate-program/graduate-program-admissions-and-financial-aid/ll-m-admissions/\",\"d\":\"hls.harvard.edu/graduate-program/graduate-program-admissions-and-financial-aid/ll-m-admissions/\",\"da\":\"\",\"h\":0,\"i\":\"hls.harvard.edu\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"LL.M. Admissions - Harvard Law School | Harvard Law School\",\"u\":\"https://hls.harvard.edu/graduate-program/graduate-program-admissions-and-financial-aid/ll-m-admissions/\"},{\"a\":\"An LL.M. is geared towards those whom either have a J.D. degree and want to gain additional training in areas such as tax law or health-care law, or those who earned a degree outside of the U.S ...\",\"ae\":null,\"c\":\"https://www.usnews.com/education/articles/getting-an-llm-degree-what-to-know\",\"d\":\"www.usnews.com/education/articles/getting-an-llm-degree-what-to-know\",\"da\":\"\",\"e\":\"2022-12-28T00:00:00.0000000\",\"h\":0,\"i\":\"www.usnews.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Getting an LL.M. Degree: What to Know | Education | U.S. News\",\"u\":\"https://www.usnews.com/education/articles/getting-an-llm-degree-what-to-know\"},{\"a\":\"To obtain an LLM degree, students must complete at least 35 but no more than 45 approved quarter units of course work. At least 26 of these units must be in Law School courses; however, see below for the policies and limitations on enrolling in courses from elsewhere in the University, and see the section on the California or New York bar exam for special unit requirements for students ...\",\"ae\":null,\"c\":\"https://law.stanford.edu/office-of-student-affairs/the-master-of-laws-llm-degree/\",\"d\":\"law.stanford.edu/office-of-student-affairs/the-master-of-laws-llm-degree/\",\"da\":\"\",\"h\":0,\"i\":\"law.stanford.edu\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"The Master of Laws (LLM) Degree | Stanford Law School\",\"u\":\"https://law.stanford.edu/office-of-student-affairs/the-master-of-laws-llm-degree/\"},{\"a\":\"Earn a Master of Laws degree from a top-ranked law school in the U.S. with a part-time, flexible and interdisciplinary curriculum. Learn from world-class faculty, seasoned academics and policymakers, and join the global Trojan Family network of more than 15,000 law school alumni.\",\"ae\":null,\"c\":\"https://gould.usc.edu/academics/degrees/online-llm/\",\"d\":\"gould.usc.edu/academics/degrees/online-llm/\",\"da\":\"\",\"h\":0,\"i\":\"gould.usc.edu\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Master of Laws (LLM) - Online - USC Gould School of Law\",\"u\":\"https://gould.usc.edu/academics/degrees/online-llm/\"},{\"a\":\"LLM in Taxation. The Master of Laws (LLM) is the degree of choice for career advancement and international credibility, particularly in today's competitive and globally focused legal environment. Early- and mid-career lawyers pursue the LLM voluntarily when looking to expand their proficiency in a specific area of law.\",\"ae\":null,\"c\":\"https://www.lawyeredu.org/LLM-degree/\",\"d\":\"www.lawyeredu.org/LLM-degree/\",\"da\":\"\",\"h\":0,\"i\":\"www.lawyeredu.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"What is an LLM | What is a Master of Laws - Lawyeredu.org\",\"u\":\"https://www.lawyeredu.org/LLM-degree/\"},{\"a\":\"Design Your Own LLM. You will choose from 300+ courses to plan a curriculum that meets your intellectual and professional interests. You can choose to specialize in one or two areas, or take a broad range of classes. You also will have the chance to write a paper in close consultation with a professor, or expand a typical research assignment into a master's thesis.\",\"ae\":null,\"c\":\"https://www.law.nyu.edu/llmjsd/master-of-laws\",\"d\":\"www.law.nyu.edu/llmjsd/master-of-laws\",\"da\":\"\",\"h\":0,\"i\":\"www.law.nyu.edu\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Master of Laws (LLM) | NYU School of Law - New York University\",\"u\":\"https://www.law.nyu.edu/llmjsd/master-of-laws\"},{\"a\":\"Learn about the Master of Laws (LLM) degree programs at USC Gould School of Law, which focus on the U.S. legal system and prepare students for leadership roles in law. Choose from various formats, eligibility criteria, and specialization tracks to suit your goals and interests.\",\"ae\":null,\"c\":\"https://gould.usc.edu/academics/degrees/llm/\",\"d\":\"gould.usc.edu/academics/degrees/llm/\",\"da\":\"\",\"h\":0,\"i\":\"gould.usc.edu\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Master of Laws (LLM) Degree Programs | USC Gould School of Law\",\"u\":\"https://gould.usc.edu/academics/degrees/llm/\"},{\"a\":\"A large language model (LLM) is a type of artificial intelligence ( AI) algorithm that uses deep learning techniques and massively large data sets to understand, summarize, generate and predict new content. The term generative AI also is closely connected with LLMs, which are, in fact, a type of generative AI that has been specifically ...\",\"ae\":null,\"b\":\"whatis\\tWhatIs.com\\twhatis.techtarget.com\",\"c\":\"https://www.techtarget.com/whatis/definition/large-language-model-LLM\",\"d\":\"www.techtarget.com/whatis/definition/large-language-model-LLM\",\"da\":\"\",\"h\":0,\"i\":\"www.techtarget.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"What are Large Language Models? | Definition from TechTarget\",\"u\":\"https://www.techtarget.com/whatis/definition/large-language-model-LLM\"},{\"a\":\"Large language models (LLM) are very large deep learning models that are pre-trained on vast amounts of data. The underlying transformer is a set of neural networks that consist of an encoder and a decoder with self-attention capabilities. The encoder and decoder extract meanings from a sequence of text and understand the relationships between words and phrases in it.\",\"ae\":null,\"b\":\"a\\tAmazon.com\\twww.amazon.com\",\"c\":\"https://aws.amazon.com/what-is/large-language-model/\",\"d\":\"aws.amazon.com/what-is/large-language-model/\",\"da\":\"products\",\"h\":0,\"i\":\"aws.amazon.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"What are Large Language Models? - LLM AI Explained - AWS\",\"u\":\"https://aws.amazon.com/what-is/large-language-model/\"},{\"a\":\"The LLM could come back with "cereal," or "rice," or "steak tartare." There's no 100% right answer, but there is a probability based on the data already ingested in the model. The ...\",\"ae\":null,\"c\":\"https://www.computerworld.com/article/3697649/what-are-large-language-models-and-how-are-they-used-in-generative-ai.html\",\"d\":\"www.computerworld.com/article/3697649/what-are-large-language-models-and-how-are-they-used-in-generative-ai.html\",\"da\":\"\",\"e\":\"2023-05-30T10:00:00.0000000\",\"h\":0,\"i\":\"www.computerworld.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"What are LLMs, and how are they used in generative AI?\",\"u\":\"https://www.computerworld.com/article/3697649/what-are-large-language-models-and-how-are-they-used-in-generative-ai.html\"},{\"a\":\"Northeastern University offers a Master of Laws (LLM) program with a 100% online learning format option designed for internationally trained lawyers and U.S.-trained lawyers to enhance their practical skills and foundational knowledge of the ever-changing U.S. legal environment, and the global practice of law. The online LLM program positions students to take advantage of Northeastern ...\",\"ae\":null,\"c\":\"https://graduate.northeastern.edu/program/master-of-laws-llm-online-17868/\",\"d\":\"graduate.northeastern.edu/program/master-of-laws-llm-online-17868/\",\"da\":\"\",\"h\":0,\"i\":\"graduate.northeastern.edu\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Master of Laws LLM-Online - Graduate Programs\",\"u\":\"https://graduate.northeastern.edu/program/master-of-laws-llm-online-17868/\"},{\"a\":\"LLM Programs. Our LLM (Master of Law) degree programs expand students' knowledge of law and legal processes and provide opportunities for them to gain expertise in a specialized field of law. To apply, students must have a JD from an ABA-accredited law school or a comparable legal degree from a university outside of the United States.\",\"ae\":null,\"c\":\"https://www.law.northwestern.edu/academics/degree-programs/llms/\",\"d\":\"www.law.northwestern.edu/academics/degree-programs/llms/\",\"da\":\"\",\"h\":0,\"i\":\"www.law.northwestern.edu\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"LLM Programs - Northwestern University Pritzker School of Law\",\"u\":\"https://www.law.northwestern.edu/academics/degree-programs/llms/\"},{\"a\":\"Large Language Models (LLMs) A large language model (LLM) is a specialized type of artificial intelligence (AI) that has been trained on vast amounts of text to understand existing content and generate original content.\",\"ae\":null,\"c\":\"https://www.gartner.com/en/information-technology/glossary/large-language-models-llm\",\"d\":\"www.gartner.com/en/information-technology/glossary/large-language-models-llm\",\"da\":\"\",\"h\":0,\"i\":\"www.gartner.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Large Language Models (LLMs) - Gartner\",\"u\":\"https://www.gartner.com/en/information-technology/glossary/large-language-models-llm\"},{\"a\":\"Large language models have limited reliability, limited understanding, limited range, and hence need human supervision. While large language models (colloquially termed "AI chatbots" in some contexts) can be very useful, machine-generated text (much like human-generated text) can contain errors or flaws, or be outright useless.\",\"ae\":null,\"b\":\"w\\tWikipedia\\ten.wikipedia.org\",\"c\":\"https://en.wikipedia.org/wiki/Wikipedia:Large_language_models\",\"d\":\"en.wikipedia.org/wiki/Wikipedia:Large_language_models\",\"da\":\"en_wikipedia_queries,nlp_fathead,nlp_wiki\",\"h\":0,\"i\":\"en.wikipedia.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Wikipedia:Large language models - Wikipedia\",\"u\":\"https://en.wikipedia.org/wiki/Wikipedia:Large_language_models\"},{\"a\":\"Large language model definition. A large language model (LLM) is a deep learning algorithm that can perform a variety of natural language processing (NLP) tasks. Large language models use transformer models and are trained using massive datasets \\u2014 hence, large. This enables them to recognize, translate, predict, or generate text or other content.\",\"ae\":null,\"c\":\"https://www.elastic.co/what-is/large-language-models\",\"d\":\"www.elastic.co/what-is/large-language-models\",\"da\":\"\",\"h\":0,\"i\":\"www.elastic.co\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"What is a large language model (LLM)? - Elastic\",\"u\":\"https://www.elastic.co/what-is/large-language-models\"},{\"a\":\"Difference Between NLP and LLM NLP is Natural Language Processing, a field of artificial intelligence (AI). It consists of the development of the algorithms. NLP is a broader field than LLM, which consists of algorithms and techniques. NLP rules two approaches i.e. Machine learning and the analyze language data. Applications of NLP are-\",\"ae\":null,\"c\":\"https://www.geeksforgeeks.org/large-language-model-llm/\",\"d\":\"www.geeksforgeeks.org/large-language-model-llm/\",\"da\":\"\",\"e\":\"2024-01-10T00:00:00.0000000\",\"h\":0,\"i\":\"www.geeksforgeeks.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"What is a Large Language Model (LLM) - GeeksforGeeks\",\"u\":\"https://www.geeksforgeeks.org/large-language-model-llm/\"},{\"a\":\"Define key LLM concepts, including Transformers and self-attention. Describe the costs and benefits of LLMs, along with common use cases. What is a language model? A language model is a machine learning model that aims to predict and generate plausible language. Autocomplete is a language model, for example.\",\"ae\":null,\"c\":\"https://developers.google.com/machine-learning/resources/intro-llms\",\"d\":\"developers.google.com/machine-learning/resources/intro-llms\",\"da\":\"\",\"e\":\"2023-08-08T00:00:00.0000000\",\"h\":0,\"i\":\"developers.google.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Introduction to Large Language Models - Google Developers\",\"u\":\"https://developers.google.com/machine-learning/resources/intro-llms\"},{\"n\":\"/d.js?q=llm&kl=wt-wt&l=wt-wt&p=&s=23&ex=-1&ct=US&sp=0&vqd=4-36212277936736004277629252433802891730\"}]);DDG.duckbar.load('images');DDG.duckbar.load('news');DDG.duckbar.load('videos');DDG.duckbar.loadModule('related_searches', {\"ads\":[],\"query\":\"llm\",\"queryEncoded\":\"llm\",\"response_type\":\"places\",\"results\":[{\"display_text\":\"what does #llm mean\",\"text\":\"what does #llm mean\",\"web_search_url\":\"?q=what%20does%20%23llm%20mean\"},{\"display_text\":\"llm meaning\",\"text\":\"llm meaning\",\"web_search_url\":\"?q=llm%20meaning\"},{\"display_text\":\"llm meaning in law\",\"text\":\"llm meaning in law\",\"web_search_url\":\"?q=llm%20meaning%20in%20law\"},{\"display_text\":\"what is llm stand for\",\"text\":\"what is llm stand for\",\"web_search_url\":\"?q=what%20is%20llm%20stand%20for\"},{\"display_text\":\"llm in artificial intelligence\",\"text\":\"llm in artificial intelligence\",\"web_search_url\":\"?q=llm%20in%20artificial%20intelligence\"},{\"display_text\":\"full meaning of llm\",\"text\":\"full meaning of llm\",\"web_search_url\":\"?q=full%20meaning%20of%20llm\"},{\"display_text\":\"what is llm in law\",\"text\":\"what is llm in law\",\"web_search_url\":\"?q=what%20is%20llm%20in%20law\"},{\"display_text\":\"examples of llm models\",\"text\":\"examples of llm models\",\"web_search_url\":\"?q=examples%20of%20llm%20models\"}],\"vqd\":{\"llm\":\"4-36212277936736004277629252433802891730\"}});if (DDG.pageLayout) DDG.pageLayout.initialize({\"mainline\":{\"items\":[[\"ad\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"dictionary_definition\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"related_searches\"]]},\"sidebar\":{\"items\":[[\"wikipedia_fathead\"]]}}, { start: 0 });DDG.deep.emit(\"load:completed\");", + "curl-cffi-POST-https://duckduckgo.com-{\"data\": {\"q\": \"MetaGPT use cases\"}}": "MetaGPT use cases at DuckDuckGo
", + "curl-cffi-GET-https://links.duckduckgo.com/d.js-{\"params\": {\"bing_market\": \"wt-WT\", \"df\": null, \"ex\": \"-1\", \"kl\": \"wt-wt\", \"l\": \"wt-wt\", \"q\": \"MetaGPT use cases\", \"s\": \"0\", \"sp\": \"0\", \"vqd\": \"4-206455801954364851330794682843954609879\"}}": "if (DDG.deep && DDG.deep.setUpstream) DDG.deep.setUpstream(\"bingv7aa\");DDG.deep.bn={'ivc':1};if (DDG.pageLayout) DDG.pageLayout.load('a',[], {\"page_load_url\":\"https://duckduckgo.com/y.js?iurl=%7B2%7DIG%3DA7C157D7FB464F86BD78A7B80D28A7BC%26CID%3D2B7157406A406D2B1C8943466B3F6C14%26Type%3DEvent.CPT%26DATA%3D0\"});DDG.deep.signalSummary = \"\";DDG.inject('DDG.Data.languages.resultLanguages', {\"en\":[\"https://incubity.ambilio.com/metagpt-deep-dive-into-multi-agent-system-with-use-cases/\",\"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\"https://ts2.pl/en/metagpt-in-action-use-cases-across-industries/\",\"https://geekflare.com/metagpt-multi-agent-framework/\",\"https://docs.deepwisdom.ai/main/en/guide/get_started/introduction.html\",\"https://www.straight.com/guides/software/a-complete-guide-to-metagpt-the-best-ai-agent-available-now/\",\"https://aibusiness.com/nlp/metagpt-text-to-app-ai-simplifies-web-dev\",\"https://levelup.gitconnected.com/metagpt-the-future-of-multi-agent-collaboration-in-ai-a-brief-guide-fd4b4429336d\",\"https://github.com/geekan/MetaGPT\",\"https://docs.deepwisdom.ai/main/en/guide/get_started/quickstart.html\",\"https://medium.com/mlearning-ai/metagpt-multi-agent-harmony-for-complex-problem-solving-97bcb8f3fe94\",\"https://docs.deepwisdom.ai/enus/guide/tutorials/concepts.html\",\"https://medium.com/aimonks/metagpt-a-framework-for-multi-agent-meta-programming-6c79f2eafb8e\",\"https://docs.deepwisdom.ai/main/en/guide/use_cases/agent/researcher.html\",\"https://github.com/PlaiD3/MetaGPT/blob/main/README.md\",\"https://generativeai.pub/analyzing-an-exciting-generative-ai-research-called-metagpt-2106385312db\",\"https://medium.com/technology-hits/autogpt-langchain-deep-lake-metagpt-a-revolutionary-framework-for-building-advanced-ai-e2c579d86494\",\"https://gpt3demo.com/apps/metagpt\",\"https://docs.deepwisdom.ai/main/en/guide/tutorials/use_memories.html\",\"https://smythos.com/ai-agents/agent-comparison/metagpt-vs-autogen/\",\"https://analyticsindiamag.com/metagpt-realising-the-gpt-4-dream/\",\"https://docs.deepwisdom.ai/main/en/guide/use_cases/agent/tutorial_assistant.html\",\"https://www.washingtonpost.com/technology/2024/01/04/nyt-ai-copyright-lawsuit-fair-use/\",\"https://blog.netwrix.com/2024/01/09/azure-storage/\",\"https://www.bloomberg.com/news/articles/2024-01-09/walmart-wmt-expands-rollout-of-generative-ai-shopping-search-tech\",\"https://health.ny.gov/press/releases/2024/docs/2024-01-08_masking_advisory.pdf\",\"https://www.bloomberg.com/news/articles/2024-01-10/lloyds-bank-manager-awarded-450-000-after-winning-case-over-racist-slur\",\"https://www.washingtonpost.com/world/2024/01/10/south-africa-israel-icj-genocide-case/\"]});DDG.deep.pageLayoutSummary = \"w29\";DDG.inject('DDG.Data.languages.adLanguages', {});if (DDG.pageLayout) DDG.pageLayout.load('d',[{\"a\":\"Some high-value business use cases where MetaGPT could be applied include: Software Development and Engineering: MetaGPT can streamline the software development lifecycle by orchestrating roles like Product Managers, Architects, Engineers, and QA Engineers. It can assist in requirements gathering, design, code generation, testing, and debugging ...\",\"ae\":null,\"c\":\"https://incubity.ambilio.com/metagpt-deep-dive-into-multi-agent-system-with-use-cases/\",\"d\":\"incubity.ambilio.com/metagpt-deep-dive-into-multi-agent-system-with-use-cases/\",\"da\":\"\",\"h\":0,\"i\":\"incubity.ambilio.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Deep Dive into Multi-Agent System with Use Cases\",\"u\":\"https://incubity.ambilio.com/metagpt-deep-dive-into-multi-agent-system-with-use-cases/\"},{\"a\":\"Published 4 months ago on September 11, 2023 By Aayush Mittal With Large Language Models (LLMs) like ChatGPT, OpenAI has witnessed a surge in enterprise and user adoption, currently raking in around $80 million in monthly revenue.\",\"ae\":null,\"c\":\"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\"d\":\"www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\"da\":\"\",\"e\":\"2023-09-11T00:00:00.0000000\",\"h\":0,\"i\":\"www.unite.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Complete Guide to the Best AI Agent Available Right Now\",\"u\":\"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\"},{\"a\":\"MetaGPT in Action: Use-cases Across Industries MetaGPT, a powerful language model developed by OpenAI, has been making waves across various industries due to its versatility and ability to generate human-like text. As artificial intelligence (AI) continues to advance, the potential applications of MetaGPT are becoming increasingly apparent.\",\"ae\":null,\"c\":\"https://ts2.pl/en/metagpt-in-action-use-cases-across-industries/\",\"d\":\"ts2.pl/en/metagpt-in-action-use-cases-across-industries/\",\"da\":\"\",\"e\":\"2023-06-12T00:00:00.0000000\",\"h\":0,\"i\":\"ts2.pl\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT in Action: Use-cases Across Industries\",\"u\":\"https://ts2.pl/en/metagpt-in-action-use-cases-across-industries/\"},{\"a\":\"MetaGPT is a multi-agent framework that takes one-line inputs to produce APIs, user stories, data structures, competitive analysis, and more. GPT is the short form for Generative Pretrained Transformers. MetaGPT framework can behave as a product manager, software engineer, and architect.\",\"ae\":null,\"c\":\"https://geekflare.com/metagpt-multi-agent-framework/\",\"d\":\"geekflare.com/metagpt-multi-agent-framework/\",\"da\":\"\",\"e\":\"2023-09-18T00:00:00.0000000\",\"h\":0,\"i\":\"geekflare.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Is This the Best Multi-Agent Framework Yet? - Geekflare\",\"u\":\"https://geekflare.com/metagpt-multi-agent-framework/\"},{\"a\":\"Software Company Multi-Role Schematic MetaGPT's Abilities MetaGPT started as a software company, but its capabilities are not limited to that. You can use this multi-agent framework in your own scenario to build your own application. For details, you can refer to Researcher under Use Cases. Let's do it. Examples (fully generated by GPT-4)\",\"ae\":null,\"c\":\"https://docs.deepwisdom.ai/main/en/guide/get_started/introduction.html\",\"d\":\"docs.deepwisdom.ai/main/en/guide/get_started/introduction.html\",\"da\":\"\",\"h\":0,\"i\":\"docs.deepwisdom.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: The Multi-Agent Framework | MetaGPT\",\"u\":\"https://docs.deepwisdom.ai/main/en/guide/get_started/introduction.html\"},{\"a\":\"MetaGPT is a multi-agent system that utilizes Large Language Models (LLMs) to perform complex tasks. ... MetaGPT has demonstrated its capabilities in various use cases, including developing a CLI ...\",\"ae\":null,\"c\":\"https://www.straight.com/guides/software/a-complete-guide-to-metagpt-the-best-ai-agent-available-now/\",\"d\":\"www.straight.com/guides/software/a-complete-guide-to-metagpt-the-best-ai-agent-available-now/\",\"da\":\"\",\"e\":\"2023-12-13T00:00:00.0000000\",\"h\":0,\"i\":\"www.straight.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"A Complete Guide to MetaGPT: The Best AI Agent Available Now\",\"u\":\"https://www.straight.com/guides/software/a-complete-guide-to-metagpt-the-best-ai-agent-available-now/\"},{\"a\":\"MetaGPT builds microapps - applications designed for specific tasks or use cases. Examples include Facebook Messenger, the project management app Trello, and even Microsoft Word. It only generates web apps - which can be viewed on mobile or desktop browsers but won't run as native apps on Android or iOS.\",\"ae\":null,\"c\":\"https://aibusiness.com/nlp/metagpt-text-to-app-ai-simplifies-web-dev\",\"d\":\"aibusiness.com/nlp/metagpt-text-to-app-ai-simplifies-web-dev\",\"da\":\"\",\"e\":\"2023-08-07T00:00:00.0000000\",\"h\":0,\"i\":\"aibusiness.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Text-To-App AI Simplifies Web Dev\",\"u\":\"https://aibusiness.com/nlp/metagpt-text-to-app-ai-simplifies-web-dev\"},{\"a\":\"Here's a simple example of how to use MetaGPT: python startup.py "Write a cli snake game" # Use code review will cost more money, but will opt for better code quality. python startup.py "Write a cli snake game" --code_review True. ... Over the last few months, we have looked into around 100 agents with various use cases, studied SDKs and ...\",\"ae\":null,\"c\":\"https://levelup.gitconnected.com/metagpt-the-future-of-multi-agent-collaboration-in-ai-a-brief-guide-fd4b4429336d\",\"d\":\"levelup.gitconnected.com/metagpt-the-future-of-multi-agent-collaboration-in-ai-a-brief-guide-fd4b4429336d\",\"da\":\"\",\"e\":\"2023-08-09T00:00:00.0000000\",\"h\":0,\"i\":\"levelup.gitconnected.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: The Future of Multi-Agent Collaboration in AI (A Brief Guide)\",\"u\":\"https://levelup.gitconnected.com/metagpt-the-future-of-multi-agent-collaboration-in-ai-a-brief-guide-fd4b4429336d\"},{\"a\":\"MetaGPT: The Multi-Agent Framework Assign different roles to GPTs to form a collaborative software entity for complex tasks. MetaGPT takes a one line requirement as input and outputs user stories / competitive analysis / requirements / data structures / APIs / documents, etc.\",\"ae\":null,\"b\":\"gh\\tGitHub\\tgithub.com\",\"c\":\"https://github.com/geekan/MetaGPT\",\"d\":\"github.com/geekan/MetaGPT\",\"da\":\"\",\"h\":0,\"i\":\"github.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: The Multi-Agent Framework - GitHub\",\"u\":\"https://github.com/geekan/MetaGPT\"},{\"a\":\"You can check this by using:</span>\n<span class=\"pl-c\"><span class=\"pl-c\">#</span> You can use conda to initialize a new python env</span>\n<span class=\"pl-c\"><span class=\"pl-c\">#</span> conda create -n metagpt python=3.9</span>\n<span class=\"pl-c\"><span class=\"pl-c\">#</span> conda activate metagpt</span>\npython3 --version\n\n<span...\",\"ae\":null,\"b\":\"gh\\tGitHub\\tgithub.com\",\"c\":\"https://github.com/geekan/MetaGPT?search=1\",\"d\":\"github.com/geekan/MetaGPT?search=1\",\"da\":\"\",\"h\":0,\"i\":\"github.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: The Multi-Agent Framework - GitHub\",\"u\":\"https://github.com/geekan/MetaGPT?search=1\"},{\"a\":\"Now, let's get started! We will create a team of agents to write software based on one line of our instruction. First, import off-the-shelf roles. python. import asyncio from metagpt.roles import ( Architect, Engineer, ProductManager, ProjectManager, ) from metagpt.team import Team. Next, initiate the team, equip it with agents, set their ...\",\"ae\":null,\"c\":\"https://docs.deepwisdom.ai/main/en/guide/get_started/quickstart.html\",\"d\":\"docs.deepwisdom.ai/main/en/guide/get_started/quickstart.html\",\"da\":\"\",\"h\":0,\"i\":\"docs.deepwisdom.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Quickstart | MetaGPT\",\"u\":\"https://docs.deepwisdom.ai/main/en/guide/get_started/quickstart.html\"},{\"a\":\"Stefan Silver \\u00b7 Follow Published in MLearning.ai \\u00b7 4 min read \\u00b7 Aug 9 4 Photo by Penfer on Unsplash Lately, there's been quite a buzz around automating problem-solving using multiagents...\",\"ae\":null,\"c\":\"https://medium.com/mlearning-ai/metagpt-multi-agent-harmony-for-complex-problem-solving-97bcb8f3fe94\",\"d\":\"medium.com/mlearning-ai/metagpt-multi-agent-harmony-for-complex-problem-solving-97bcb8f3fe94\",\"da\":\"\",\"e\":\"2023-08-09T00:00:00.0000000\",\"h\":0,\"i\":\"medium.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Multi-Agent Harmony for Complex Problem Solving\",\"u\":\"https://medium.com/mlearning-ai/metagpt-multi-agent-harmony-for-complex-problem-solving-97bcb8f3fe94\"},{\"a\":\"Concepts. After this tutorial, you will be able to: Understand MetaGPT's concept of agent and environment. How agents interact with each other and what a multi-agent collaboration may look like. The goal is to provide an intuitive and simplified explanation of the concepts so that users have a background to further explore the tutorial series.\",\"ae\":null,\"c\":\"https://docs.deepwisdom.ai/enus/guide/tutorials/concepts.html\",\"d\":\"docs.deepwisdom.ai/enus/guide/tutorials/concepts.html\",\"da\":\"\",\"h\":0,\"i\":\"docs.deepwisdom.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Concepts | MetaGPT\",\"u\":\"https://docs.deepwisdom.ai/enus/guide/tutorials/concepts.html\"},{\"a\":\"Capabilities/Use Case of MetaGPT MetaGPT has many potential applications and use cases in various fields and scenarios that involve multi-agent collaboration and coordination. Some of...\",\"ae\":null,\"c\":\"https://medium.com/aimonks/metagpt-a-framework-for-multi-agent-meta-programming-6c79f2eafb8e\",\"d\":\"medium.com/aimonks/metagpt-a-framework-for-multi-agent-meta-programming-6c79f2eafb8e\",\"da\":\"\",\"e\":\"2023-08-03T00:00:00.0000000\",\"h\":0,\"i\":\"medium.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: A Framework for Multi-Agent Meta Programming\",\"u\":\"https://medium.com/aimonks/metagpt-a-framework-for-multi-agent-meta-programming-6c79f2eafb8e\"},{\"a\":\"The metagpt.roles.researcher module provides a command-line interface for executing the functionalities of the Researcher. An example is as follows: bash. python3 -m metagpt.roles.researcher "dataiku vs. datarobot". Log output: log.txt Report output: dataiku vs. datarobot.md.\",\"ae\":null,\"c\":\"https://docs.deepwisdom.ai/main/en/guide/use_cases/agent/researcher.html\",\"d\":\"docs.deepwisdom.ai/main/en/guide/use_cases/agent/researcher.html\",\"da\":\"\",\"h\":0,\"i\":\"docs.deepwisdom.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Researcher: Search Web and Write Reports | MetaGPT\",\"u\":\"https://docs.deepwisdom.ai/main/en/guide/use_cases/agent/researcher.html\"},{\"a\":\"You can check this by using:</span>\npython --version\n\n<span class=\"pl-c\"><span class=\"pl-c\">#</span> Step 3: Clone the repository to your local machine, and install it.</span>\ngit clone https://github.com/geekan/metagpt\n<span class=\"pl-c1\">cd</span> metagpt\npython setup.py install</pre></div>\n<h3 tabindex=\"-1\" dir=\"auto\"><a id=\...\",\"ae\":null,\"b\":\"gh\\tGitHub\\tgithub.com\",\"c\":\"https://github.com/PlaiD3/MetaGPT/blob/main/README.md\",\"d\":\"github.com/PlaiD3/MetaGPT/blob/main/README.md\",\"da\":\"\",\"h\":0,\"i\":\"github.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Multi-Agent Meta Programming Framework - GitHub\",\"u\":\"https://github.com/PlaiD3/MetaGPT/blob/main/README.md\"},{\"a\":\"MetaGPT, as a cutting-edge framework, is not just a theoretical marvel but has been tested, showcasing its prowess in real-world applications. ... These articles cover a wide range of topics related to Generative AI, from introductions and use cases to exploring its potential and understanding its underlying layers. Happy reading!\",\"ae\":null,\"c\":\"https://generativeai.pub/analyzing-an-exciting-generative-ai-research-called-metagpt-2106385312db\",\"d\":\"generativeai.pub/analyzing-an-exciting-generative-ai-research-called-metagpt-2106385312db\",\"da\":\"translations\",\"e\":\"2023-08-14T00:00:00.0000000\",\"h\":0,\"i\":\"generativeai.pub\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Analyzing an exciting Generative AI research called MetaGPT.\",\"u\":\"https://generativeai.pub/analyzing-an-exciting-generative-ai-research-called-metagpt-2106385312db\"},{\"a\":\"Here are 10 compelling use cases that demonstrate the vast potential of LangChain: Uses-Cases of LangChain. 1. Conversational AI and Chatbots ... MetaGPT, or multimodal Generative Pretrained ...\",\"ae\":null,\"c\":\"https://medium.com/technology-hits/autogpt-langchain-deep-lake-metagpt-a-revolutionary-framework-for-building-advanced-ai-e2c579d86494\",\"d\":\"medium.com/technology-hits/autogpt-langchain-deep-lake-metagpt-a-revolutionary-framework-for-building-advanced-ai-e2c579d86494\",\"da\":\"\",\"e\":\"2023-08-28T00:00:00.0000000\",\"h\":0,\"i\":\"medium.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"AutoGPT \\u2014 LangChain \\u2014 Deep Lake \\u2014 MetaGPT: A ... - Medium\",\"u\":\"https://medium.com/technology-hits/autogpt-langchain-deep-lake-metagpt-a-revolutionary-framework-for-building-advanced-ai-e2c579d86494\"},{\"a\":\"MetaGPT takes a one-line requirement as input and outputs user stories / competitive analysis/requirements/data structures / APIs / documents, etc. Internally, MetaGPT includes product managers/architects/project managers/engineers. It provides the entire process of a software company along with carefully orchestrated SOPs.\",\"ae\":null,\"c\":\"https://gpt3demo.com/apps/metagpt\",\"d\":\"gpt3demo.com/apps/metagpt\",\"da\":\"\",\"h\":0,\"i\":\"gpt3demo.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT | Discover AI use cases - GPT-3 Demo\",\"u\":\"https://gpt3demo.com/apps/metagpt\"},{\"a\":\"Retrieve memory. When recorded memories are needed, such as serving as context for a LLM call, you can use self.get_memories. The function definition is as follows: python. def get_memories(self, k=0) -> list [Message]: """A wrapper to return the most recent k memories of this role, return all when k=0""" return self.rc.memory.get (k=k) For ...\",\"ae\":null,\"c\":\"https://docs.deepwisdom.ai/main/en/guide/tutorials/use_memories.html\",\"d\":\"docs.deepwisdom.ai/main/en/guide/tutorials/use_memories.html\",\"da\":\"\",\"h\":0,\"i\":\"docs.deepwisdom.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Use Memories | MetaGPT\",\"u\":\"https://docs.deepwisdom.ai/main/en/guide/tutorials/use_memories.html\"},{\"a\":\"6 Conclusion Introduction Are you struggling to choose between MetaGPT Vs AutoGen? Comparing these two leading companies can help you make an informed decision. MetaGPT is a powerful tool designed for software developers, project managers, startups, technology companies, and AI enthusiasts.\",\"ae\":null,\"c\":\"https://smythos.com/ai-agents/agent-comparison/metagpt-vs-autogen/\",\"d\":\"smythos.com/ai-agents/agent-comparison/metagpt-vs-autogen/\",\"da\":\"\",\"h\":0,\"i\":\"smythos.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT Vs AutoGen: A Comprehensive Comparison\",\"u\":\"https://smythos.com/ai-agents/agent-comparison/metagpt-vs-autogen/\"},{\"a\":\"While some are just wrappers of OpenAI's APIs with added functionality like Forefront.ai or AnonChatGPT, others, like MemeCam or Bing Chat use the GPT-4 API to facilitate new use-cases altogether. OpenAI now needs to move faster, or risk their dream being stolen by others who are on the bleeding edge. Anirudh VK\",\"ae\":null,\"c\":\"https://analyticsindiamag.com/metagpt-realising-the-gpt-4-dream/\",\"d\":\"analyticsindiamag.com/metagpt-realising-the-gpt-4-dream/\",\"da\":\"\",\"e\":\"2023-04-26T00:00:00.0000000\",\"h\":0,\"i\":\"analyticsindiamag.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT \\u2014 Realising the GPT-4 Dream - Analytics India Magazine\",\"u\":\"https://analyticsindiamag.com/metagpt-realising-the-gpt-4-dream/\"},{\"a\":\"Override the _act method. The _act method is responsible for executing the action.Use todo = self.rc.todo to get the next action to be executed from the context, and then execute the run method of the action.Here, it first obtains the tutorial directory structure through WriteDirectory, then chunks the directory, generates a WriteContent action for each chunk, and initializes the newly added ...\",\"ae\":null,\"c\":\"https://docs.deepwisdom.ai/main/en/guide/use_cases/agent/tutorial_assistant.html\",\"d\":\"docs.deepwisdom.ai/main/en/guide/use_cases/agent/tutorial_assistant.html\",\"da\":\"\",\"h\":0,\"i\":\"docs.deepwisdom.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Tutorial Assistant: Generate technology tutorial | MetaGPT\",\"u\":\"https://docs.deepwisdom.ai/main/en/guide/use_cases/agent/tutorial_assistant.html\"},{\"a\":\"AI's future could hinge on one thorny legal question. A lawsuit accuses OpenAI and Microsoft of violating the New York Times's copyright. But the law is anything but clear. By Will Oremus. and ...\",\"ae\":null,\"c\":\"https://www.washingtonpost.com/technology/2024/01/04/nyt-ai-copyright-lawsuit-fair-use/\",\"d\":\"www.washingtonpost.com/technology/2024/01/04/nyt-ai-copyright-lawsuit-fair-use/\",\"da\":\"news,translations\",\"e\":\"2024-01-04T12:01:54.0000000\",\"h\":0,\"i\":\"www.washingtonpost.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"AI copyright lawsuit hinges on the legal concept of 'fair use' - The ...\",\"u\":\"https://www.washingtonpost.com/technology/2024/01/04/nyt-ai-copyright-lawsuit-fair-use/\"},{\"a\":\"Here are some common use cases for Azure Table Storage: Centralized storage of logs, telemetry data and monitoring data. Storage of catalog and shopping cart data for e-commerce applications. Scalable task scheduling and metadata storage. Storage of sensory data and IoT telemetry data.\",\"ae\":null,\"c\":\"https://blog.netwrix.com/2024/01/09/azure-storage/\",\"d\":\"blog.netwrix.com/2024/01/09/azure-storage/\",\"da\":\"translations\",\"e\":\"2024-01-09T00:00:00.0000000\",\"h\":0,\"i\":\"blog.netwrix.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Understanding Six Popular Azure Storage Types and Their Use Cases\",\"u\":\"https://blog.netwrix.com/2024/01/09/azure-storage/\"},{\"a\":\"January 10, 2024 at 8:10 AM PST. Walmart Inc. opened up access to a generative artificial intelligence tool that allows shoppers to search for products by specific use cases, rather than look up ...\",\"ae\":null,\"c\":\"https://www.bloomberg.com/news/articles/2024-01-09/walmart-wmt-expands-rollout-of-generative-ai-shopping-search-tech\",\"d\":\"www.bloomberg.com/news/articles/2024-01-09/walmart-wmt-expands-rollout-of-generative-ai-shopping-search-tech\",\"da\":\"news,translations\",\"e\":\"2024-01-09T16:10:00.0000000\",\"h\":0,\"i\":\"www.bloomberg.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Walmart Expands Rollout of Generative AI Shopping Search, Tech\",\"u\":\"https://www.bloomberg.com/news/articles/2024-01-09/walmart-wmt-expands-rollout-of-generative-ai-shopping-search-tech\"},{\"a\":\"The purpose of this advisory is to alert healthcare providers and facilities to substantial increases in cases of influenza and COVID-19, at least partially driven by an emerging SARS-CoV-2 variant, and to recommend that healthcare and residential facilities advocate strongly for the use of masks within their facility to prevent transmission\",\"ae\":null,\"c\":\"https://health.ny.gov/press/releases/2024/docs/2024-01-08_masking_advisory.pdf\",\"d\":\"health.ny.gov/press/releases/2024/docs/2024-01-08_masking_advisory.pdf\",\"da\":\"translations\",\"e\":\"2024-01-08T00:00:00.0000000\",\"h\":0,\"i\":\"health.ny.gov\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"PDF Health Advisory: Nys Department of Health Recommends Masking in ...\",\"u\":\"https://health.ny.gov/press/releases/2024/docs/2024-01-08_masking_advisory.pdf\"},{\"a\":\"2:29. The ex- Lloyds Banking Group Plc manager who won his unfair dismissal case over his use of a racist slur was awarded more than \\u00a3450,000 ($572,560) from an employment tribunal that said he ...\",\"ae\":null,\"c\":\"https://www.bloomberg.com/news/articles/2024-01-10/lloyds-bank-manager-awarded-450-000-after-winning-case-over-racist-slur\",\"d\":\"www.bloomberg.com/news/articles/2024-01-10/lloyds-bank-manager-awarded-450-000-after-winning-case-over-racist-slur\",\"da\":\"news,translations\",\"e\":\"2024-01-10T13:25:00.0000000\",\"h\":0,\"i\":\"www.bloomberg.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Lloyds Bank Manager Awarded \\u00a3450,000 After Winning Case Over Racist ...\",\"u\":\"https://www.bloomberg.com/news/articles/2024-01-10/lloyds-bank-manager-awarded-450-000-after-winning-case-over-racist-slur\"},{\"a\":\"The ICJ case adds to international pressure on Israel to scale back or end its war against Hamas, which health officials in Gaza say has killed more than 23,000 people \\u2014 many of them women and ...\",\"ae\":null,\"c\":\"https://www.washingtonpost.com/world/2024/01/10/south-africa-israel-icj-genocide-case/\",\"d\":\"www.washingtonpost.com/world/2024/01/10/south-africa-israel-icj-genocide-case/\",\"da\":\"news,translations\",\"e\":\"2024-01-10T22:24:00.0000000\",\"h\":0,\"i\":\"www.washingtonpost.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"What to know about the genocide case against Israel at the ICJ\",\"u\":\"https://www.washingtonpost.com/world/2024/01/10/south-africa-israel-icj-genocide-case/\"},{\"n\":\"/d.js?q=MetaGPT%20use%20cases&kl=wt-wt&l=wt-wt&p=&s=29&ex=-1&ct=US&sp=0&vqd=4-206455801954364851330794682843954609879\"}]);DDG.duckbar.load('images');DDG.duckbar.load('news');DDG.duckbar.load('videos');DDG.duckbar.loadModule('related_searches');if (DDG.pageLayout) DDG.pageLayout.initialize({\"mainline\":{\"items\":[[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"]]}}, { start: 0 });DDG.deep.emit(\"load:completed\");", + "curl-cffi-POST-https://duckduckgo.com-{\"data\": {\"q\": \"The roadmap of MetaGPT\"}}": "The roadmap of MetaGPT at DuckDuckGo
", + "curl-cffi-GET-https://links.duckduckgo.com/d.js-{\"params\": {\"bing_market\": \"wt-WT\", \"df\": null, \"ex\": \"-1\", \"kl\": \"wt-wt\", \"l\": \"wt-wt\", \"q\": \"The roadmap of MetaGPT\", \"s\": \"0\", \"sp\": \"0\", \"vqd\": \"4-25941261128344049410840372626152530092\"}}": "if (DDG.deep && DDG.deep.setUpstream) DDG.deep.setUpstream(\"bingv7aa\");DDG.deep.bn={'ivc':1};if (DDG.pageLayout) DDG.pageLayout.load('a',[], {\"page_load_url\":\"https://duckduckgo.com/y.js?iurl=%7B2%7DIG%3DF478763C5197469DB7B1E366C8182CF2%26CID%3D192D84C7D0B167C5000690C1D1D466C6%26Type%3DEvent.CPT%26DATA%3D0\"});DDG.deep.signalSummary = \"\";DDG.inject('DDG.Data.languages.resultLanguages', {\"en\":[\"https://github.com/geekan/MetaGPT/blob/main/docs/ROADMAP.md\",\"https://github.com/geekan/MetaGPT\",\"https://www.almabetter.com/bytes/articles/metagpt\",\"https://lablab.ai/blog/this-week-in-ai-exploring-the-latest-from-metagpt-and-gpt4-and-more\",\"https://arxiv.org/abs/2308.00352\",\"https://github.com/geekan/MetaGPT/blob/main/README.md\",\"https://github.com/PlaiD3/MetaGPT/blob/main/README.md\",\"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\"https://www.linkedin.com/pulse/metagpt-important-conceptual-advance-multi-agent-systems-brad-edwards\",\"https://www.straight.com/guides/software/a-complete-guide-to-metagpt-the-best-ai-agent-available-now/\",\"https://generativeai.pub/analyzing-an-exciting-generative-ai-research-called-metagpt-2106385312db\",\"https://docs.deepwisdom.ai/main/en/guide/get_started/introduction.html\",\"https://geekflare.com/metagpt-multi-agent-framework/\",\"https://www.marktechpost.com/2023/08/09/meet-metagpt-the-open-source-ai-framework-that-transforms-gpts-into-engineers-architects-and-managers/\",\"https://www.louisbouchard.ai/metagpt/\",\"https://pypi.org/project/metagpt/\",\"https://www.reddit.com/r/ChatGPT/comments/14qhn00/metagpt_the_roadmap_has_been_released_come_and/\",\"https://medium.com/aimonks/metagpt-a-framework-for-multi-agent-meta-programming-6c79f2eafb8e\",\"https://xthemadgenius.medium.com/how-to-use-metagpt-to-operate-as-a-full-engineering-team-c0f6e53c1dc3\",\"https://medium.com/@smraiyyan/metagpt-unleashed-crafting-your-virtual-software-company-from-scratch-6ea60cd70da1\",\"https://github.com/Ditto190/MetaGPT/blob/main/docs/ROADMAP.md\"],\"zh-CN\":[\"https://zhuanlan.zhihu.com/p/677608276\"]});DDG.deep.pageLayoutSummary = \"w1i1w4v1w18\";DDG.inject('DDG.Data.languages.adLanguages', {});if (DDG.pageLayout) DDG.pageLayout.load('d',[{\"a\":\"MetaGPT is an open source framework for building innovative AI powered applications with minimal coding. It leverages the power of GPT-3 and other models to generate various software artifacts from natural language inputs. Learn how to use MetaGPT and contribute to its development in this roadmap.\",\"ae\":null,\"b\":\"gh\\tGitHub\\tgithub.com\",\"c\":\"https://github.com/geekan/MetaGPT/blob/main/docs/ROADMAP.md\",\"d\":\"github.com/geekan/MetaGPT/blob/main/docs/ROADMAP.md\",\"da\":\"\",\"h\":0,\"i\":\"github.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Roadmap - GitHub\",\"u\":\"https://github.com/geekan/MetaGPT/blob/main/docs/ROADMAP.md\"},{\"a\":\"MetaGPT: The Multi-Agent Framework Assign different roles to GPTs to form a collaborative software entity for complex tasks. MetaGPT takes a one line requirement as input and outputs user stories / competitive analysis / requirements / data structures / APIs / documents, etc.\",\"ae\":null,\"b\":\"gh\\tGitHub\\tgithub.com\",\"c\":\"https://github.com/geekan/MetaGPT\",\"d\":\"github.com/geekan/MetaGPT\",\"da\":\"\",\"h\":0,\"i\":\"github.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: The Multi-Agent Framework - GitHub\",\"u\":\"https://github.com/geekan/MetaGPT\"},{\"a\":\"Understanding MetaGPT MetaGPT, a concept originating from a research paper that received significant attention, represents a leap forward in Artificial Intelligence, specifically in multi-agent collaboration using large language models (LLMs).\",\"ae\":null,\"c\":\"https://www.almabetter.com/bytes/articles/metagpt\",\"d\":\"www.almabetter.com/bytes/articles/metagpt\",\"da\":\"\",\"e\":\"2023-08-28T00:00:00.0000000\",\"h\":0,\"i\":\"www.almabetter.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: The Future of Multi-Agent Collaboration in AI\",\"u\":\"https://www.almabetter.com/bytes/articles/metagpt\"},{\"a\":\"MetaGPT is a groundbreaking multi-agent framework that is transforming the way software development is approached. By taking a single line of requirement as input, MetaGPT outputs a comprehensive array of development components, including user stories, competitive analysis, requirements, data structures, APIs, and documents.\",\"ae\":null,\"c\":\"https://lablab.ai/blog/this-week-in-ai-exploring-the-latest-from-metagpt-and-gpt4-and-more\",\"d\":\"lablab.ai/blog/this-week-in-ai-exploring-the-latest-from-metagpt-and-gpt4-and-more\",\"da\":\"\",\"e\":\"2023-08-11T00:00:00.0000000\",\"h\":0,\"i\":\"lablab.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"This Week in AI: Exploring the Latest from MetaGPT and GPT-4 and more..\",\"u\":\"https://lablab.ai/blog/this-week-in-ai-exploring-the-latest-from-metagpt-and-gpt4-and-more\"},{\"a\":\"MetaGPT utilizes an assembly line paradigm to assign diverse roles to various agents, efficiently breaking down complex tasks into subtasks involving many agents working together. On collaborative software engineering benchmarks, MetaGPT generates more coherent solutions than previous chat-based multi-agent systems.\",\"ae\":null,\"b\":\"arx\\tarXiv.org\\tarxiv.org\",\"c\":\"https://arxiv.org/abs/2308.00352\",\"d\":\"arxiv.org/abs/2308.00352\",\"da\":\"translations\",\"e\":\"2023-08-01T00:00:00.0000000\",\"h\":0,\"i\":\"arxiv.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework\",\"u\":\"https://arxiv.org/abs/2308.00352\"},{\"a\":\"MetaGPT takes a one line requirement as input and outputs user stories / competitive analysis / requirements / data structures / APIs / documents, etc. \n Internally, MetaGPT includes product managers / architects / project managers / engineers.\",\"ae\":null,\"b\":\"gh\\tGitHub\\tgithub.com\",\"c\":\"https://github.com/geekan/MetaGPT/blob/main/README.md\",\"d\":\"github.com/geekan/MetaGPT/blob/main/README.md\",\"da\":\"\",\"h\":0,\"i\":\"github.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: The Multi-Agent Framework - GitHub\",\"u\":\"https://github.com/geekan/MetaGPT/blob/main/README.md\"},{\"a\":\"arXiv.org\",\"ae\":null,\"b\":\"arx\\tarXiv.org\\tarxiv.org\",\"c\":\"https://arxiv.org/pdf/2308.00352.pdf\",\"d\":\"arxiv.org/pdf/2308.00352.pdf\",\"da\":\"translations\",\"h\":0,\"i\":\"arxiv.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"PDF arXiv.org\",\"u\":\"https://arxiv.org/pdf/2308.00352.pdf\"},{\"a\":\"MetaGPT takes a one line requirement as input and outputs user stories / competitive analysis / requirements / data structures / APIs / documents, etc. \n; Internally, MetaGPT includes product managers / architects / project managers / engineers. It provides the entire process of a software company along with carefully orchestrated SOPs.\n \n\",\"ae\":null,\"b\":\"gh\\tGitHub\\tgithub.com\",\"c\":\"https://github.com/PlaiD3/MetaGPT/blob/main/README.md\",\"d\":\"github.com/PlaiD3/MetaGPT/blob/main/README.md\",\"da\":\"\",\"h\":0,\"i\":\"github.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Multi-Agent Meta Programming Framework - GitHub\",\"u\":\"https://github.com/PlaiD3/MetaGPT/blob/main/README.md\"},{\"a\":\"MetaGPT's architecture is divided into two layers: the Foundational Components Layer and the Collaboration Layer. Foundational Components Layer: This layer focuses on individual agent operations and facilitates system-wide information exchange. It introduces core building blocks such as Environment, Memory, Roles, Actions, and Tools.\",\"ae\":null,\"c\":\"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\"d\":\"www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\"da\":\"\",\"e\":\"2023-09-11T00:00:00.0000000\",\"h\":0,\"i\":\"www.unite.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Complete Guide to the Best AI Agent Available Right Now\",\"u\":\"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\"},{\"a\":\"Published Aug 6, 2023 + Follow Recent advances in large language models (LLMs) have opened up new opportunities for developing intelligent software agents capable of replicating human-level...\",\"ae\":null,\"b\":\"li\\tLinkedIn\\twww.linkedin.com\",\"c\":\"https://www.linkedin.com/pulse/metagpt-important-conceptual-advance-multi-agent-systems-brad-edwards\",\"d\":\"www.linkedin.com/pulse/metagpt-important-conceptual-advance-multi-agent-systems-brad-edwards\",\"da\":\"\",\"e\":\"2023-08-06T00:00:00.0000000\",\"h\":0,\"i\":\"www.linkedin.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Important Conceptual Advance in Multi-Agent Systems - LinkedIn\",\"u\":\"https://www.linkedin.com/pulse/metagpt-important-conceptual-advance-multi-agent-systems-brad-edwards\"},{\"a\":\"1. Enhanced Operational Efficiency. MetaGPT is designed to store, retrieve, and share information at varying levels, reducing redundancy and enhancing operational efficiency. This means that ...\",\"ae\":null,\"c\":\"https://www.straight.com/guides/software/a-complete-guide-to-metagpt-the-best-ai-agent-available-now/\",\"d\":\"www.straight.com/guides/software/a-complete-guide-to-metagpt-the-best-ai-agent-available-now/\",\"da\":\"\",\"e\":\"2023-12-13T00:00:00.0000000\",\"h\":0,\"i\":\"www.straight.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"A Complete Guide to MetaGPT: The Best AI Agent Available Now\",\"u\":\"https://www.straight.com/guides/software/a-complete-guide-to-metagpt-the-best-ai-agent-available-now/\"},{\"a\":\"The MetaGPT approach showcases its ability to decompose highlevel tasks into detailed actionable components handled by distinct roles (ProductManager, Architect, ProjectManager, Engineer, QA Engineer), thereby facilitating role-specific expertise and coordination. This methodology mirrors human software development teams.\",\"ae\":null,\"c\":\"https://generativeai.pub/analyzing-an-exciting-generative-ai-research-called-metagpt-2106385312db\",\"d\":\"generativeai.pub/analyzing-an-exciting-generative-ai-research-called-metagpt-2106385312db\",\"da\":\"translations\",\"e\":\"2023-08-14T00:00:00.0000000\",\"h\":0,\"i\":\"generativeai.pub\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Analyzing an exciting Generative AI research called MetaGPT.\",\"u\":\"https://generativeai.pub/analyzing-an-exciting-generative-ai-research-called-metagpt-2106385312db\"},{\"a\":\"Internally, MetaGPT includes product managers / architects / project managers / engineers. It provides the entire process of a software company along with carefully orchestrated SOPs. Code = SOP (Team) is the core philosophy. We materialize SOP and apply it to teams composed of LLMs. Software Company Multi-Role Schematic.\",\"ae\":null,\"c\":\"https://docs.deepwisdom.ai/main/en/guide/get_started/introduction.html\",\"d\":\"docs.deepwisdom.ai/main/en/guide/get_started/introduction.html\",\"da\":\"\",\"h\":0,\"i\":\"docs.deepwisdom.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: The Multi-Agent Framework | MetaGPT\",\"u\":\"https://docs.deepwisdom.ai/main/en/guide/get_started/introduction.html\"},{\"a\":\"MetaGPT is a multi-agent framework that takes one-line inputs to produce APIs, user stories, data structures, competitive analysis, and more. GPT is the short form for Generative Pretrained Transformers. MetaGPT framework can behave as a product manager, software engineer, and architect. This framework can act as an entire software company with ...\",\"ae\":null,\"c\":\"https://geekflare.com/metagpt-multi-agent-framework/\",\"d\":\"geekflare.com/metagpt-multi-agent-framework/\",\"da\":\"\",\"e\":\"2023-09-18T00:00:00.0000000\",\"h\":0,\"i\":\"geekflare.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Is This the Best Multi-Agent Framework Yet? - Geekflare\",\"u\":\"https://geekflare.com/metagpt-multi-agent-framework/\"},{\"a\":\"MetaGPT manages far more software complexity than GPT-3.5 or other open-source frameworks like AutoGPT and AgentVerse, measured by lines of produced code. Additionally, MetaGPT generates high-quality requirement papers, design artifacts, flowcharts, and interface specifications throughout the automated end-to-end process. ...\",\"ae\":null,\"c\":\"https://www.marktechpost.com/2023/08/09/meet-metagpt-the-open-source-ai-framework-that-transforms-gpts-into-engineers-architects-and-managers/\",\"d\":\"www.marktechpost.com/2023/08/09/meet-metagpt-the-open-source-ai-framework-that-transforms-gpts-into-engineers-architects-and-managers/\",\"da\":\"translations\",\"e\":\"2023-08-09T00:00:00.0000000\",\"h\":0,\"i\":\"www.marktechpost.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Meet MetaGPT: The Open-Source AI Framework That Transforms GPTs into ...\",\"u\":\"https://www.marktechpost.com/2023/08/09/meet-metagpt-the-open-source-ai-framework-that-transforms-gpts-into-engineers-architects-and-managers/\"},{\"a\":\"MetaGPT is a new paper and open-source work that is making a lot of noise on GitHub! The researchers developed a new framework for combining or chaining large language models and mitigating hallucination risks by integrating human standardized operating procedures (SOPs) into the chaining process. This new design scheme allows the system to ...\",\"ae\":null,\"c\":\"https://www.louisbouchard.ai/metagpt/\",\"d\":\"www.louisbouchard.ai/metagpt/\",\"da\":\"\",\"e\":\"2023-08-27T00:00:00.0000000\",\"h\":0,\"i\":\"www.louisbouchard.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Mitigating AI Hallucinations: Exploring MetaGPT's Collaborative Framework\",\"u\":\"https://www.louisbouchard.ai/metagpt/\"},{\"a\":\"MetaGPT takes a one line requirement as input and outputs user stories / competitive analysis / requirements / data structures / APIs / documents, etc. Internally, MetaGPT includes product managers / architects / project managers / engineers. It provides the entire process of a software company along with carefully orchestrated SOPs.\",\"ae\":null,\"c\":\"https://pypi.org/project/metagpt/\",\"d\":\"pypi.org/project/metagpt/\",\"da\":\"\",\"e\":\"2024-01-10T00:00:00.0000000\",\"h\":0,\"i\":\"pypi.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"metagpt \\u00b7 PyPI\",\"u\":\"https://pypi.org/project/metagpt/\"},{\"a\":\"Hey u/embessoaat, if your post is a ChatGPT conversation screenshot, please reply with the conversation link or prompt. Thanks! We have a public discord server.There's a free Chatgpt bot, Open Assistant bot (Open-source model), AI image generator bot, Perplexity AI bot, \\ud83e\\udd16 GPT-4 bot (Now with Visual capabilities (cloud vision)!) and channel for latest prompts.\",\"ae\":null,\"b\":\"r\\tReddit\\twww.reddit.com\",\"c\":\"https://www.reddit.com/r/ChatGPT/comments/14qhn00/metagpt_the_roadmap_has_been_released_come_and/\",\"d\":\"www.reddit.com/r/ChatGPT/comments/14qhn00/metagpt_the_roadmap_has_been_released_come_and/\",\"da\":\"translations\",\"h\":0,\"i\":\"www.reddit.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: The roadmap has been released! Come and take a look ... - Reddit\",\"u\":\"https://www.reddit.com/r/ChatGPT/comments/14qhn00/metagpt_the_roadmap_has_been_released_come_and/\"},{\"a\":\"Business: MetaGPT can be used to create and execute business programs that can optimize or automate various processes, such as scheduling, planning, budgeting, marketing, etc. MetaGPT can also...\",\"ae\":null,\"c\":\"https://medium.com/aimonks/metagpt-a-framework-for-multi-agent-meta-programming-6c79f2eafb8e\",\"d\":\"medium.com/aimonks/metagpt-a-framework-for-multi-agent-meta-programming-6c79f2eafb8e\",\"da\":\"\",\"e\":\"2023-08-03T00:00:00.0000000\",\"h\":0,\"i\":\"medium.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: A Framework for Multi-Agent Meta Programming\",\"u\":\"https://medium.com/aimonks/metagpt-a-framework-for-multi-agent-meta-programming-6c79f2eafb8e\"},{\"a\":\"MetaGPT is an innovative solution that allows us to assign different roles to GPTs, forging a collaborative software force. In this guide, we'll explore how to harness the power of MetaGPT for...\",\"ae\":null,\"c\":\"https://xthemadgenius.medium.com/how-to-use-metagpt-to-operate-as-a-full-engineering-team-c0f6e53c1dc3\",\"d\":\"xthemadgenius.medium.com/how-to-use-metagpt-to-operate-as-a-full-engineering-team-c0f6e53c1dc3\",\"da\":\"translations\",\"e\":\"2023-08-12T00:00:00.0000000\",\"h\":0,\"i\":\"xthemadgenius.medium.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"How to use MetaGPT to Operate as a full Engineering Team\",\"u\":\"https://xthemadgenius.medium.com/how-to-use-metagpt-to-operate-as-a-full-engineering-team-c0f6e53c1dc3\"},{\"a\":\"MetaGPT, available on Github (crossed 13,000 stars), aims to change the way we make software.This exciting tool can take a single line of what you want to do and turn it into many things like user ...\",\"ae\":null,\"c\":\"https://medium.com/@smraiyyan/metagpt-unleashed-crafting-your-virtual-software-company-from-scratch-6ea60cd70da1\",\"d\":\"medium.com/@smraiyyan/metagpt-unleashed-crafting-your-virtual-software-company-from-scratch-6ea60cd70da1\",\"da\":\"translations\",\"e\":\"2023-08-07T00:00:00.0000000\",\"h\":0,\"i\":\"medium.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT Lets You Create Your Own Virtual Software Company from ... - Medium\",\"u\":\"https://medium.com/@smraiyyan/metagpt-unleashed-crafting-your-virtual-software-company-from-scratch-6ea60cd70da1\"},{\"a\":\"\\u57282023\\u5e7412\\u670819\\u65e5\\u65f6\\uff0c\\u542c\\u4e86\\u6797\\u4e49\\u7ae0\\u8001\\u5e08\\u5173\\u4e8e"\\u57fa\\u4e8eMetaGPT\\u8fdb\\u884c\\u667a\\u80fd\\u4f53\\u5f00\\u53d1"\\u7684\\u8bb2\\u5ea7\\uff1a \\u89c9\\u5f97\\u65b0\\u5947\\u6709\\u8da3\\uff0c\\u5982\\u679c\\u80fd\\u8fd9\\u6837\\u5728\\u5de5\\u4f5c\\u751f\\u6d3b\\u4e2d\\u5b8c\\u6210\\u81ea\\u5df1\\u7684\\u4efb\\u52a1\\uff0c\\u90a3\\u7b80\\u76f4\\u662f\\u4e8b\\u534a\\u529f\\u500d\\u3002\\u4e8e\\u662f\\u8fd9\\u4e24\\u5929\\u53c8\\u5b66\\u4e60\\u4e86\\u300aMetaGPT\\u667a\\u80fd\\u4f53\\u5f00\\u53d1\\u5165\\u95e8\\u300b\\u6559\\u2026\",\"ae\":null,\"c\":\"https://zhuanlan.zhihu.com/p/677608276\",\"d\":\"zhuanlan.zhihu.com/p/677608276\",\"da\":\"translations\",\"e\":\"2024-01-12T00:00:00.0000000\",\"h\":0,\"i\":\"zhuanlan.zhihu.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"\\u5b66\\u4e60\\u7b14\\u8bb0-\\u300aMetaGPT\\u667a\\u80fd\\u4f53\\u5f00\\u53d1\\u5165\\u95e8\\u300b\\u6559\\u7a0b - \\u77e5\\u4e4e\",\"u\":\"https://zhuanlan.zhihu.com/p/677608276\"},{\"a\":\"Roadmap \n Long-term Objective \n. Enable MetaGPT to self-evolve, accomplishing self-training, fine-tuning, optimization, utilization, and updates. \n Short-term Objective \n \n; Become the multi-agent framework with the highest ROI. \n; Support fully automatic implementation of medium-sized projects (around 2000 lines of code). \n\",\"ae\":null,\"b\":\"gh\\tGitHub\\tgithub.com\",\"c\":\"https://github.com/Ditto190/MetaGPT/blob/main/docs/ROADMAP.md\",\"d\":\"github.com/Ditto190/MetaGPT/blob/main/docs/ROADMAP.md\",\"da\":\"translations\",\"h\":0,\"i\":\"github.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Roadmap - GitHub\",\"u\":\"https://github.com/Ditto190/MetaGPT/blob/main/docs/ROADMAP.md\"},{\"n\":\"/d.js?q=The%20roadmap%20of%20MetaGPT&kl=wt-wt&l=wt-wt&p=&s=23&ex=-1&ct=US&sp=0&vqd=4-25941261128344049410840372626152530092\"}]);DDG.duckbar.load('images', {\"ads\":[],\"query\":\"The roadmap of MetaGPT\",\"queryEncoded\":\"The%20roadmap%20of%20MetaGPT\",\"response_type\":\"places\",\"results\":[{\"height\":720,\"image\":\"https://i.ytimg.com/vi/8cxLdYtwx4M/maxresdefault.jpg\",\"image_token\":\"25476bee58d891e0b100edecfc9022b60c5c458e8d078f04908c2fe341449aad\",\"source\":\"Bing\",\"thumbnail\":\"https://tse1.mm.bing.net/th?id=OIP.Sbc4rHZKxrQ7JJKFfx4pggHaEK&pid=Api\",\"thumbnail_token\":\"b8ca3fc5d5109341b5fa2a52479d696225049ce7a678a01730e9440c44454ef0\",\"title\":\"How to use metagpt || What is MetaGPT || Meta AI Tool - YouTube\",\"url\":\"https://www.youtube.com/watch?v=8cxLdYtwx4M\",\"width\":1280},{\"height\":1699,\"image\":\"https://cdn-cashy-static-assets.lucidchart.com/lucidspark/marketing/blog/2020Q4/product-roadmap/product-roadmap-example.png\",\"image_token\":\"45bef86b333d99b8975de0658ef5da261356dcfea369bc8c529fe98c02e9508f\",\"source\":\"Bing\",\"thumbnail\":\"https://tse1.mm.bing.net/th?id=OIP.zBKuJobHBxYMSdOL3b7oOAHaG1&pid=Api\",\"thumbnail_token\":\"4480400578522cbc135bd7ce753dee35cbbb49e2709c4e5775b3f6334842c409\",\"title\":\"How to Build a Product Roadmap | Lucidspark\",\"url\":\"https://lucidspark.com/blog/how-to-build-a-product-roadmap\",\"width\":1839},{\"height\":688,\"image\":\"https://fanpu.io/assets/img/summaries/metagpt-overview.webp\",\"image_token\":\"7b40dc9efbd841a6810483fe46865f2a8fbc6def3902f385bb02c8199488d8d1\",\"source\":\"Bing\",\"thumbnail\":\"https://tse1.mm.bing.net/th?id=OIP.jEHUvpk8FOPdH12J79rFXQHaFR&pid=Api\",\"thumbnail_token\":\"8490d5575bc2435a2ae2adde324f594a1fd905d74bc8c8849565a99d1cf8e658\",\"title\":\"MetaGPT: Meta Programming for Multi-Agent Collaborative Framework | Fan ...\",\"url\":\"https://fanpu.io/summaries/2023-08-11-metagpt-meta-programming-for-multi-agent-collaborative-framework/\",\"width\":966},{\"height\":920,\"image\":\"https://roadmunk.com/guides/content/images/2020/09/Timeline-Roadmap-1.png\",\"image_token\":\"96f5e93a0d50706ce051a1024537055b9f1f1f7e2db4da2f158f13b98d56cd9d\",\"source\":\"Bing\",\"thumbnail\":\"https://tse1.mm.bing.net/th?id=OIP.hchaQb3VwheAODLNSOIAdQHaEe&pid=Api\",\"thumbnail_token\":\"28d9e72b560992d38447bd1a9d050d154d32f1eb0bf8690fe33d04c56ecdd618\",\"title\":\"What is a roadmap? The guide to roadmapping - Roadmunk\",\"url\":\"https://roadmunk.com/guides/roadmap-definition/\",\"width\":1520},{\"height\":920,\"image\":\"https://lh3.googleusercontent.com/H1-gvtbro-_27q3NnYbGO37i7a_xTqVRQ0ZvqJtRJBkfRjuPMg-11djNlPDLFJpjY8hKCzKwIlKiy040Us5unlwBPLUyqEMHIOUm7qqcEobhB-Uqsf2qHEyzEQywl9dkdErjkkrZ\",\"image_token\":\"8912b09257d9b4cef5ed84acd84b034f77aca601cf6d10094d87836b7b2bf7b5\",\"source\":\"Bing\",\"thumbnail\":\"https://tse1.mm.bing.net/th?id=OIP.qjGOCXm2nvDzRX8JRwxTagHaEe&pid=Api\",\"thumbnail_token\":\"7b3a7ba7cbcec30ecf21e2ac3f06daf1f8d55a9e3a7b4c657b9e1f1656c21f01\",\"title\":\"What is a product roadmap and how to create it? - Weje.io\",\"url\":\"https://weje.io/blog/product-roadmap\",\"width\":1520},{\"height\":3500,\"image\":\"https://static.vecteezy.com/system/resources/previews/000/680/342/original/infographic-business-roadmap-timeline.jpg\",\"image_token\":\"f78e25fbc66da7d380c8b378b3b51d2e3aabb98e01b39239b0184e8eff3f3b1d\",\"source\":\"Bing\",\"thumbnail\":\"https://tse1.mm.bing.net/th?id=OIP.1XTASxs0KABNJjUYwMRIWQHaFL&pid=Api\",\"thumbnail_token\":\"b41dcc02238e5f5150208de2a5f4d508fbecccc0040c24314e76107185be772a\",\"title\":\"Business Roadmap Vector Art, Icons, and Graphics for Free Download\",\"url\":\"https://www.vecteezy.com/free-vector/business-roadmap\",\"width\":5000},{\"height\":3871,\"image\":\"https://uniserveit.com/uploads/Building-A-Technology-Roadmap.jpg\",\"image_token\":\"2cd4a4a7699e095734ebd1278facc076305f4cc59854186bef8b98152a794f08\",\"source\":\"Bing\",\"thumbnail\":\"https://tse1.mm.bing.net/th?id=OIP.2ZnbiyLbkRcKHXL1_6u8JwHaE2&pid=Api\",\"thumbnail_token\":\"de73a3f195d39fb167707545a4761d3a15d0ef308a4b29e2e244202c52d76628\",\"title\":\"How To Build A Technology Roadmap | Uniserve IT Soltutions\",\"url\":\"https://uniserveit.com/blog/building-a-technology-roadmap\",\"width\":5903},{\"height\":940,\"image\":\"https://media.nngroup.com/media/editor/2020/10/28/screen-shot-2020-10-28-at-12537-pm.png\",\"image_token\":\"e90c0ef520ee067896af8acc0653d9ea2082d41fc2d82075c551a61de427ee42\",\"source\":\"Bing\",\"thumbnail\":\"https://tse1.mm.bing.net/th?id=OIP.RDV9bbgkuW_SJ28N8n_R9AHaDu&pid=Api\",\"thumbnail_token\":\"c242005da7b72897176ef634488b5025eb2d4925e6aae4aa6cec3ace673f62e3\",\"title\":\"The 6 steps to roadmapping (2022)\",\"url\":\"https://edduls.pics/article/the-6-steps-to-roadmapping\",\"width\":1872},{\"height\":720,\"image\":\"https://slidevilla.com/wp-content/uploads/2019/02/b7548f226f6de734c5dfa5f141a5918d-6.jpg\",\"image_token\":\"c6ac89b3b79b7e7d2a8f1246dbe131f95256f0ee2de78c478b92733ef40e63e1\",\"source\":\"Bing\",\"thumbnail\":\"https://tse1.mm.bing.net/th?id=OIP.PJUe_oduaD-FLmTMHHYp2wHaFj&pid=Api\",\"thumbnail_token\":\"e9193de0027c0317d3ea8ac3f6ffd3e30ec2043df987c29e4c002e9b2476681d\",\"title\":\"Roadmap with milestones powerpoint template - Slidevilla\",\"url\":\"https://slidevilla.com/shop/powerpoint-templates/roadmap-with-milestones/\",\"width\":960},{\"height\":2050,\"image\":\"https://www.jibility.com/wp-content/uploads/2021/09/digital-transformation-example-roadmap-detail.png\",\"image_token\":\"45a07b34c3b25676ca4f531482db5c858f6bbfd24ca0e593808aebfbb150acaa\",\"source\":\"Bing\",\"thumbnail\":\"https://tse1.mm.bing.net/th?id=OIP.sbKPggSDMYdKDZ5jbz14iwHaFJ&pid=Api\",\"thumbnail_token\":\"2a087acc29b5f757b8f7469d08f6c80701eb0393627932e93ecf7206714120c1\",\"title\":\"Strategic Roadmap Tool | Jibility | Professional Plan from $39\",\"url\":\"https://www.jibility.com/pricing/\",\"width\":2951},{\"height\":2095,\"image\":\"https://media.nngroup.com/media/articles/opengraph_images/6_Steps_Roadmapping_Social-Media-Posts_2020-38.png\",\"image_token\":\"855d0a05e27aee26c8f4ca738a99fd20f7b0efa43bfa177a22ce0717463e8a1b\",\"source\":\"Bing\",\"thumbnail\":\"https://tse1.mm.bing.net/th?id=OIP.bkZKyld4JdobfUy5xZaMzAHaD4&pid=Api\",\"thumbnail_token\":\"a0046653e2b02e1b76f8bc160d0c7734914e5cea3a37d5d61534943ccafb3f00\",\"title\":\"The 6 Steps to Roadmapping\",\"url\":\"https://www.nngroup.com/articles/roadmapping-steps/\",\"width\":4001},{\"height\":1440,\"image\":\"https://www.ciloart.com/files/free-process-roadmap-timeline-infographics-for-powerpoint-templates.jpg\",\"image_token\":\"f8a130afc6893bb6cec1a071b2f3fe6fec48bc8f15eb0efc0ad415a4e2c9d162\",\"source\":\"Bing\",\"thumbnail\":\"https://tse1.mm.bing.net/th?id=OIP.wDWVEEfr5zg9vz7HAnK0DQHaEK&pid=Api\",\"thumbnail_token\":\"fafeddafec61e36c6e39575442bfd415d85c909580b432618d4bc79efcc5b9b5\",\"title\":\"Roadmap ppt template free - plmnoble\",\"url\":\"https://plmnoble.weebly.com/blog/roadmap-ppt-template-free\",\"width\":2560},{\"height\":576,\"image\":\"https://cdn.infodiagram.com/c/a84123/team-roadmap-engineering-chart-development-innovation-technology.png\",\"image_token\":\"aaf2d11a998569ac9cebb8b0441fca9a1a5f9b88a902737cca46a41b2a3d5b1f\",\"source\":\"Bing\",\"thumbnail\":\"https://tse1.mm.bing.net/th?id=OIP.IBaC7bT304-c6nyTTJiijAHaEK&pid=Api\",\"thumbnail_token\":\"788dc194a239068cff396bf9d134d6dd3aa22adb27fa480345951cb142fc37ec\",\"title\":\"Technology Roadmap PPT Template\",\"url\":\"https://www.infodiagram.co.uk/slides/roadmap-technology-template/\",\"width\":1024},{\"height\":2048,\"image\":\"https://1.bp.blogspot.com/-Dv1SChkX87k/YD3c7mfegKI/AAAAAAAARSI/8thS6TtRC30DiAwzBXXfKw1IwgLp695JQCLcBGAsYHQ/s2048/Wed%2BRoadmap.jpeg\",\"image_token\":\"9744a089f465662e7961b05bbf9859438a69af2f7d7190af5059835b9c3001d6\",\"source\":\"Bing\",\"thumbnail\":\"https://tse1.mm.bing.net/th?id=OIP.xTQzIRZCy3BS7DuA2YWXVgHaLG&pid=Api\",\"thumbnail_token\":\"d3f95531ab49a8166b2eb7b49aafe8b09b08ecb4224151456a63fd892ae077f7\",\"title\":\"Learn Web Development as an absolute Beginner Roadmap & What Skills you ...\",\"url\":\"https://codewithwastik.blogspot.com/2021/03/learn-web-development-as-absolute.html\",\"width\":1367},{\"height\":1440,\"image\":\"https://www.itce.com/wp-content/uploads/2018/11/SAFe-Implementation-Roadmap-ITCE-1920x1440.png\",\"image_token\":\"922478c3965e75a6f91ba2aec69832a2a725feffed976e0422377ce7e1c61146\",\"source\":\"Bing\",\"thumbnail\":\"https://tse1.mm.bing.net/th?id=OIP.Mkqd375BYhPyxasV59LLPwHaFj&pid=Api\",\"thumbnail_token\":\"2319f3ca85bae584140ce010bacecfa89823bc3ac15760fc2bc620eaa4c09492\",\"title\":\"SAFe Roadmap Implementation - ITCE\",\"url\":\"https://www.itce.com/services/safe-implementation-roadmap/\",\"width\":1920},{\"height\":1214,\"image\":\"https://graphicpanda.net/wp-content/uploads/2019/12/09.jpg\",\"image_token\":\"8c3fec9753bddf54dbbc4334adcb7c3845ebc2c9ac9d9919a625b4e81b931227\",\"source\":\"Bing\",\"thumbnail\":\"https://tse1.mm.bing.net/th?id=OIP.aer6Fq6Foz4Ugx0PGAUX1wHaE8&pid=Api\",\"thumbnail_token\":\"5cda767e92851048094f8b0eb9e5782c590d178766a35e12f102bdaf12eebb55\",\"title\":\"Top 48 Best Roadmap Infographics of 2019\",\"url\":\"https://graphicpanda.net/top-48-best-roadmap-infographics-of-2019/\",\"width\":1820},{\"height\":858,\"image\":\"https://47billion.com/wp-content/uploads/2022/02/Roadmap-for-Transforming-into-a-Data-Driven-Organization-e1645684780855.png\",\"image_token\":\"89aa4e1ff6bbf4325203cf48edcd389e876ad5b4aece7c79c127206fd7af53e3\",\"source\":\"Bing\",\"thumbnail\":\"https://tse1.mm.bing.net/th?id=OIP.8_44ACp4csLk-5DvQxJ1WgHaF4&pid=Api\",\"thumbnail_token\":\"e34f7de48ebea075b768b643bdca97ca0f5780511cbd5d512360baf9f61e5a74\",\"title\":\"Roadmap for Transforming into a Data-Driven Organization - 47billion.com\",\"url\":\"https://47billion.com/blog/roadmap-for-transforming-into-a-data-driven-organization/\",\"width\":1080},{\"height\":1656,\"image\":\"https://business-docs.co.uk/wp-content/uploads/2021/06/BDUK43StrategyRoadmapTemplatePowerpoint16x90501.png\",\"image_token\":\"4c7ed30a8f7f563a9a9c556ff684737bd5c57e4f2fa9643543a0961a6ea81748\",\"source\":\"Bing\",\"thumbnail\":\"https://tse1.mm.bing.net/th?id=OIP.9SJg6AkiAlFFaiSaR_wt-AHaEQ&pid=Api\",\"thumbnail_token\":\"1aa46a29b71fb20bce0482e7d652e25fd9b0068d27c153dc8934c0cdabe8be87\",\"title\":\"Strategy Roadmap Template PowerPoint - Present your strategic plans!\",\"url\":\"https://business-docs.co.uk/downloads/strategy-roadmap-template-powerpoint/\",\"width\":2880},{\"height\":1163,\"image\":\"https://i.pinimg.com/originals/55/17/fb/5517fbb701b86448db0e2027c3143d24.png\",\"image_token\":\"26a8a386dc179392ce51475382e003749b26eab3020410abee2451f837227e1e\",\"source\":\"Bing\",\"thumbnail\":\"https://tse1.mm.bing.net/th?id=OIP.m3dqCemXYTHDwnTNHr9CzgHaEj&pid=Api\",\"thumbnail_token\":\"be77ab5985646acf21a6268137d9c9108f60731a90cd9a36e116e30ba89ba445\",\"title\":\"Marketing Roadmap - Template and Examples | Roadmunk | Marketing ...\",\"url\":\"https://www.pinterest.de/pin/588704982531603300/\",\"width\":1893},{\"height\":1170,\"image\":\"https://d2slcw3kip6qmk.cloudfront.net/marketing/blog/2019Q4/technology-roadmap/it-roadmap-example.png\",\"image_token\":\"093a2a27e4cd0988223d8947a600db27208ce6c77f522eaca579c3eddab2d6e7\",\"source\":\"Bing\",\"thumbnail\":\"https://tse1.mm.bing.net/th?id=OIP.WRAD5IO2lrB7GGe7fLSJwgHaFN&pid=Api\",\"thumbnail_token\":\"910b3be01ddda5ea65a713080fb458385fb72d8911f434b59d08c18cf7ad5d38\",\"title\":\"What Is a Product Roadmap and How to Create One? | LaunchPad Lab\",\"url\":\"https://launchpadlab.com/blog/what-is-a-product-roadmap-and-why-you-need-one/\",\"width\":1662}],\"vqd\":{\"The%20roadmap%20of%20MetaGPT\":\"4-25941261128344049410840372626152530092\"}});DDG.duckbar.load('news');DDG.duckbar.load('videos', {\"ads\":[],\"query\":\"The roadmap of MetaGPT\",\"queryEncoded\":\"The%20roadmap%20of%20MetaGPT\",\"response_type\":\"places\",\"results\":[{\"content\":\"https://www.youtube.com/watch?v=uT75J_KG_aY\",\"description\":\"In this video, we review MetaGPT, a new project that aims to recreate an entire engineering organization using AI. MetaGPT is a CEO, Product Manager, Architect, Project Manager, Engineering, and QA. Write a simple prompt, and you get everything from the requirements to the PRDs to the code and tests. How To Find Me: Become a Patron \\ud83d\\udd25 - https ...\",\"duration\":\"6:36\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/uT75J_KG_aY?autoplay=1\",\"image_token\":\"57974159b78b309485721c0bce280219d9927e071e542a34777864767d6cb8d4\",\"images\":{\"large\":\"https://tse3.mm.bing.net/th?id=OVP.BbSKV8N1vyYv-3m8vyuCoQEsDh&pid=Api\",\"medium\":\"https://tse3.mm.bing.net/th?id=OVP.BbSKV8N1vyYv-3m8vyuCoQEsDh&pid=Api\",\"motion\":\"https://tse3.mm.bing.net/th?id=OM1.bsXxoMoJ9ZWQBw&pid=Api\",\"small\":\"https://tse3.mm.bing.net/th?id=OVP.BbSKV8N1vyYv-3m8vyuCoQEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-08-14T14:09:10.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":75408},\"title\":\"How To Install MetaGPT - Build A Startup With One Prompt!!\",\"uploader\":\"Matthew Berman\"},{\"content\":\"https://www.youtube.com/watch?v=YtxMderNrzU\",\"description\":\"Subscribe to my Newsletter (My AI updates and news clearly explained): https://louisbouchard.substack.com/ References: Read the full article: https://www.louisbouchard.ai/metagpt/ Hong et al., 2023: MetaGPT, https://arxiv.org/pdf/2308.00352.pdf Code: https://github.com/geekan/MetaGPT/blob/main/README.md Twitter: https://twitter.com/Whats_AI ...\",\"duration\":\"7:38\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/YtxMderNrzU?autoplay=1\",\"image_token\":\"2e0774ace2e34bbe23ece04e80b7bb2ee976fd8ef7f53001e8f8b137763561dc\",\"images\":{\"large\":\"https://tse2.mm.bing.net/th?id=OVP.HP81CZ34ap22GZZG2l024QHgFo&pid=Api\",\"medium\":\"https://tse2.mm.bing.net/th?id=OVP.HP81CZ34ap22GZZG2l024QHgFo&pid=Api\",\"motion\":\"https://tse2.mm.bing.net/th?id=OM2.xArTjo5bOxSBhg&pid=Api\",\"small\":\"https://tse2.mm.bing.net/th?id=OVP.HP81CZ34ap22GZZG2l024QHgFo&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-08-27T15:05:12.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":9594},\"title\":\"MetaGPT: Redefining Multi-Agent Collaboration for Complex Tasks\",\"uploader\":\"What's AI by Louis Bouchard\"},{\"content\":\"https://www.youtube.com/watch?v=nqZlTV_L6Ao\",\"description\":\"Welcome to our video review! \\ud83c\\udfa5 Dive into the world of MetaGPT, a revolutionary project that's redefining the boundaries of AI. \\ud83e\\udd16 Imagine having an entire engineering team - from CEO to QA - compacted into one AI system. Just input a prompt, and voila! You're handed everything from requirements, PRDs, to the actual code and tests. Let ...\",\"duration\":\"14:15\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/nqZlTV_L6Ao?autoplay=1\",\"image_token\":\"9d13b27084400da23ef8d8567bd6b5c8a3758d4129f2b28c3619c0e2e1ba8276\",\"images\":{\"large\":\"https://tse4.mm.bing.net/th?id=OVP.VBEy5DF-0BQshjEkqA9T0wHgFo&pid=Api\",\"medium\":\"https://tse4.mm.bing.net/th?id=OVP.VBEy5DF-0BQshjEkqA9T0wHgFo&pid=Api\",\"motion\":\"https://tse4.mm.bing.net/th?id=OM2.N7S3-wAngkj7VA&pid=Api\",\"small\":\"https://tse4.mm.bing.net/th?id=OVP.VBEy5DF-0BQshjEkqA9T0wHgFo&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-09-04T11:45:06.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":23248},\"title\":\"\\ud83d\\ude80 MetaGPT Setup: Launch a Startup with One \\u270d\\ufe0f Prompt!\",\"uploader\":\"Prompt Engineering\"},{\"content\":\"https://www.youtube.com/watch?v=12X4pupy4No\",\"description\":\"Simple as Plug n Play Visit www.MetaIDT.com\",\"duration\":\"1:00\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/12X4pupy4No?autoplay=1\",\"image_token\":\"3bf00a4528bef3408e273fd9403d2bce8428fc915c51ba5d0b09527abb7b47ce\",\"images\":{\"large\":\"https://tse2.mm.bing.net/th?id=OVP.gqgtA4cHlbRoVhYkFEkUuQEsDh&pid=Api\",\"medium\":\"https://tse2.mm.bing.net/th?id=OVP.gqgtA4cHlbRoVhYkFEkUuQEsDh&pid=Api\",\"motion\":\"https://tse2.mm.bing.net/th?id=OM1.dgsWk4LJc4VGrQ_1691420084&pid=Api\",\"small\":\"https://tse2.mm.bing.net/th?id=OVP.gqgtA4cHlbRoVhYkFEkUuQEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-07-17T09:43:32.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":470},\"title\":\"MetaGPT Key Drive Installation Guide\",\"uploader\":\"MetaGPT\"},{\"content\":\"https://www.youtube.com/watch?v=EgipcKPhqME\",\"description\":\"In this video I provide a great demo and overview of a project called MetaGPT. Have you ever wondered if each person in a development project (such as the project manager, developers, architects, QA testers, etc.) were all AI's and how they'd behave? MetaGPT is doing just that. Not only are all the docs, designs, and tasks delivered, but also a ...\",\"duration\":\"7:35\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/EgipcKPhqME?autoplay=1\",\"image_token\":\"624d4ccdb6d1605da1e388e85c9124957bcba9c70a11a575e751ba6fc09bc5f8\",\"images\":{\"large\":\"https://tse4.mm.bing.net/th?id=OVP.hG0c3nw7X-uz0gzUjnOVNwEsDh&pid=Api\",\"medium\":\"https://tse4.mm.bing.net/th?id=OVP.hG0c3nw7X-uz0gzUjnOVNwEsDh&pid=Api\",\"motion\":\"https://tse4.mm.bing.net/th?id=OM1.8F2lEMy1JlCKsQ_1698986522&pid=Api\",\"small\":\"https://tse4.mm.bing.net/th?id=OVP.hG0c3nw7X-uz0gzUjnOVNwEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-09-24T08:00:11.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":1587},\"title\":\"MetaGPT Tutorial | It builds an entire project (with working source code) with just one prompt!!\",\"uploader\":\"CraceCasts\"},{\"content\":\"https://www.youtube.com/watch?v=T_wBUpzxxPY\",\"description\":\"In this video i talk about this awesome project called MetaGPT in my video. Now, MetaGPT is like an all-in-one AI powerhouse. It can do everything from being a CEO to a QA tester for an engineering organization. And the cool thing is, you just give it a simple prompt, and it spits out everything you need - requirements, PRDs, code, and tests ...\",\"duration\":\"4:00\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/T_wBUpzxxPY?autoplay=1\",\"image_token\":\"ef14791d7faff848cb15177567e9f4f9c04ccae4fafc7ef7386e69df3a012010\",\"images\":{\"large\":\"https://tse1.mm.bing.net/th?id=OVP.EWCOFStB_tQza4SLrUA0AAEsDh&pid=Api\",\"medium\":\"https://tse1.mm.bing.net/th?id=OVP.EWCOFStB_tQza4SLrUA0AAEsDh&pid=Api\",\"motion\":\"https://tse1.mm.bing.net/th?id=OM1.itG5pHJg6MKYzg_1696190983&pid=Api\",\"small\":\"https://tse1.mm.bing.net/th?id=OVP.EWCOFStB_tQza4SLrUA0AAEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-09-11T10:41:22.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":368},\"title\":\"MetaGPT Installation Guide: From Setup to Startup With One Prompt!\",\"uploader\":\"Py Man\"},{\"content\":\"https://www.youtube.com/watch?v=AwnltW8n74A\",\"description\":\"MetaGPT is a framework that uses GPT-4 to automate multiple roles within a software company. For example product managers, software architects, project managers and software engineers. It writes code, documentation, user stories, competitive analysis, and creates diagrams. Github: https://github.com/geekan/MetaGPT #ai #gpt4\",\"duration\":\"7:53\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/AwnltW8n74A?autoplay=1\",\"image_token\":\"b7c3d4481f0f7b7b7c7c43d3da07368a2feb28b2fbdbd8b86b8d5c64b19833fd\",\"images\":{\"large\":\"https://tse1.mm.bing.net/th?id=OVP.J4PD9qpp3rIqXai84Jsu2wEsDh&pid=Api\",\"medium\":\"https://tse1.mm.bing.net/th?id=OVP.J4PD9qpp3rIqXai84Jsu2wEsDh&pid=Api\",\"motion\":\"https://tse1.mm.bing.net/th?id=OM1.EGElEVpTnZYCdQ_1691938448&pid=Api\",\"small\":\"https://tse1.mm.bing.net/th?id=OVP.J4PD9qpp3rIqXai84Jsu2wEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-08-06T23:09:15.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":6497},\"title\":\"MetaGPT - Multi-Agent Framework with GPT-4\",\"uploader\":\"Tosh Velaga\"},{\"content\":\"https://www.youtube.com/watch?v=D80u__nYYWw\",\"description\":\"Learn how to quickly build a roadmap alongside the same table and board views you already know and love in GitHub Projects. With Senior Product Manager, Riley Broughten and Developer Advocate, Kedasha Kerr (@itsthatladydev) Blog: https://gh.io/roadmaps-changelog Project Roadmaps Docs: https://gh.io/roadmaps Tell us what you think!: https://gh ...\",\"duration\":\"7:01\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/D80u__nYYWw?autoplay=1\",\"image_token\":\"2a89cec713d7aae159325e0cb365581ed2715f02621e9f83a9738ffa92664166\",\"images\":{\"large\":\"https://tse3.mm.bing.net/th?id=OVP.46T5385YNZojaEJclzxHKQEsDh&pid=Api\",\"medium\":\"https://tse3.mm.bing.net/th?id=OVP.46T5385YNZojaEJclzxHKQEsDh&pid=Api\",\"motion\":\"https://tse3.mm.bing.net/th?id=OM.jWPqoPJyqHDaVw_1685085608&pid=Api\",\"small\":\"https://tse3.mm.bing.net/th?id=OVP.46T5385YNZojaEJclzxHKQEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-04-18T13:48:05.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":19906},\"title\":\"Learn how to use Project Roadmaps - GitHub Checkout\",\"uploader\":\"GitHub\"},{\"content\":\"https://www.youtube.com/watch?v=pJwR5pv0_gs\",\"description\":\"Multi agent framework tutorial of MetaGPT & chatDev; Check the Hubspot x Jasper research of Using Generative AI to Scale Your Content Operations: https://offers.hubspot.com/generative-ai-for-content-operations?utm_source=youtube&utm_medium=social&utm_campaign=CR0087Sep2023_AIJason/partner_youtube \\ud83d\\udd17 Links - Follow me on twitter: https ...\",\"duration\":\"13:41\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/pJwR5pv0_gs?autoplay=1\",\"image_token\":\"18ac54a8e5144c74f2010219781c47c295099a6eed7479645733832910d19aec\",\"images\":{\"large\":\"https://tse4.mm.bing.net/th?id=OVP.LJ0SK8DLWjCcwVVh-PEcOwHgFo&pid=Api\",\"medium\":\"https://tse4.mm.bing.net/th?id=OVP.LJ0SK8DLWjCcwVVh-PEcOwHgFo&pid=Api\",\"motion\":\"https://tse4.mm.bing.net/th?id=OM2.PxMMOsse4Yi_FQ&pid=Api\",\"small\":\"https://tse4.mm.bing.net/th?id=OVP.LJ0SK8DLWjCcwVVh-PEcOwHgFo&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-09-08T11:36:03.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":167793},\"title\":\"Build AI agent workforce - Multi agent framework with MetaGPT & chatDev\",\"uploader\":\"AI Jason\"},{\"content\":\"https://www.youtube.com/watch?v=q16Gi9pTG_M\",\"description\":\"In this captivating video, we explore the core concept of MetaGPT, which centers on task distribution and coordination among individual GPT agents. Each agent is bestowed with specific roles that capitalize on their unique strengths and expertise. Imagine one GPT excelling in natural language understanding, while another showcases prowess in ...\",\"duration\":\"14:56\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/q16Gi9pTG_M?autoplay=1\",\"image_token\":\"bee3657ef83c9da2bc4ccfea770244e18958f5789a39d0136c3a049cc22a0e54\",\"images\":{\"large\":\"https://tse4.mm.bing.net/th?id=OVP.eiPUmQWRU1sE-01-x5Kn7gEsDh&pid=Api\",\"medium\":\"https://tse4.mm.bing.net/th?id=OVP.eiPUmQWRU1sE-01-x5Kn7gEsDh&pid=Api\",\"motion\":\"https://tse4.mm.bing.net/th?id=OM2.eWDmjf8nvrSrhw&pid=Api\",\"small\":\"https://tse4.mm.bing.net/th?id=OVP.eiPUmQWRU1sE-01-x5Kn7gEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-07-25T00:37:40.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":14365},\"title\":\"MetaGPT: Deploy POWERFUL Autonomous Ai Agents BETTER Than SUPERAGI! (Installation Tutorial)\",\"uploader\":\"WorldofAI\"}],\"vqd\":{\"The%20roadmap%20of%20MetaGPT\":\"4-25941261128344049410840372626152530092\"}});DDG.duckbar.loadModule('related_searches');if (DDG.pageLayout) DDG.pageLayout.initialize({\"mainline\":{\"items\":[[\"organic\"],[\"images\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"videos\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"]]}}, { start: 0 });DDG.deep.emit(\"load:completed\");", + "curl-cffi-POST-https://duckduckgo.com-{\"data\": {\"q\": \"The function of MetaGPT\"}}": "The function of MetaGPT at DuckDuckGo
", + "curl-cffi-GET-https://links.duckduckgo.com/d.js-{\"params\": {\"bing_market\": \"wt-WT\", \"df\": null, \"ex\": \"-1\", \"kl\": \"wt-wt\", \"l\": \"wt-wt\", \"q\": \"The function of MetaGPT\", \"s\": \"0\", \"sp\": \"0\", \"vqd\": \"4-148519746540767190220111387879117509726\"}}": "if (DDG.deep && DDG.deep.setUpstream) DDG.deep.setUpstream(\"bingv7aa\");DDG.deep.bn={'ivc':1};if (DDG.pageLayout) DDG.pageLayout.load('a',[], {\"page_load_url\":\"https://duckduckgo.com/y.js?iurl=%7B2%7DIG%3D3A2CFCB179EC4A63AF1E2047F34A7CBB%26CID%3D3280021F2FA667E6037316192E5B66D4%26Type%3DEvent.CPT%26DATA%3D0\"});DDG.deep.signalSummary = \"\";DDG.inject('DDG.Data.languages.resultLanguages', {\"en\":[\"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\"https://www.straight.com/guides/software/a-complete-guide-to-metagpt-the-best-ai-agent-available-now/\",\"https://levelup.gitconnected.com/metagpt-the-future-of-multi-agent-collaboration-in-ai-a-brief-guide-fd4b4429336d\",\"https://geekflare.com/metagpt-multi-agent-framework/\",\"https://medium.com/aimonks/metagpt-a-framework-for-multi-agent-meta-programming-6c79f2eafb8e\",\"https://arxiv.org/abs/2308.00352\",\"https://github.com/geekan/MetaGPT\",\"https://medium.com/@reddy.khoushik/metagpt-the-multi-agent-framework-revolutionizing-software-collaboration-38e48397021f\",\"https://www.marktechpost.com/2023/08/09/meet-metagpt-the-open-source-ai-framework-that-transforms-gpts-into-engineers-architects-and-managers/\",\"https://ai-scholar.tech/en/articles/agent-simulation/meta-gpt\",\"https://docs.deepwisdom.ai/main/en/guide/tutorials/multi_agent_101.html\",\"https://lablab.ai/blog/this-week-in-ai-exploring-the-latest-from-metagpt-and-gpt4-and-more\",\"https://analyticsindiamag.com/metagpt-realising-the-gpt-4-dream/\",\"https://docs.deepwisdom.ai/main/en/guide/tutorials/agent_101.html\",\"https://www.almabetter.com/bytes/articles/metagpt\",\"https://medium.com/@korolalexei/metagpt-a-multi-agent-framework-revolutionizing-software-development-f585fe1aa950\",\"https://github.com/PlaiD3/MetaGPT/blob/main/README.md\",\"https://eightify.app/summary/computer-science-and-technology/metagpt-advanced-autonomous-ai-agents-installation-tutorial\",\"https://ar5iv.labs.arxiv.org/html/2308.00352\",\"https://www.freegpttools.org/metagpt\",\"https://github.com/geekan/MetaGPT/releases\",\"https://theventurecation.com/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\"https://blogs.windows.com/windowsexperience/2024/01/04/introducing-a-new-copilot-key-to-kick-off-the-year-of-ai-powered-windows-pcs/\",\"https://www.instagram.com/richfieldmusic/p/C1bpt1eucbU/\"]});DDG.deep.pageLayoutSummary = \"w25\";DDG.inject('DDG.Data.languages.adLanguages', {});if (DDG.pageLayout) DDG.pageLayout.load('d',[{\"a\":\"To actualize an agile, flexible software architecture that can adapt to dynamic programming tasks. Agile Development SOPs act as a meta-function here, coordinating agents to auto-generate code based on defined inputs.\",\"ae\":null,\"c\":\"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\"d\":\"www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\"da\":\"\",\"e\":\"2023-09-11T00:00:00.0000000\",\"h\":0,\"i\":\"www.unite.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Complete Guide to the Best AI Agent Available Right Now\",\"u\":\"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\"},{\"a\":\"MetaGPT is a multi-agent system that utilizes Large Language Models (LLMs) to perform complex tasks. It is designed to overcome the limitations of LLMs in fostering effective collaboration and...\",\"ae\":null,\"c\":\"https://www.straight.com/guides/software/a-complete-guide-to-metagpt-the-best-ai-agent-available-now/\",\"d\":\"www.straight.com/guides/software/a-complete-guide-to-metagpt-the-best-ai-agent-available-now/\",\"da\":\"\",\"e\":\"2023-12-13T00:00:00.0000000\",\"h\":0,\"i\":\"www.straight.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"A Complete Guide to MetaGPT: The Best AI Agent Available Now\",\"u\":\"https://www.straight.com/guides/software/a-complete-guide-to-metagpt-the-best-ai-agent-available-now/\"},{\"a\":\"1 Created by Bing In the ever-evolving world of artificial intelligence, one term has recently taken the spotlight: MetaGPT. As the digital landscape becomes more competitive, understanding and leveraging the capabilities of MetaGPT can be a game-changer for businesses, developers, and AI enthusiasts alike.\",\"ae\":null,\"c\":\"https://levelup.gitconnected.com/metagpt-the-future-of-multi-agent-collaboration-in-ai-a-brief-guide-fd4b4429336d\",\"d\":\"levelup.gitconnected.com/metagpt-the-future-of-multi-agent-collaboration-in-ai-a-brief-guide-fd4b4429336d\",\"da\":\"\",\"e\":\"2023-08-09T00:00:00.0000000\",\"h\":0,\"i\":\"levelup.gitconnected.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: The Future of Multi-Agent Collaboration in AI (A Brief Guide)\",\"u\":\"https://levelup.gitconnected.com/metagpt-the-future-of-multi-agent-collaboration-in-ai-a-brief-guide-fd4b4429336d\"},{\"a\":\"MetaGPT is a multi-agent framework that takes one-line inputs to produce APIs, user stories, data structures, competitive analysis, and more. GPT is the short form for Generative Pretrained Transformers. MetaGPT framework can behave as a product manager, software engineer, and architect. This framework can act as an entire software company with ...\",\"ae\":null,\"c\":\"https://geekflare.com/metagpt-multi-agent-framework/\",\"d\":\"geekflare.com/metagpt-multi-agent-framework/\",\"da\":\"\",\"e\":\"2023-09-18T00:00:00.0000000\",\"h\":0,\"i\":\"geekflare.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Is This the Best Multi-Agent Framework Yet? - Geekflare\",\"u\":\"https://geekflare.com/metagpt-multi-agent-framework/\"},{\"a\":\"Gaming: MetaGPT can be used to create and control intelligent agents that can cooperate or compete with human players or other agents in various games, such as board games, card games, video...\",\"ae\":null,\"c\":\"https://medium.com/aimonks/metagpt-a-framework-for-multi-agent-meta-programming-6c79f2eafb8e\",\"d\":\"medium.com/aimonks/metagpt-a-framework-for-multi-agent-meta-programming-6c79f2eafb8e\",\"da\":\"\",\"e\":\"2023-08-03T00:00:00.0000000\",\"h\":0,\"i\":\"medium.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: A Framework for Multi-Agent Meta Programming\",\"u\":\"https://medium.com/aimonks/metagpt-a-framework-for-multi-agent-meta-programming-6c79f2eafb8e\"},{\"a\":\"You can check this by using:</span>\n<span class=\"pl-c\"><span class=\"pl-c\">#</span> You can use conda to initialize a new python env</span>\n<span class=\"pl-c\"><span class=\"pl-c\">#</span> conda create -n metagpt python=3.9</span>\n<span class=\"pl-c\"><span class=\"pl-c\">#</span> conda activate metagpt</span>\npython3 --version\n\n<span...\",\"ae\":null,\"b\":\"gh\\tGitHub\\tgithub.com\",\"c\":\"https://github.com/geekan/MetaGPT/blob/main/README.md\",\"d\":\"github.com/geekan/MetaGPT/blob/main/README.md\",\"da\":\"\",\"h\":0,\"i\":\"github.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: The Multi-Agent Framework - GitHub\",\"u\":\"https://github.com/geekan/MetaGPT/blob/main/README.md\"},{\"a\":\"MetaGPT utilizes an assembly line paradigm to assign diverse roles to various agents, efficiently breaking down complex tasks into subtasks involving many agents working together. On collaborative software engineering benchmarks, MetaGPT generates more coherent solutions than previous chat-based multi-agent systems.\",\"ae\":null,\"b\":\"arx\\tarXiv.org\\tarxiv.org\",\"c\":\"https://arxiv.org/abs/2308.00352\",\"d\":\"arxiv.org/abs/2308.00352\",\"da\":\"translations\",\"e\":\"2023-08-01T00:00:00.0000000\",\"h\":0,\"i\":\"arxiv.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework\",\"u\":\"https://arxiv.org/abs/2308.00352\"},{\"a\":\"MetaGPT: The Multi-Agent Framework Assign different roles to GPTs to form a collaborative software entity for complex tasks. MetaGPT takes a one line requirement as input and outputs user stories / competitive analysis / requirements / data structures / APIs / documents, etc.\",\"ae\":null,\"b\":\"gh\\tGitHub\\tgithub.com\",\"c\":\"https://github.com/geekan/MetaGPT\",\"d\":\"github.com/geekan/MetaGPT\",\"da\":\"\",\"h\":0,\"i\":\"github.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: The Multi-Agent Framework - GitHub\",\"u\":\"https://github.com/geekan/MetaGPT\"},{\"a\":\"MetaGPT's innovative approach to collaborative AI has the potential to reshape the landscape of software development. By harnessing the collective power of specialized AI roles, developers can ...\",\"ae\":null,\"c\":\"https://medium.com/@reddy.khoushik/metagpt-the-multi-agent-framework-revolutionizing-software-collaboration-38e48397021f\",\"d\":\"medium.com/@reddy.khoushik/metagpt-the-multi-agent-framework-revolutionizing-software-collaboration-38e48397021f\",\"da\":\"translations\",\"e\":\"2023-08-18T00:00:00.0000000\",\"h\":0,\"i\":\"medium.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: The Multi-Agent Framework Revolutionizing Software ... - Medium\",\"u\":\"https://medium.com/@reddy.khoushik/metagpt-the-multi-agent-framework-revolutionizing-software-collaboration-38e48397021f\"},{\"a\":\"MetaGPT streamlines the coordination between interdependent jobs by formalizing the artifacts that human experts exchange. Agents are connected by a shared environment that offers insight into activities and shared use of tools and resources. All communications between agents are contained in this environment.\",\"ae\":null,\"c\":\"https://www.marktechpost.com/2023/08/09/meet-metagpt-the-open-source-ai-framework-that-transforms-gpts-into-engineers-architects-and-managers/\",\"d\":\"www.marktechpost.com/2023/08/09/meet-metagpt-the-open-source-ai-framework-that-transforms-gpts-into-engineers-architects-and-managers/\",\"da\":\"translations\",\"e\":\"2023-08-09T00:00:00.0000000\",\"h\":0,\"i\":\"www.marktechpost.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Meet MetaGPT: The Open-Source AI Framework That Transforms GPTs into ...\",\"u\":\"https://www.marktechpost.com/2023/08/09/meet-metagpt-the-open-source-ai-framework-that-transforms-gpts-into-engineers-architects-and-managers/\"},{\"a\":\"This paper presents MetaGPT, a multi-agent framework that extends complex problem solving capabilities by encoding SOPs that incorporate real-world expertise into LLM agents, and shows through experiments that it can generate more consistent and comprehensive solutionsthan existing methods.\",\"ae\":null,\"c\":\"https://ai-scholar.tech/en/articles/agent-simulation/meta-gpt\",\"d\":\"ai-scholar.tech/en/articles/agent-simulation/meta-gpt\",\"da\":\"\",\"e\":\"2023-08-18T00:00:00.0000000\",\"h\":0,\"i\":\"ai-scholar.tech\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT, a multi-agent framework in which AI consistently develops ...\",\"u\":\"https://ai-scholar.tech/en/articles/agent-simulation/meta-gpt\"},{\"a\":\"The core advantage of MetaGPT also lies in the easy and flexible development of a team of agents. Under MetaGPT framework, users can enable interactions between agents with a minimal amount of codes. ... we need three steps to set up the team and make it function: Define each role capable of intended actions; Think about the Standard Operating ...\",\"ae\":null,\"c\":\"https://docs.deepwisdom.ai/main/en/guide/tutorials/multi_agent_101.html\",\"d\":\"docs.deepwisdom.ai/main/en/guide/tutorials/multi_agent_101.html\",\"da\":\"translations\",\"h\":0,\"i\":\"docs.deepwisdom.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MultiAgent 101 | MetaGPT\",\"u\":\"https://docs.deepwisdom.ai/main/en/guide/tutorials/multi_agent_101.html\"},{\"a\":\"MetaGPT is a groundbreaking multi-agent framework that is transforming the way software development is approached. By taking a single line of requirement as input, MetaGPT outputs a comprehensive array of development components, including user stories, competitive analysis, requirements, data structures, APIs, and documents.\",\"ae\":null,\"c\":\"https://lablab.ai/blog/this-week-in-ai-exploring-the-latest-from-metagpt-and-gpt4-and-more\",\"d\":\"lablab.ai/blog/this-week-in-ai-exploring-the-latest-from-metagpt-and-gpt4-and-more\",\"da\":\"\",\"e\":\"2023-08-11T00:00:00.0000000\",\"h\":0,\"i\":\"lablab.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"This Week in AI: Exploring the Latest from MetaGPT and GPT-4 and more..\",\"u\":\"https://lablab.ai/blog/this-week-in-ai-exploring-the-latest-from-metagpt-and-gpt4-and-more\"},{\"a\":\"MetaGPT then asks for a few additional details, such as the required inputs from the user. Subscribe to our Newsletter. ... One only needs to look at the success of AutoGPT, an open-source project looking to allow GPT-4 to function autonomously. Other similar projects include BabyAGI, a GPT API powered task management system, ...\",\"ae\":null,\"c\":\"https://analyticsindiamag.com/metagpt-realising-the-gpt-4-dream/\",\"d\":\"analyticsindiamag.com/metagpt-realising-the-gpt-4-dream/\",\"da\":\"\",\"e\":\"2023-04-26T00:00:00.0000000\",\"h\":0,\"i\":\"analyticsindiamag.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT \\u2014 Realising the GPT-4 Dream - Analytics India Magazine\",\"u\":\"https://analyticsindiamag.com/metagpt-realising-the-gpt-4-dream/\"},{\"a\":\"In MetaGPT, class Action is the logical abstraction for an action. Users may use LLM to empower this Action by simply invoking the self._aask function, which will make LLM api call under the hood. In our scenario, we define a SimpleWriteCode subclassed Action.\",\"ae\":null,\"c\":\"https://docs.deepwisdom.ai/main/en/guide/tutorials/agent_101.html\",\"d\":\"docs.deepwisdom.ai/main/en/guide/tutorials/agent_101.html\",\"da\":\"translations\",\"h\":0,\"i\":\"docs.deepwisdom.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Agent 101 | MetaGPT\",\"u\":\"https://docs.deepwisdom.ai/main/en/guide/tutorials/agent_101.html\"},{\"a\":\"Understanding MetaGPT MetaGPT, a concept originating from a research paper that received significant attention, represents a leap forward in Artificial Intelligence, specifically in multi-agent collaboration using large language models (LLMs).\",\"ae\":null,\"c\":\"https://www.almabetter.com/bytes/articles/metagpt\",\"d\":\"www.almabetter.com/bytes/articles/metagpt\",\"da\":\"\",\"e\":\"2023-08-28T00:00:00.0000000\",\"h\":0,\"i\":\"www.almabetter.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: The Future of Multi-Agent Collaboration in AI\",\"u\":\"https://www.almabetter.com/bytes/articles/metagpt\"},{\"a\":\"MetaGPT is a trending GitHub repository that simulates different roles in a software company using GPT-4. It's like a software company in a box (or CLI to be precise).\",\"ae\":null,\"c\":\"https://medium.com/@korolalexei/metagpt-a-multi-agent-framework-revolutionizing-software-development-f585fe1aa950\",\"d\":\"medium.com/@korolalexei/metagpt-a-multi-agent-framework-revolutionizing-software-development-f585fe1aa950\",\"da\":\"translations\",\"e\":\"2023-08-09T00:00:00.0000000\",\"h\":0,\"i\":\"medium.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: A Multi-Agent Framework Revolutionizing Software ... - Medium\",\"u\":\"https://medium.com/@korolalexei/metagpt-a-multi-agent-framework-revolutionizing-software-development-f585fe1aa950\"},{\"a\":\"You can check this by using:</span>\npython --version\n\n<span class=\"pl-c\"><span class=\"pl-c\">#</span> Step 3: Clone the repository to your local machine, and install it.</span>\ngit clone https://github.com/geekan/metagpt\n<span class=\"pl-c1\">cd</span> metagpt\npython setup.py install</pre></div>\n<h3 tabindex=\"-1\" dir=\"auto\"><a id=\...\",\"ae\":null,\"b\":\"gh\\tGitHub\\tgithub.com\",\"c\":\"https://github.com/PlaiD3/MetaGPT/blob/main/README.md\",\"d\":\"github.com/PlaiD3/MetaGPT/blob/main/README.md\",\"da\":\"\",\"h\":0,\"i\":\"github.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Multi-Agent Meta Programming Framework - GitHub\",\"u\":\"https://github.com/PlaiD3/MetaGPT/blob/main/README.md\"},{\"a\":\"\\u2014 MetaGPT is a multi-agent framework that enables collaboration among AI agents to tackle complex tasks and achieve collective intelligence. How does MetaGPT work? \\u2014 MetaGPT assigns specific roles to GPT agents based on their strengths and expertise, allowing them to collaborate, communicate, and share information to effectively tackle ...\",\"ae\":null,\"c\":\"https://eightify.app/summary/computer-science-and-technology/metagpt-advanced-autonomous-ai-agents-installation-tutorial\",\"d\":\"eightify.app/summary/computer-science-and-technology/metagpt-advanced-autonomous-ai-agents-installation-tutorial\",\"da\":\"\",\"h\":0,\"i\":\"eightify.app\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Advanced Autonomous AI Agents Installation Tutorial\",\"u\":\"https://eightify.app/summary/computer-science-and-technology/metagpt-advanced-autonomous-ai-agents-installation-tutorial\"},{\"a\":\"Therefore, we introduce MetaGPT, an innovative framework that incorporates efficient human workflows as a meta programming approach into LLM-based multi-agent collaboration. Specifically, MetaGPT encodes Standardized Operating Procedures (SOPs) into prompts to enhance structured coordination. ... SOPs act as a meta-function, taking the team and ...\",\"ae\":null,\"b\":\"arx\\tarXiv.org\\tarxiv.org\",\"c\":\"https://ar5iv.labs.arxiv.org/html/2308.00352\",\"d\":\"ar5iv.labs.arxiv.org/html/2308.00352\",\"da\":\"translations\",\"e\":\"2023-09-05T00:00:00.0000000\",\"h\":0,\"i\":\"ar5iv.labs.arxiv.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Meta Programming for Multi-Agent Collaborative Framework\",\"u\":\"https://ar5iv.labs.arxiv.org/html/2308.00352\"},{\"a\":\"Discover MetaGPT, a cutting-edge technology that harnesses Standardized Operating Procedures (SOPs) to orchestrate Large Language Model (LLM)-driven multi-agent systems, revolutionizing software development and collaborative task resolution. Explore its key features, delve into the core mechanisms, and learn how it enhances collaboration efficiency.\",\"ae\":null,\"c\":\"https://www.freegpttools.org/metagpt\",\"d\":\"www.freegpttools.org/metagpt\",\"da\":\"\",\"h\":0,\"i\":\"www.freegpttools.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Unlocking the Power of MetaGPT: A Multi-Agent Framework for Complex ...\",\"u\":\"https://www.freegpttools.org/metagpt\"},{\"a\":\"Message Function: Retained for event notification, weakened data transportation. Configuration Optimization: Default to gpt-4-1106-preview. ~/.metagpt for highest priority config, reading config.yaml. METAGPT_PROJECT_ROOT for workspace path specification. project_name specification via command line, generated by ProductManager. CLI Support\",\"ae\":null,\"b\":\"gh\\tGitHub\\tgithub.com\",\"c\":\"https://github.com/geekan/MetaGPT/releases\",\"d\":\"github.com/geekan/MetaGPT/releases\",\"da\":\"\",\"h\":0,\"i\":\"github.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Releases \\u00b7 geekan/MetaGPT \\u00b7 GitHub\",\"u\":\"https://github.com/geekan/MetaGPT/releases\"},{\"a\":\"SOPs act as a meta-function here, coordinating agents to auto-generate code based on defined inputs. In simple terms, it's as if you've turned a highly coordinated team of software engineers into an adaptable, intelligent software system. ... MetaGPT's architecture is divided into two layers: the Foundational Components Layer and the ...\",\"ae\":null,\"c\":\"https://theventurecation.com/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\"d\":\"theventurecation.com/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\"da\":\"\",\"e\":\"2023-09-11T00:00:00.0000000\",\"h\":0,\"i\":\"theventurecation.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Complete Guide to the Best AI Agent Available Right Now\",\"u\":\"https://theventurecation.com/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\"},{\"a\":\"Today, we are excited to take the next significant step forward and introduce a new Copilot key to Windows 11 PCs. In this new year, we will be ushering in a significant shift toward a more personal and intelligent computing future where AI will be seamlessly woven into Windows from the system, to the silicon, to the hardware.\",\"ae\":null,\"c\":\"https://blogs.windows.com/windowsexperience/2024/01/04/introducing-a-new-copilot-key-to-kick-off-the-year-of-ai-powered-windows-pcs/\",\"d\":\"blogs.windows.com/windowsexperience/2024/01/04/introducing-a-new-copilot-key-to-kick-off-the-year-of-ai-powered-windows-pcs/\",\"da\":\"translations\",\"e\":\"2024-01-04T00:00:00.0000000\",\"h\":0,\"i\":\"blogs.windows.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Introducing a new Copilot key to kick off the year of AI-powered ...\",\"u\":\"https://blogs.windows.com/windowsexperience/2024/01/04/introducing-a-new-copilot-key-to-kick-off-the-year-of-ai-powered-windows-pcs/\"},{\"a\":\"11 Cosmic Contingencies About 600,000 words between ChatGPT and I later. Please see pinned post on profile for modification of the Einstein field equations including contributions from quantum ram...\",\"ae\":null,\"c\":\"https://www.instagram.com/richfieldmusic/p/C1bpt1eucbU/\",\"d\":\"www.instagram.com/richfieldmusic/p/C1bpt1eucbU/\",\"da\":\"\",\"h\":0,\"i\":\"www.instagram.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"\\u042f\\u13c6\\u13df\\u13bb\\u0192\\u13c6\\u13ac\\u13de\\u13a0 on Instagram\",\"u\":\"https://www.instagram.com/richfieldmusic/p/C1bpt1eucbU/\"},{\"n\":\"/d.js?q=The%20function%20of%20MetaGPT&kl=wt-wt&l=wt-wt&p=&s=25&ex=-1&ct=US&sp=0&vqd=4-148519746540767190220111387879117509726\"}]);DDG.duckbar.load('images');DDG.duckbar.load('news');DDG.duckbar.load('videos');DDG.duckbar.loadModule('related_searches');if (DDG.pageLayout) DDG.pageLayout.initialize({\"mainline\":{\"items\":[[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"]]}}, { start: 0 });DDG.deep.emit(\"load:completed\");", + "curl-cffi-POST-https://duckduckgo.com-{\"data\": {\"q\": \"What llm MetaGPT support\"}}": "What llm MetaGPT support at DuckDuckGo
", + "curl-cffi-GET-https://links.duckduckgo.com/d.js-{\"params\": {\"bing_market\": \"wt-WT\", \"df\": null, \"ex\": \"-1\", \"kl\": \"wt-wt\", \"l\": \"wt-wt\", \"q\": \"What llm MetaGPT support\", \"s\": \"0\", \"sp\": \"0\", \"vqd\": \"4-73487995398881375915343809280473758117\"}}": "if (DDG.deep && DDG.deep.setUpstream) DDG.deep.setUpstream(\"bingv7aa\");DDG.deep.bn={'ivc':1};if (DDG.pageLayout) DDG.pageLayout.load('a',[], {\"page_load_url\":\"https://duckduckgo.com/y.js?iurl=%7B2%7DIG%3D0C43B15D5A884367BD85DF1F28ABDA06%26CID%3D26CF35F2E42B60EF229421F4E5D6611D%26Type%3DEvent.CPT%26DATA%3D0\"});DDG.deep.signalSummary = \"\";DDG.inject('DDG.Data.languages.resultLanguages', {\"en\":[\"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\"https://docs.deepwisdom.ai/main/en/guide/tutorials/integration_with_open_llm.html\",\"https://www.straight.com/guides/software/a-complete-guide-to-metagpt-the-best-ai-agent-available-now/\",\"https://mathaware.org/ai/metagpt-encoding-sops-with-llm-agents/\",\"https://arxiv.org/abs/2308.00352\",\"https://github.com/geekan/MetaGPT\",\"https://www.louisbouchard.ai/metagpt/\",\"https://lablab.ai/blog/this-week-in-ai-exploring-the-latest-from-metagpt-and-gpt4-and-more\",\"https://www.linkedin.com/pulse/what-metagpt-llm-agents-collaborating-solve-complex-bouchard-\",\"https://towardsai.net/p/machine-learning/what-is-metagpt-llm-agents-collaborating-to-solve-complex-tasks\",\"https://medium.com/mlearning-ai/metagpt-multi-agent-harmony-for-complex-problem-solving-97bcb8f3fe94\",\"https://medium.com/@yousra.aoudi/navigating-the-future-metagpts-innovative-approach-to-multi-agent-collaboration-ed1cc5835011\",\"https://openreview.net/forum?id=VtmBAGCN7o\",\"https://hackernoon.com/autogpt-langchain-deep-lake-metagpt-building-the-ultimate-llm-app\",\"https://generativeai.pub/analyzing-an-exciting-generative-ai-research-called-metagpt-2106385312db\",\"https://ar5iv.labs.arxiv.org/html/2308.00352\",\"https://louisbouchard.substack.com/p/what-is-metagpt-llm-agents-collaborating\",\"https://developer.nvidia.com/blog/supercharging-llm-applications-on-windows-pcs-with-nvidia-rtx-systems/\",\"https://github.com/geekan/MetaGPT/releases\",\"https://github.com/geekan/MetaGPT-docs/blob/main/src/en/guide/tutorials/integration_with_open_llm.md\",\"https://mathaware.org/ai/empowering-ai-with-llm-based-agents-metagpt-framework-transforms-human-sops/\",\"https://stackshare.io/metagpt\",\"https://medium.com/gta-generative-tech-advances/metagpt-an-interesting-approach-to-multi-agent-collaboration-5ace263c4fd8\",\"https://nvidianews.nvidia.com/news/generative-ai-rtx-pcs-and-workstations\",\"https://www.intel.com/content/www/us/en/developer/articles/technical/finetuning-llms-on-intel-gpus-using-bigdl-llm.html\",\"https://arxiv.org/abs/2401.05778\",\"https://www.ft.com/content/116f3541-bf2f-483e-a36a-ed3618548f9b\"],\"zh-CN\":[\"https://community.modelscope.cn/659cb258d4226e0eb42708e5.html\"]});DDG.deep.pageLayoutSummary = \"w29\";DDG.inject('DDG.Data.languages.adLanguages', {});if (DDG.pageLayout) DDG.pageLayout.load('d',[{\"a\":\"Enter MetaGPT \\u2014 a Multi-agent system that utilizes Large Language models by Sirui Hong fuses Standardized Operating Procedures (SOPs) with LLM-based multi-agent systems. This emerging paradigm disrupts the existing limitations of LLMs in fostering effective collaboration and task decomposition in complex, real-world applications.\",\"ae\":null,\"c\":\"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\"d\":\"www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\",\"da\":\"\",\"e\":\"2023-09-11T00:00:00.0000000\",\"h\":0,\"i\":\"www.unite.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Complete Guide to the Best AI Agent Available Right Now\",\"u\":\"https://www.unite.ai/metagpt-complete-guide-to-the-best-ai-agent-available-right-now/\"},{\"a\":\"The methods of integrating open source LLM and integrating some non-openai closed source models (such as Baidu Wenxinyiyan, iFLYTEK Spark, Zhipu ChatGLM, etc.) are similar, the main difference is the configuration. For details on the configuration of other closed-source LLMs, please refer to other LLM configuration documents under the online ...\",\"ae\":null,\"c\":\"https://docs.deepwisdom.ai/main/en/guide/tutorials/integration_with_open_llm.html\",\"d\":\"docs.deepwisdom.ai/main/en/guide/tutorials/integration_with_open_llm.html\",\"da\":\"\",\"e\":\"2023-12-21T00:00:00.0000000\",\"h\":0,\"i\":\"docs.deepwisdom.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Integration with open LLM | MetaGPT\",\"u\":\"https://docs.deepwisdom.ai/main/en/guide/tutorials/integration_with_open_llm.html\"},{\"a\":\"MetaGPT is a multi-agent system that utilizes Large Language Models (LLMs) to perform complex tasks. It is designed to overcome the limitations of LLMs in fostering effective collaboration and...\",\"ae\":null,\"c\":\"https://www.straight.com/guides/software/a-complete-guide-to-metagpt-the-best-ai-agent-available-now/\",\"d\":\"www.straight.com/guides/software/a-complete-guide-to-metagpt-the-best-ai-agent-available-now/\",\"da\":\"\",\"e\":\"2023-12-13T00:00:00.0000000\",\"h\":0,\"i\":\"www.straight.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"A Complete Guide to MetaGPT: The Best AI Agent Available Now\",\"u\":\"https://www.straight.com/guides/software/a-complete-guide-to-metagpt-the-best-ai-agent-available-now/\"},{\"a\":\"The advent of MetaGPT and similar LLM agents represents a significant leap forward in the realm of process management and optimization. By addressing the complexities of encoding SOPs and acknowledging the nuanced nature of their integration, these intelligent systems are poised to redefine operational efficiency.\",\"ae\":null,\"c\":\"https://mathaware.org/ai/metagpt-encoding-sops-with-llm-agents/\",\"d\":\"mathaware.org/ai/metagpt-encoding-sops-with-llm-agents/\",\"da\":\"\",\"h\":0,\"i\":\"mathaware.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Encoding SOPs with LLM Agents | MathAware AI\",\"u\":\"https://mathaware.org/ai/metagpt-encoding-sops-with-llm-agents/\"},{\"a\":\"Remarkable progress has been made on automated problem solving through societies of agents based on large language models (LLMs). Existing LLM-based multi-agent systems can already solve simple dialogue tasks.\",\"ae\":null,\"b\":\"arx\\tarXiv.org\\tarxiv.org\",\"c\":\"https://arxiv.org/abs/2308.00352\",\"d\":\"arxiv.org/abs/2308.00352\",\"da\":\"translations\",\"e\":\"2023-08-01T00:00:00.0000000\",\"h\":0,\"i\":\"arxiv.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework\",\"u\":\"https://arxiv.org/abs/2308.00352\"},{\"a\":\"agent multi-agent gpt hacktoberfest llm metagpt Resources. Readme License. MIT license Activity. Stars. 33.2k stars Watchers. 802 watching Forks. 3.9k forks Report repository Releases 11. Patch release: v0.6.4 Latest Jan 12, 2024 + 10 releases Packages 0. No packages published . Contributors 55 + 41 contributors Languages.\",\"ae\":null,\"b\":\"gh\\tGitHub\\tgithub.com\",\"c\":\"https://github.com/geekan/MetaGPT\",\"d\":\"github.com/geekan/MetaGPT\",\"da\":\"\",\"h\":0,\"i\":\"github.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: The Multi-Agent Framework - GitHub\",\"u\":\"https://github.com/geekan/MetaGPT\"},{\"a\":\"Aug 27, 2023 \\u2022 6 min read Watch the video! MetaGPT: Redefining Multi-Agent Collaboration for Complex Tasks Watch on Thanks to GPT and the recent large language models, we've seen the popularization of a new type of AI-based system\\u2026 agents. An agent is basically an AI model like ChatGPT that can access and interact with one or more applications.\",\"ae\":null,\"c\":\"https://www.louisbouchard.ai/metagpt/\",\"d\":\"www.louisbouchard.ai/metagpt/\",\"da\":\"\",\"e\":\"2023-08-27T00:00:00.0000000\",\"h\":0,\"i\":\"www.louisbouchard.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Mitigating AI Hallucinations: Exploring MetaGPT's Collaborative Framework\",\"u\":\"https://www.louisbouchard.ai/metagpt/\"},{\"a\":\"MetaGPT is a groundbreaking multi-agent framework that is transforming the way software development is approached. By taking a single line of requirement as input, MetaGPT outputs a comprehensive array of development components, including user stories, competitive analysis, requirements, data structures, APIs, and documents.\",\"ae\":null,\"c\":\"https://lablab.ai/blog/this-week-in-ai-exploring-the-latest-from-metagpt-and-gpt4-and-more\",\"d\":\"lablab.ai/blog/this-week-in-ai-exploring-the-latest-from-metagpt-and-gpt4-and-more\",\"da\":\"\",\"e\":\"2023-08-11T00:00:00.0000000\",\"h\":0,\"i\":\"lablab.ai\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"This Week in AI: Exploring the Latest from MetaGPT and GPT-4 and more..\",\"u\":\"https://lablab.ai/blog/this-week-in-ai-exploring-the-latest-from-metagpt-and-gpt4-and-more\"},{\"a\":\"This iteration focuses on MetaGPT, a new approach to improving collaborations between AI agents (e.g., ChatGPT-based entities mimicking human roles). ... 3D-LLM Unleashes Language Models into the ...\",\"ae\":null,\"b\":\"li\\tLinkedIn\\twww.linkedin.com\",\"c\":\"https://www.linkedin.com/pulse/what-metagpt-llm-agents-collaborating-solve-complex-bouchard-\",\"d\":\"www.linkedin.com/pulse/what-metagpt-llm-agents-collaborating-solve-complex-bouchard-\",\"da\":\"\",\"e\":\"2023-08-28T00:00:00.0000000\",\"h\":0,\"i\":\"www.linkedin.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"What is MetaGPT? LLM Agents Collaborating to Solve Complex Tasks - LinkedIn\",\"u\":\"https://www.linkedin.com/pulse/what-metagpt-llm-agents-collaborating-solve-complex-bouchard-\"},{\"a\":\"Latest Machine Learning What is MetaGPT? LLM Agents Collaborating to Solve Complex Tasks August 28, 2023 Last Updated on August 29, 2023 by Editorial Team Author (s): Louis Bouchard Watch the video! This member-only story is on us. Upgrade to access all of Medium. Originally published on louisbouchard.ai, read it 2 days before on my blog!\",\"ae\":null,\"c\":\"https://towardsai.net/p/machine-learning/what-is-metagpt-llm-agents-collaborating-to-solve-complex-tasks\",\"d\":\"towardsai.net/p/machine-learning/what-is-metagpt-llm-agents-collaborating-to-solve-complex-tasks\",\"da\":\"\",\"e\":\"2023-08-29T00:00:00.0000000\",\"h\":0,\"i\":\"towardsai.net\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"What is MetaGPT? LLM Agents Collaborating to Solve Complex Tasks\",\"u\":\"https://towardsai.net/p/machine-learning/what-is-metagpt-llm-agents-collaborating-to-solve-complex-tasks\"},{\"a\":\"You know how those multi-agent systems powered by Large Language Models (LLMs) have the potential to mimic and jazz up human workflows? But, the real world's a tangled place, and these systems...\",\"ae\":null,\"c\":\"https://medium.com/mlearning-ai/metagpt-multi-agent-harmony-for-complex-problem-solving-97bcb8f3fe94\",\"d\":\"medium.com/mlearning-ai/metagpt-multi-agent-harmony-for-complex-problem-solving-97bcb8f3fe94\",\"da\":\"\",\"e\":\"2023-08-09T00:00:00.0000000\",\"h\":0,\"i\":\"medium.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Multi-Agent Harmony for Complex Problem Solving\",\"u\":\"https://medium.com/mlearning-ai/metagpt-multi-agent-harmony-for-complex-problem-solving-97bcb8f3fe94\"},{\"a\":\"T he essence of MetaGPT is the seamless integration of SOPs to craft a highly coordinated LLM-based multi-agent ecosystem. With a focus on emulating human-like roles and intricate workflows, it...\",\"ae\":null,\"c\":\"https://medium.com/@yousra.aoudi/navigating-the-future-metagpts-innovative-approach-to-multi-agent-collaboration-ed1cc5835011\",\"d\":\"medium.com/@yousra.aoudi/navigating-the-future-metagpts-innovative-approach-to-multi-agent-collaboration-ed1cc5835011\",\"da\":\"translations\",\"e\":\"2023-08-10T00:00:00.0000000\",\"h\":0,\"i\":\"medium.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Navigating the Future: MetaGPT's Innovative Approach to ... - Medium\",\"u\":\"https://medium.com/@yousra.aoudi/navigating-the-future-metagpts-innovative-approach-to-multi-agent-collaboration-ed1cc5835011\"},{\"a\":\"Recently, remarkable progress has been made on automated problem solving through societies of agents based on large language models (LLMs). Previous LLM-based multi-agent systems can already solve simple dialogue tasks.\",\"ae\":null,\"c\":\"https://openreview.net/forum?id=VtmBAGCN7o\",\"d\":\"openreview.net/forum?id=VtmBAGCN7o\",\"da\":\"\",\"e\":\"2023-09-22T00:00:00.0000000\",\"h\":0,\"i\":\"openreview.net\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Meta Programming for Multi-Agent Collaborative Framework\",\"u\":\"https://openreview.net/forum?id=VtmBAGCN7o\"},{\"a\":\"Deep Lake is a remarkable answer to the problem of storing gigabytes of data for LLMs \\u2014 efficiently, easily, and practically. Its unique configuration allows the optimal usage of finances. OpenAI's LLM Operational Cost Daily is on average 700,000 USD a day. Some are even predicting bankruptcy for the company.\",\"ae\":null,\"c\":\"https://hackernoon.com/autogpt-langchain-deep-lake-metagpt-building-the-ultimate-llm-app\",\"d\":\"hackernoon.com/autogpt-langchain-deep-lake-metagpt-building-the-ultimate-llm-app\",\"da\":\"\",\"e\":\"2023-08-29T00:00:00.0000000\",\"h\":0,\"i\":\"hackernoon.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Deep Lake \\u2014 MetaGPT: Building the Ultimate LLM App - HackerNoon\",\"u\":\"https://hackernoon.com/autogpt-langchain-deep-lake-metagpt-building-the-ultimate-llm-app\"},{\"a\":\"The MetaGPT approach showcases its ability to decompose highlevel tasks into detailed actionable components handled by distinct roles (ProductManager, Architect, ProjectManager, Engineer, QA Engineer), thereby facilitating role-specific expertise and coordination. This methodology mirrors human software development teams.\",\"ae\":null,\"c\":\"https://generativeai.pub/analyzing-an-exciting-generative-ai-research-called-metagpt-2106385312db\",\"d\":\"generativeai.pub/analyzing-an-exciting-generative-ai-research-called-metagpt-2106385312db\",\"da\":\"translations\",\"e\":\"2023-08-14T00:00:00.0000000\",\"h\":0,\"i\":\"generativeai.pub\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Analyzing an exciting Generative AI research called MetaGPT.\",\"u\":\"https://generativeai.pub/analyzing-an-exciting-generative-ai-research-called-metagpt-2106385312db\"},{\"a\":\"In this work, we present MetaGPT, a promising framework for collaborative agents using SOPs that leverages LLMs to mimic efficient human workflows. MetaGPT is a meta programming technology that utilizes SOPs to coordinate LLM-based multi-agent systems. Specifically, to encode SOPs into prompts, MetaGPT manages multi-agents through role ...\",\"ae\":null,\"b\":\"arx\\tarXiv.org\\tarxiv.org\",\"c\":\"https://ar5iv.labs.arxiv.org/html/2308.00352\",\"d\":\"ar5iv.labs.arxiv.org/html/2308.00352\",\"da\":\"translations\",\"e\":\"2023-09-05T00:00:00.0000000\",\"h\":0,\"i\":\"ar5iv.labs.arxiv.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: Meta Programming for Multi-Agent Collaborative Framework\",\"u\":\"https://ar5iv.labs.arxiv.org/html/2308.00352\"},{\"a\":\"MetaGPT: Meta Programming for Multi-Agent Collaborative Framework. Topsakal, O., & Akinci, T.C. (2023). Creating Large Language Model Applications Utilizing LangChain: A Primer on Developing LLM ...\",\"ae\":null,\"c\":\"https://medium.com/technology-hits/autogpt-langchain-deep-lake-metagpt-a-revolutionary-framework-for-building-advanced-ai-e2c579d86494\",\"d\":\"medium.com/technology-hits/autogpt-langchain-deep-lake-metagpt-a-revolutionary-framework-for-building-advanced-ai-e2c579d86494\",\"da\":\"\",\"e\":\"2023-08-28T00:00:00.0000000\",\"h\":0,\"i\":\"medium.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"AutoGPT \\u2014 LangChain \\u2014 Deep Lake \\u2014 MetaGPT: A ... - Medium\",\"u\":\"https://medium.com/technology-hits/autogpt-langchain-deep-lake-metagpt-a-revolutionary-framework-for-building-advanced-ai-e2c579d86494\"},{\"a\":\"Louis , starting with a trending topic: AI agents! This iteration focuses on MetaGPT, a new approach to improving collaborations between AI agents (e.g., ChatGPT-based entities mimicking human roles).\",\"ae\":null,\"c\":\"https://louisbouchard.substack.com/p/what-is-metagpt-llm-agents-collaborating\",\"d\":\"louisbouchard.substack.com/p/what-is-metagpt-llm-agents-collaborating\",\"da\":\"\",\"e\":\"2023-08-28T00:00:00.0000000\",\"h\":0,\"i\":\"louisbouchard.substack.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"What is MetaGPT? LLM Agents Collaborating to Solve Complex Tasks\",\"u\":\"https://louisbouchard.substack.com/p/what-is-metagpt-llm-agents-collaborating\"},{\"a\":\"Today, LLM-powered applications are running predominantly in the cloud. However, many use cases that would benefit from running LLMs locally on Windows PCs, including gaming, creativity, productivity, and developer experiences. AT CES 2024, NVIDIA announced several developer tools to accelerate LLM inference and development on NVIDIA RTX ...\",\"ae\":null,\"c\":\"https://developer.nvidia.com/blog/supercharging-llm-applications-on-windows-pcs-with-nvidia-rtx-systems/\",\"d\":\"developer.nvidia.com/blog/supercharging-llm-applications-on-windows-pcs-with-nvidia-rtx-systems/\",\"da\":\"\",\"e\":\"2024-01-08T00:00:00.0000000\",\"h\":0,\"i\":\"developer.nvidia.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Supercharging LLM Applications on Windows PCs with NVIDIA RTX Systems\",\"u\":\"https://developer.nvidia.com/blog/supercharging-llm-applications-on-windows-pcs-with-nvidia-rtx-systems/\"},{\"a\":\"Supported Ollama as underlying LLM #603 by @better629; Enabled MetaGPT to be used as a dependency for web applications, such as https: ... PIP Support: pip install metagpt is now available for installing and using metagpt, enabling direct access to the command-line version of metagpt.\",\"ae\":null,\"b\":\"gh\\tGitHub\\tgithub.com\",\"c\":\"https://github.com/geekan/MetaGPT/releases\",\"d\":\"github.com/geekan/MetaGPT/releases\",\"da\":\"\",\"h\":0,\"i\":\"github.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Releases \\u00b7 geekan/MetaGPT \\u00b7 GitHub\",\"u\":\"https://github.com/geekan/MetaGPT/releases\"},{\"a\":\"If you want to support <code>http: //ip:11434/api/chat</code>, you can do as follows:</p>\n<div class=\"highlight highlight-source-shell notranslate position-relative overflow-auto\" dir=\"auto\" data-snippet-clipboard-copy-content=\"service ollama stop\n\nOLLAMA_HOST=0.0.0.0 OLLAMA_ORIGINS=* ollama serve # one terminal\n\nollama run llama2 # ot...\",\"ae\":null,\"b\":\"gh\\tGitHub\\tgithub.com\",\"c\":\"https://github.com/geekan/MetaGPT-docs/blob/main/src/en/guide/tutorials/integration_with_open_llm.md\",\"d\":\"github.com/geekan/MetaGPT-docs/blob/main/src/en/guide/tutorials/integration_with_open_llm.md\",\"da\":\"\",\"h\":0,\"i\":\"github.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Integration with open LLM - GitHub\",\"u\":\"https://github.com/geekan/MetaGPT-docs/blob/main/src/en/guide/tutorials/integration_with_open_llm.md\"},{\"a\":\"With LLM-based agents empowered by the MetaGPT framework, companies can streamline their workflows and improve productivity. These agents can assist employees by automating repetitive tasks, generating reports, and even coming up with creative solutions to problems. The MetaGPT framework allows for fine-tuning the LLM-based agents based on ...\",\"ae\":null,\"c\":\"https://mathaware.org/ai/empowering-ai-with-llm-based-agents-metagpt-framework-transforms-human-sops/\",\"d\":\"mathaware.org/ai/empowering-ai-with-llm-based-agents-metagpt-framework-transforms-human-sops/\",\"da\":\"\",\"h\":0,\"i\":\"mathaware.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Empowering AI with LLM-based Agents: MetaGPT Framework Transforms Human ...\",\"u\":\"https://mathaware.org/ai/empowering-ai-with-llm-based-agents-metagpt-framework-transforms-human-sops/\"},{\"a\":\"Check out popular companies that use MetaGPT and some tools that integrate with MetaGPT. ... On top of llm, there is a CLI application, llm-cli, which provides a convenient interface for running inference on supported models. Chroma. It is an open-source embedding database. Chroma makes it easy to build LLM apps by making knowledge, facts, and ...\",\"ae\":null,\"c\":\"https://stackshare.io/metagpt\",\"d\":\"stackshare.io/metagpt\",\"da\":\"\",\"h\":0,\"i\":\"stackshare.io\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT - Reviews, Pros & Cons | Companies using MetaGPT - StackShare\",\"u\":\"https://stackshare.io/metagpt\"},{\"a\":\"MetaGPT is the maestro who brings harmony to this chaos. By encoding Standardized Operating Procedures (SOPs) into prompts, MetaGPT ensures structured collaboration akin to a well-rehearsed ...\",\"ae\":null,\"c\":\"https://medium.com/gta-generative-tech-advances/metagpt-an-interesting-approach-to-multi-agent-collaboration-5ace263c4fd8\",\"d\":\"medium.com/gta-generative-tech-advances/metagpt-an-interesting-approach-to-multi-agent-collaboration-5ace263c4fd8\",\"da\":\"\",\"e\":\"2023-08-15T00:00:00.0000000\",\"h\":0,\"i\":\"medium.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MetaGPT: An Interesting Approach to Multi-Agent Collaboration\",\"u\":\"https://medium.com/gta-generative-tech-advances/metagpt-an-interesting-approach-to-multi-agent-collaboration-5ace263c4fd8\"},{\"a\":\"NVIDIA recently extended TensorRT to text-based applications with TensorRT-LLM for Windows, an open-source library for accelerating LLMs. The latest update to TensorRT-LLM, available now, adds Phi-2 to the growing list of pre-optimized models for PC, which run up to 5x faster compared to other inference backends.\",\"ae\":null,\"c\":\"https://nvidianews.nvidia.com/news/generative-ai-rtx-pcs-and-workstations\",\"d\":\"nvidianews.nvidia.com/news/generative-ai-rtx-pcs-and-workstations\",\"da\":\"\",\"e\":\"2024-01-08T00:00:00.0000000\",\"h\":0,\"i\":\"nvidianews.nvidia.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"NVIDIA Brings Generative AI to Millions, With Tensor Core GPUs, LLMs ...\",\"u\":\"https://nvidianews.nvidia.com/news/generative-ai-rtx-pcs-and-workstations\"},{\"a\":\"The BigDL LLM library extends support for fine-tuning LLMs to a variety of Intel GPUs, including the Intel\\u00ae Data Center GPU Flex 170 and Intel\\u00ae Arc\\u2122 series graphics. Specifically, using the Intel\\u00ae Data Center GPU Flex 170 hardware as an example, you can complete the fine-tuning of the Llama 2 7B model in approximately 2 hours on a single ...\",\"ae\":null,\"b\":\"ark\\tIntel Processor Specification\\tark.intel.com\",\"c\":\"https://www.intel.com/content/www/us/en/developer/articles/technical/finetuning-llms-on-intel-gpus-using-bigdl-llm.html\",\"d\":\"www.intel.com/content/www/us/en/developer/articles/technical/finetuning-llms-on-intel-gpus-using-bigdl-llm.html\",\"da\":\"\",\"e\":\"2023-12-22T00:00:00.0000000\",\"h\":0,\"i\":\"www.intel.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Fine-tuning Llama 2 models on Intel\\u00ae Data Center GPUs using BigDL LLM\",\"u\":\"https://www.intel.com/content/www/us/en/developer/articles/technical/finetuning-llms-on-intel-gpus-using-bigdl-llm.html\"},{\"a\":\"Large language models (LLMs) have strong capabilities in solving diverse natural language processing tasks. However, the safety and security issues of LLM systems have become the major obstacle to their widespread application. Many studies have extensively investigated risks in LLM systems and developed the corresponding mitigation strategies. Leading-edge enterprises such as OpenAI, Google ...\",\"ae\":null,\"b\":\"arx\\tarXiv.org\\tarxiv.org\",\"c\":\"https://arxiv.org/abs/2401.05778\",\"d\":\"arxiv.org/abs/2401.05778\",\"da\":\"translations\",\"e\":\"2024-01-11T09:29:00.0000000\",\"h\":0,\"i\":\"arxiv.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"[2401.05778] Risk Taxonomy, Mitigation, and Assessment Benchmarks of ...\",\"u\":\"https://arxiv.org/abs/2401.05778\"},{\"a\":\"It carries on regardless and gives a definitive answer anyway," the BIS paper noted. The tendency of LLMs to make hilariously confident mistakes (eg inventing case law) is sometimes innocuously ...\",\"ae\":null,\"c\":\"https://www.ft.com/content/116f3541-bf2f-483e-a36a-ed3618548f9b\",\"d\":\"www.ft.com/content/116f3541-bf2f-483e-a36a-ed3618548f9b\",\"da\":\"\",\"e\":\"2024-01-05T12:13:51.0000000\",\"h\":0,\"i\":\"www.ft.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"BIS vs LLM - Financial Times\",\"u\":\"https://www.ft.com/content/116f3541-bf2f-483e-a36a-ed3618548f9b\"},{\"a\":\"\\u5927\\u578b\\u8bed\\u8a00\\u6a21\\u578b\\uff08LLM\\uff09\\u7684\\u51fa\\u73b0\\u5e26\\u706b\\u4e86Agent\\u3002\\u5229\\u7528LLM\\u7406\\u89e3\\u4eba\\u7c7b\\u610f\\u56fe\\u3001\\u751f\\u6210\\u590d\\u6742\\u8ba1\\u5212\\u5e76\\u4e14\\u80fd\\u591f\\u81ea\\u4e3b\\u884c\\u52a8\\u7684\\u80fd\\u529b\\u3002Agent\\u5177\\u6709\\u65e0\\u4e0e\\u4f26\\u6bd4\\u7684\\u80fd\\u529b\\uff0c\\u80fd\\u591f\\u505a\\u51fa\\u7c7b\\u4f3c\\u4e8e\\u4eba\\u7c7b\\u590d\\u6742\\u6027\\u7684\\u51b3\\u7b56\\u548c\\u5b8c\\u6210\\u4e00\\u4e9b\\u590d\\u6742\\u7684\\u5de5\\u4f5c\\u3002\\u76ee\\u524d\\u5e02\\u9762\\u4e0a\\u5df2\\u7ecf\\u51fa\\u73b0\\u975e\\u5e38\\u591a\\u5f97Agent\\u6846\\u67b6\\uff1aXAgent, AutoGPT\\u3001BabyAGI\\u3001CAMEL\\u3001MetaGPT\\u3001AutoGen\\u3001DSPy\\u3001AutoAgents\\u3001OpenAgents ...\",\"ae\":null,\"c\":\"https://community.modelscope.cn/659cb258d4226e0eb42708e5.html\",\"d\":\"community.modelscope.cn/659cb258d4226e0eb42708e5.html\",\"da\":\"translations\",\"e\":\"2024-01-09T00:00:00.0000000\",\"h\":0,\"i\":\"community.modelscope.cn\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"\\u5982\\u4f55\\u63d0\\u5347\\u5927\\u6a21\\u578bAgent\\u7684\\u80fd\\u529b \\u2014\\u2014LLM Agent\\u6846\\u67b6 modelscope-agent \\u5b9e\\u6218\",\"u\":\"https://community.modelscope.cn/659cb258d4226e0eb42708e5.html\"},{\"n\":\"/d.js?q=What%20llm%20MetaGPT%20support&kl=wt-wt&l=wt-wt&p=&s=29&ex=-1&ct=US&sp=0&vqd=4-73487995398881375915343809280473758117\"}]);DDG.duckbar.load('images');DDG.duckbar.load('news');DDG.duckbar.load('videos');DDG.duckbar.loadModule('related_searches');if (DDG.pageLayout) DDG.pageLayout.initialize({\"mainline\":{\"items\":[[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"]]}}, { start: 0 });DDG.deep.emit(\"load:completed\");", + "curl-cffi-POST-https://duckduckgo.com-{\"data\": {\"q\": \"dataiku\"}}": "dataiku at DuckDuckGo
", + "curl-cffi-POST-https://duckduckgo.com-{\"data\": {\"q\": \"datarobot\"}}": "datarobot at DuckDuckGo
", + "curl-cffi-GET-https://links.duckduckgo.com/d.js-{\"params\": {\"bing_market\": \"wt-WT\", \"df\": null, \"ex\": \"-1\", \"kl\": \"wt-wt\", \"l\": \"wt-wt\", \"q\": \"datarobot\", \"s\": \"0\", \"sp\": \"0\", \"vqd\": \"4-305438614248843041332096319235456803678\"}}": "DDG.search.altIsNavigational = 1;DDG.search.isNavigational = 1;if (DDG.deep && DDG.deep.setUpstream) DDG.deep.setUpstream(\"bingv7aa\");DDG.deep.bn={'ivc':1};if (DDG.pageLayout) DDG.pageLayout.load('a',[{\"a\":\"\\u9ad8\\u7cbe\\u5ea6\\u306a\\u6a5f\\u68b0\\u5b66\\u7fd2\\u30e2\\u30c7\\u30eb\\u3092\\u69cb\\u7bc9\\u3001\\u5b9f\\u88c5\\u3001\\u904b\\u7528\\u3002DataRobot\\u306f\\u793e\\u5185\\u30c7\\u30fc\\u30bf\\u304b\\u3089\\u65b0\\u3057\\u3044\\u4fa1\\u5024\\u3092\\u5275\\u9020\\u3057\\u307e\\u3059. DataRobot\\u306f\\u4f01\\u696d\\u306e\\u8ab2\\u984c\\u89e3\\u6c7a\\u306b\\u7279\\u5316\\u3002\\u610f\\u601d\\u6c7a\\u5b9a\\u306e\\u81ea\\u52d5\\u5316\\u304b\\u3089\\u9700\\u8981\\u4e88\\u6e2c\\u3001\\u8981\\u56e0\\u5206\\u6790\\u307e\\u3067\\u3053\\u306a\\u3059AI\\u30c4\\u30fc\\u30eb\",\"adext\":{\"callout\":{\"t\":\"Accelerate Time to Impact \\u00b7 Trust Your AI Models\",\"tid\":\"6\"},\"filterlinks\":{\"l\":[],\"tid\":\"\"},\"sitelinks\":{\"l\":[{\"snippet\":\"Explore the DataRobot AI Platform Get Started With a 30-Day Trial\",\"targetUrl\":\"https://duckduckgo.com/y.js?ad_domain=datarobot.com&ad_provider=bingv7aa&ad_type=txad&eddgt=4t4PwhnMmcY7atJfp41CAw%3D%3D&rut=e62e2cb72625106a43222549f786956db024f90f62131cbc4a16a28c8968f74c&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De8CWdnSSqYRxodv2hK6GQYijVUCUw3ed58BZbCDycwDIpFPyWnwhXm4uvKBLQcGynMpDz_AvA1H9RQnhR9goEmV9Ya3_wHg3eI0DOkTWJinjiudX5vGJJNWufOQNDYFdDHFVXChhUck3LNxc3D4UakPcG3Fgqa6IzJHoPcJWXkWGOqxuYGxVGkxJO3aqSinbclEqi5d0At_9OjMeDihix9q58SuTw%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuZGF0YXJvYm90LmNvbSUyZnRyaWFsJTJmJTNmdXRtX21lZGl1bSUzZHNlYXJjaCUyNnV0bV9zb3VyY2UlM2RiaW5nJTI2dXRtX2NhbXBhaWduJTNkRnJlZVRyaWFsMjAyM1dXMDgxNkdQU2FkZXh0JTI2Y2FtcGFpZ25pZCUzZDUzMDcwODA5OSUyNmFkZ3JvdXBpZCUzZDEzNTAyMDI3NzQyMTc2OTglMjZhZGlkJTNkJTI2bXNjbGtpZCUzZDFlNjcxOTNiY2UzZDFlNzBhZTQ2N2M5ODMxM2UxOTI0%26rlid%3D1e67193bce3d1e70ae467c98313e1924&vqd=4-304301655096002530435728103296846286110&iurl=%7B1%7DIG%3DEFC9E56534984E60A01E1AD18F949C41%26CID%3D17342A5B7AD96D85041C3E5D7B5C6C74%26ID%3DDevEx%2C5065.1\",\"text\":\"DataRobot Free Trial\"},{\"snippet\":\"Unlock Your AI Success in 2023 Tips on the Path of Value-Driven AI\",\"targetUrl\":\"https://duckduckgo.com/y.js?ad_domain=datarobot.com&ad_provider=bingv7aa&ad_type=txad&eddgt=4t4PwhnMmcY7atJfp41CAw%3D%3D&rut=25d9bbfaa9cbfaaf08d9a6b444b633a6ce8d72930552b70b34d90df428188bf0&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De8qLHGkuDcZ8xxbn%2DtJPQw1jVUCUxi3KfzBWZ8dNgaK8%2DSoRj4nbRrRFfQVuiV7AIxRtZS4mXd7F1paHlN1BDqwOZNm7_O3fvHO59UXgQmhFNY1DQ7kMdZ%2Did8G0JnG_Kzxief1jx20D1mrfhnsLHb6_5GY1RZ4vzYXUnjih6Hh8nrgoRbxzYS5s8evGIka2E8%2Dq%2DIbZtX%2DEHSMg1sz2yxt770lqc%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuZGF0YXJvYm90LmNvbSUyZnJlc291cmNlcyUyZmFpc3VjY2VzczIwMjMlMmYlM2Z1dG1fbWVkaXVtJTNkc2VhcmNoJTI2dXRtX3NvdXJjZSUzZGJpbmclMjZ1dG1fY2FtcGFpZ24lM2RDb250ZW50MTBLZXlzdG9BSVN1Y2Nlc3MyMDIzV1cwNTIyR1BTYWRleHQlMjZ1dG1fdGVybSUzZGRhdGFyb2JvdCUyNnV0bV9jb250ZW50JTNkYWRfZXh0JTI2Y2FtcGFpZ25pZCUzZDUzMDcwODA5OSUyNmFkZ3JvdXBpZCUzZDEzNTAyMDI3NzQyMTc2OTglMjZhZGlkJTNkJTI2bXNjbGtpZCUzZDBiNTgwOTJlZGMzNjFlNGFiOTgyNmM4ZDIzMzViNTFm%26rlid%3D0b58092edc361e4ab9826c8d2335b51f&vqd=4-17900397268746114651881769682671895645&iurl=%7B1%7DIG%3DEFC9E56534984E60A01E1AD18F949C41%26CID%3D17342A5B7AD96D85041C3E5D7B5C6C74%26ID%3DDevEx%2C5067.1\",\"text\":\"10 Keys to AI Success\"},{\"snippet\":\"Our Platform Includes Four Fully Integrated Products. Read More.\",\"targetUrl\":\"https://duckduckgo.com/y.js?ad_domain=datarobot.com&ad_provider=bingv7aa&ad_type=txad&eddgt=4t4PwhnMmcY7atJfp41CAw%3D%3D&rut=9b50c0e6e845f83a4b6122341b7007b55cc637bf24bc814ebfbdb41c4570e842&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De80Tber2J5eyzs7ehxic3bQzVUCUyYDJpssb8FQM4q5TzHPQTbxhVuzWgr30VBdcQ%2Du_fAfiqmWEHQ13X%2DWe_zzqhxfJqe8TH1WdLsIIKUrxWqqxfZyPQuZ818htUh82k2s2Co_K3ZgklXSA%2Duj9j4sghZ155%2DCpGDXbSizpxOVw8TwgUDyW_ZxEZDxtS0Rk5iH4G6PuzQvhP02YdMD_rMZTOt42M%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuZGF0YXJvYm90LmNvbSUyZnByb2R1Y3QlMmYlM2ZjYW1wYWlnbmlkJTNkNTMwNzA4MDk5JTI2YWRncm91cGlkJTNkMTM1MDIwMjc3NDIxNzY5OCUyNmFkaWQlM2QlMjZtc2Nsa2lkJTNkZDQ0ZDJiNzQ0YjEzMWFmZTcyZDQ0ZWQ3YjQxMDY3MDI%26rlid%3Dd44d2b744b131afe72d44ed7b4106702&vqd=4-135405568356283083312160903053804829572&iurl=%7B1%7DIG%3DEFC9E56534984E60A01E1AD18F949C41%26CID%3D17342A5B7AD96D85041C3E5D7B5C6C74%26ID%3DDevEx%2C5069.1\",\"text\":\"Product Overview\"}],\"tid\":\"7\\t9[8]\\t11[10]\\t13[12]\",\"type\":\"EnhancedSiteLink\"},\"tid\":\"1\"},\"ae\":{\"callout\":[\"Accelerate Time to Impact \\u00b7 Trust Your AI Models\"]},\"c\":\"https://duckduckgo.com/y.js?ad_domain=datarobot.com&ad_provider=bingv7aa&ad_type=txad&eddgt=4t4PwhnMmcY7atJfp41CAw%3D%3D&rut=4e8bc6239074926ffa0595d2f0838a03d95a9f20653372406dff4b73bb47a802&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De87ziwl19iJFSSt9AnpU7A2TVUCUyhd2uAWuXw7lbtLG3tobaae8IKDFy1RftUyaK7hmjFguIcBXLMF0__8U%2DYMqp1lRmGet90G40qSPB8wuC4MyZjuA8D06WqXRZsdl4uyGEuNXLmrp1n4swCoXfw3cJtq1Sl1iqwNQBdH4Ev%2DJQUf54N5TQ274OfwCR8PMamVlYCWg%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuZGF0YXJvYm90LmNvbSUyZmpwJTJmbHAlMmZhaS1mb3ItYnVzaW5lc3MlMmYlM2Z1dG1fbWVkaXVtJTNkc2VhcmNoJTI2dXRtX3NvdXJjZSUzZGJpbmclMjZ1dG1fY2FtcGFpZ24lM2RERU1PMjAyM0FsbFByb2R1Y3RzSlAwNjI2QlBTJTI2dXRtX3Rlcm0lM2RkYXRhcm9ib3QlMjZ1dG1fY29udGVudCUzZERSX2JyYW5kZWRfcnNhJTI2Y2FtcGFpZ25pZCUzZDUzMDcwODA5OSUyNmFkZ3JvdXBpZCUzZDEzNTAyMDI3NzQyMTc2OTglMjZhZGlkJTNkJTI2bXNjbGtpZCUzZDdlMDA4NWE0OTQ3MzE5ODdlNzJjZTAwNzZiNzAxMmFm%26rlid%3D7e0085a494731987e72ce0076b7012af&vqd=4-129716528454193272263152190229962811420&iurl=%7B1%7DIG%3DEFC9E56534984E60A01E1AD18F949C41%26CID%3D17342A5B7AD96D85041C3E5D7B5C6C74%26ID%3DDevEx%2C5060.1\",\"d\":\"datarobot.com\",\"h\":0,\"i\":\"\",\"k\":0,\"m\":0,\"o\":\"\",\"p\":1,\"relevancy\":{\"abstract\":\"%E9%AB%98%E7%B2%BE%E5%BA%A6%E3%81%AA%E6%A9%9F%E6%A2%B0%E5%AD%A6%E7%BF%92%E3%83%A2%E3%83%87%E3%83%AB%E3%82%92%E6%A7%8B%E7%AF%89%E3%80%81%E5%AE%9F%E8%A3%85%E3%80%81%E9%81%8B%E7%94%A8%E3%80%82%3Cb%3EDataRobot%3C%2Fb%3E%E3%81%AF%E7%A4%BE%E5%86%85%E3%83%87%E3%83%BC%E3%82%BF%E3%81%8B%E3%82%89%E6%96%B0%E3%81%97%E3%81%84%E4%BE%A1%E5%80%A4%E3%82%92%E5%89%B5%E9%80%A0%E3%81%97%E3%81%BE%E3%81%99.%20DataRobot%E3%81%AF%E4%BC%81%E6%A5%AD%E3%81%AE%E8%AA%B2%E9%A1%8C%E8%A7%A3%E6%B1%BA%E3%81%AB%E7%89%B9%E5%8C%96%E3%80%82%E6%84%8F%E6%80%9D%E6%B1%BA%E5%AE%9A%E3%81%AE%E8%87%AA%E5%8B%95%E5%8C%96%E3%81%8B%E3%82%89%E9%9C%80%E8%A6%81%E4%BA%88%E6%B8%AC%E3%80%81%E8%A6%81%E5%9B%A0%E5%88%86%E6%9E%90%E3%81%BE%E3%81%A7%E3%81%93%E3%81%AA%E3%81%99AI%E3%83%84%E3%83%BC%E3%83%AB\",\"adx_name\":\"none\",\"is_good_v10\":1,\"organic_ranks\":[\"0\",1,2,3,5,6,7,8,9,10,11,12,13,14,15,18],\"q\":\"datarobot\",\"q_words\":1,\"q_words_fuzzy\":1,\"q_words_in_ad\":1,\"root_domain\":\"datarobot.com\",\"start\":\"0\",\"title\":\"%E3%83%87%E3%83%BC%E3%82%BF%E3%81%8B%E3%82%89%E6%96%B0%E3%81%97%E3%81%84%E4%BE%A1%E5%80%A4%E3%82%92%20%2D%20%E7%A4%BE%E5%86%85%E3%83%87%E3%83%BC%E3%82%BF%E3%81%8B%E3%82%89%E4%BE%A1%E5%80%A4%E5%89%B5%E5%87%BA\"},\"s\":\"bingv7aa\",\"t\":\"\\u30c7\\u30fc\\u30bf\\u304b\\u3089\\u65b0\\u3057\\u3044\\u4fa1\\u5024\\u3092 - \\u793e\\u5185\\u30c7\\u30fc\\u30bf\\u304b\\u3089\\u4fa1\\u5024\\u5275\\u51fa\",\"tid\":\"1,6,7,9[8],11[10],13[12]\",\"u\":\"https://duckduckgo.com/y.js?ad_domain=datarobot.com&ad_provider=bingv7aa&ad_type=txad&eddgt=4t4PwhnMmcY7atJfp41CAw%3D%3D&rut=4e8bc6239074926ffa0595d2f0838a03d95a9f20653372406dff4b73bb47a802&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De87ziwl19iJFSSt9AnpU7A2TVUCUyhd2uAWuXw7lbtLG3tobaae8IKDFy1RftUyaK7hmjFguIcBXLMF0__8U%2DYMqp1lRmGet90G40qSPB8wuC4MyZjuA8D06WqXRZsdl4uyGEuNXLmrp1n4swCoXfw3cJtq1Sl1iqwNQBdH4Ev%2DJQUf54N5TQ274OfwCR8PMamVlYCWg%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuZGF0YXJvYm90LmNvbSUyZmpwJTJmbHAlMmZhaS1mb3ItYnVzaW5lc3MlMmYlM2Z1dG1fbWVkaXVtJTNkc2VhcmNoJTI2dXRtX3NvdXJjZSUzZGJpbmclMjZ1dG1fY2FtcGFpZ24lM2RERU1PMjAyM0FsbFByb2R1Y3RzSlAwNjI2QlBTJTI2dXRtX3Rlcm0lM2RkYXRhcm9ib3QlMjZ1dG1fY29udGVudCUzZERSX2JyYW5kZWRfcnNhJTI2Y2FtcGFpZ25pZCUzZDUzMDcwODA5OSUyNmFkZ3JvdXBpZCUzZDEzNTAyMDI3NzQyMTc2OTglMjZhZGlkJTNkJTI2bXNjbGtpZCUzZDdlMDA4NWE0OTQ3MzE5ODdlNzJjZTAwNzZiNzAxMmFm%26rlid%3D7e0085a494731987e72ce0076b7012af&vqd=4-129716528454193272263152190229962811420&iurl=%7B1%7DIG%3DEFC9E56534984E60A01E1AD18F949C41%26CID%3D17342A5B7AD96D85041C3E5D7B5C6C74%26ID%3DDevEx%2C5060.1\"}], {\"page_load_url\":\"https://duckduckgo.com/y.js?ifu=%7B3%7Dappid%3D055AAD1BA669BEB8B048128DC89A107C678B527B%26rguid%3Dbff00ba70e8f4285a2ad22f803500ac4&iurl=%7B2%7DIG%3DEFC9E56534984E60A01E1AD18F949C41%26CID%3D17342A5B7AD96D85041C3E5D7B5C6C74%26Type%3DEvent.CPT%26DATA%3D0\",\"visibility_url\":\"https://duckduckgo.com/y.js?ivu=%7B4%7Dtype%3Dmv%26reqver%3D1.0%26rg%3Dbff00ba70e8f4285a2ad22f803500ac4\"});DDG.deep.signalSummary = \"\";DDG.inject('DDG.Data.languages.resultLanguages', {\"en\":[\"https://www.datarobot.com/\",\"https://www.datarobot.com/trial/\",\"https://www.datarobot.com/pricing/\",\"https://www.datarobot.com/platform/new/datarobot-9-0/\",\"https://www.linkedin.com/company/datarobot/\",\"https://docs.datarobot.com/\",\"https://docs.datarobot.com/en/docs/get-started/index.html\",\"https://www.datarobot.com/newsroom/press/datarobot-launches-pathfinder-a-comprehensive-library-of-100-ai-use-cases/\",\"https://www.datarobot.com/blog/introducing-the-datarobot-use-case-value-tracker/\",\"https://docs.datarobot.com/en/docs/data/index.html\",\"https://app.datarobot.com/\",\"https://pathfinder.datarobot.com/integrations\",\"https://docs.datarobot.com/en/docs/api/api-quickstart/index.html\",\"https://docs.datarobot.com/en/docs/data/connect-data/data-conn.html\",\"https://app.datarobot.com/sign-in\",\"https://learn.datarobot.com/\",\"https://www.carahsoft.com/datarobot\",\"https://www.afa.org/company/datarobot/\",\"https://www.datarobot.com/use-cases/\",\"https://www.nomuraholdings.com/top.html\",\"https://www.nomura.com/\"],\"es\":[\"https://vivevirtual.es/inteligencia-artificial/tutoriales-ia/como-integrar-datarobot-en-squarespace/\"]});DDG.deep.pageLayoutSummary = \"a1w22v1r1\";DDG.inject('DDG.Data.languages.adLanguages', {});if (DDG.pageLayout) DDG.pageLayout.load('d',[{\"a\":\"DataRobot is a leading platform for generative and predictive AI that lets you focus on tangible business outcomes, not infrastructure. Learn how DataRobot helps customers across industries and organizations scale AI with enterprise monitoring, governance, and open ecosystems.\",\"ae\":null,\"c\":\"https://www.datarobot.com/\",\"d\":\"www.datarobot.com\",\"da\":\"\",\"h\":0,\"i\":\"www.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot AI Platform | Deliver Value from AI\",\"u\":\"https://www.datarobot.com/\"},{\"a\":\"DataRobot is a single platform that streamlines your predictive and generative AI workflows. Start your free 30-day trial to experience how to fast-track preparing data, running experiments, and testing models, and to automate your AI processes with a single solution.\",\"ae\":null,\"c\":\"https://www.datarobot.com/trial/\",\"d\":\"www.datarobot.com/trial/\",\"da\":\"\",\"h\":0,\"i\":\"www.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot AI Platform Free Trial | DataRobot AI Platform\",\"u\":\"https://www.datarobot.com/trial/\"},{\"a\":\"DataRobot offers a range of pricing plans to suit your business needs and goals, from Essentials to Business Critical. Learn how to customize your solution, get support, and see the ROI and savings of DataRobot AI Platform.\",\"ae\":null,\"c\":\"https://www.datarobot.com/pricing/\",\"d\":\"www.datarobot.com/pricing/\",\"da\":\"\",\"h\":0,\"i\":\"www.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Pricing | DataRobot AI Platform\",\"u\":\"https://www.datarobot.com/pricing/\"},{\"a\":\"DataRobot AI Platform 9.0 is a complete and open AI lifecycle platform that leverages machine learning and supports Experimentation, Production, and Compliance. Learn about the new features, such as collaborative experimentation, workbench, data preparation, notebooks, drift management, and more, and how they integrate with Snowflake, GitHub, SAP, and Kubernetes.\",\"ae\":null,\"c\":\"https://www.datarobot.com/platform/new/datarobot-9-0/\",\"d\":\"www.datarobot.com/platform/new/datarobot-9-0/\",\"da\":\"translations\",\"h\":0,\"i\":\"www.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot AI Platform 9.0 Release | DataRobot AI Platform\",\"u\":\"https://www.datarobot.com/platform/new/datarobot-9-0/\"},{\"a\":\"DataRobot is the Value-Driven AI company, empowering organizations to accelerate AI from idea to impact. With over a decade at the forefront of AI innovation, we know what it takes to make a real ...\",\"ae\":null,\"b\":\"li\\tLinkedIn\\twww.linkedin.com\",\"c\":\"https://www.linkedin.com/company/datarobot/\",\"d\":\"www.linkedin.com/company/datarobot/\",\"da\":\"\",\"h\":0,\"i\":\"www.linkedin.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot | LinkedIn\",\"u\":\"https://www.linkedin.com/company/datarobot/\"},{\"a\":\"Learn how to use DataRobot, a platform for data science and machine learning, with UI and API docs, tutorials, and release information. Find out how to get started, manage the platform, and access additional resources for modeling success.\",\"ae\":null,\"c\":\"https://docs.datarobot.com/\",\"d\":\"docs.datarobot.com\",\"da\":\"\",\"h\":0,\"i\":\"docs.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot Product Documentation\",\"u\":\"https://docs.datarobot.com/\"},{\"a\":\"Learn how to use DataRobot's value-driven AI to analyze data, create and deploy models, and work with code-first notebooks. Explore the topics of workbench, data preparation, model building, model exploration, model deployment, and more.\",\"ae\":null,\"c\":\"https://docs.datarobot.com/en/docs/get-started/index.html\",\"d\":\"docs.datarobot.com/en/docs/get-started/index-html\",\"da\":\"\",\"e\":\"2023-08-21T00:00:00.0000000\",\"h\":0,\"i\":\"docs.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Get started: DataRobot docs\",\"u\":\"https://docs.datarobot.com/en/docs/get-started/index.html\"},{\"a\":\"DataRobot is the leader in enterprise AI, delivering trusted AI technology and enablement services to global enterprises competing in today's Intelligence Revolution. DataRobot's enterprise AI platform democratizes data science with end-to-end automation for building, deploying, and managing machine learning models. This platform maximizes ...\",\"ae\":null,\"c\":\"https://www.datarobot.com/newsroom/press/datarobot-launches-pathfinder-a-comprehensive-library-of-100-ai-use-cases/\",\"d\":\"www.datarobot.com/newsroom/press/datarobot-launches-pathfinder-a-comprehensive-library-of-100-ai-use-cases/\",\"da\":\"translations\",\"h\":0,\"i\":\"www.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot Launches Pathfinder: A Comprehensive Library of 100+ AI Use ...\",\"u\":\"https://www.datarobot.com/newsroom/press/datarobot-launches-pathfinder-a-comprehensive-library-of-100-ai-use-cases/\"},{\"a\":\"In Release 6.1, we are thrilled to introduce the DataRobot Use Case Value Tracker, designed to help you with the business operationalization of your AI.As a centralized hub to collaborate with team members on any machine learning initiative from start to finish, it provides a systematic way to manage, monitor, and track the return on investment generated from your AI efforts.\",\"ae\":null,\"c\":\"https://www.datarobot.com/blog/introducing-the-datarobot-use-case-value-tracker/\",\"d\":\"www.datarobot.com/blog/introducing-the-datarobot-use-case-value-tracker/\",\"da\":\"\",\"h\":0,\"i\":\"www.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Introducing the DataRobot Use Case Value Tracker\",\"u\":\"https://www.datarobot.com/blog/introducing-the-datarobot-use-case-value-tracker/\"},{\"a\":\"Learn how to import, transform, analyze, and manage data for machine learning projects using DataRobot tools and visualizations. Find out the data requirements, limitations, and best practices for different data types, sources, and scenarios.\",\"ae\":null,\"c\":\"https://docs.datarobot.com/en/docs/data/index.html\",\"d\":\"docs.datarobot.com/en/docs/data/index-html\",\"da\":\"\",\"e\":\"2023-06-15T00:00:00.0000000\",\"h\":0,\"i\":\"docs.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Data: DataRobot docs\",\"u\":\"https://docs.datarobot.com/en/docs/data/index.html\"},{\"a\":\"DataRobot\",\"ae\":null,\"c\":\"https://app.datarobot.com/\",\"d\":\"app.datarobot.com\",\"da\":\"\",\"h\":0,\"i\":\"app.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot\",\"u\":\"https://app.datarobot.com/\"},{\"a\":\"DataRobot does not share customer data, personal data, or sensitive use cases on Pathfinder. The AI applications and frameworks shared in Pathfinder are widespread and commonplace across their respective industries. This tool was developed to provide a framework for how you can solve AI use cases with problems that are within the context of ...\",\"ae\":null,\"c\":\"https://pathfinder.datarobot.com/integrations\",\"d\":\"pathfinder.datarobot.com/integrations\",\"da\":\"\",\"h\":0,\"i\":\"pathfinder.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Integrations | Explore 100+ AI Use Cases | DataRobot Pathfinder\",\"u\":\"https://pathfinder.datarobot.com/integrations\"},{\"a\":\"API quickstart. The DataRobot API provides a programmatic alternative to the web interface for creating and managing DataRobot projects. The API can be used via REST or with DataRobot's Python or R clients in Windows, UNIX, and OS X environments. This guide walks you through setting up your environment and then you can follow a sample problem ...\",\"ae\":null,\"c\":\"https://docs.datarobot.com/en/docs/api/api-quickstart/index.html\",\"d\":\"docs.datarobot.com/en/docs/api/api-quickstart/index-html\",\"da\":\"\",\"e\":\"2023-08-07T00:00:00.0000000\",\"h\":0,\"i\":\"docs.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"API Quickstart: DataRobot docs\",\"u\":\"https://docs.datarobot.com/en/docs/api/api-quickstart/index.html\"},{\"a\":\"From the Data Connections tab, select the data connection in the left-panel connections list. Click the Delete button in the upper right ( ). DataRobot prompts for confirmation. Click Delete to remove the data connection. If there are data sources dependent on the data connection, DataRobot returns a notification.\",\"ae\":null,\"c\":\"https://docs.datarobot.com/en/docs/data/connect-data/data-conn.html\",\"d\":\"docs.datarobot.com/en/docs/data/connect-data/data-conn.html\",\"da\":\"\",\"e\":\"2023-11-15T00:00:00.0000000\",\"h\":0,\"i\":\"docs.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Data connections: DataRobot docs - DataRobot AI Platform\",\"u\":\"https://docs.datarobot.com/en/docs/data/connect-data/data-conn.html\"},{\"a\":\"Sign in to DataRobot, the leading enterprise AI platform that democratizes data science and automates machine learning. DataRobot offers comprehensive solutions for MLOps, AI use cases, and AI cloud. Learn from DataRobot University and Algorithmia, and join the Intelligence Revolution.\",\"ae\":null,\"c\":\"https://app.datarobot.com/sign-in\",\"d\":\"app.datarobot.com/sign-in\",\"da\":\"\",\"h\":0,\"i\":\"app.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot\",\"u\":\"https://app.datarobot.com/sign-in\"},{\"a\":\"This article will help you better navigate DataRobot University. View Details. What's New in 9.0. Class. Learn about new features in the DataRobot 9.0 release. View Details. Integrating Snowflake with DataRobot. Class. Connect Snowflake to DataRobot to build models and make predictions.\",\"ae\":null,\"c\":\"https://learn.datarobot.com/\",\"d\":\"learn.datarobot.com\",\"da\":\"\",\"h\":0,\"i\":\"learn.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot University\",\"u\":\"https://learn.datarobot.com/\"},{\"a\":\"DataRobot is the leader in enterprise AI, delivering trusted AI technology and ROI enablement services to global enterprises. DataRobot's enterprise AI platform democratizes data science with end-to-end automation for building, deploying, and managing machine learning models. This platform maximizes value to the mission by delivering AI at ...\",\"ae\":null,\"c\":\"https://www.carahsoft.com/datarobot\",\"d\":\"www.carahsoft.com/datarobot\",\"da\":\"\",\"h\":0,\"i\":\"www.carahsoft.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot - Enterprise AI Cloud Platform | Carahsoft\",\"u\":\"https://www.carahsoft.com/datarobot\"},{\"a\":\"DataRobot is the leading Augmented Intelligence platform, delivering trusted AI technology and ROI enablement services to global enterprises competing in today's Intelligence Revolution. Its enterprise AI platform maximizes business value by delivering AI at scale and continuously optimizing performance over time.\",\"ae\":null,\"c\":\"https://www.afa.org/company/datarobot/\",\"d\":\"www.afa.org/company/datarobot/\",\"da\":\"\",\"e\":\"2024-01-09T00:00:00.0000000\",\"h\":0,\"i\":\"www.afa.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot - Air & Space Forces Association\",\"u\":\"https://www.afa.org/company/datarobot/\"},{\"a\":\"DataRobot is a platform for generative and predictive AI that helps you deploy and run AI solutions across various industries and outcomes. Learn how DataRobot customers use their platform to solve business problems, innovate, and drive value with AI.\",\"ae\":null,\"c\":\"https://www.datarobot.com/use-cases/\",\"d\":\"www.datarobot.com/use-cases/\",\"da\":\"\",\"h\":0,\"i\":\"www.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Machine Learning Use Cases | DataRobot AI Platform\",\"u\":\"https://www.datarobot.com/use-cases/\"},{\"a\":\"Para integrar DataRobot en tu sitio de Squarespace, iniciar con la configuraci\\u00f3n de una cuenta en la plataforma de DataRobot es esencial. El proceso es directo y f\\u00e1cil de seguir y asegurar\\u00e1 que dispongas de todas las herramientas anal\\u00edticas avanzadas para potenciar tu sitio web.. Crear Tu Cuenta de DataRobot. Visita el sitio web de DataRobot y localiza la opci\\u00f3n de registro.\",\"ae\":null,\"c\":\"https://vivevirtual.es/inteligencia-artificial/tutoriales-ia/como-integrar-datarobot-en-squarespace/\",\"d\":\"vivevirtual.es/inteligencia-artificial/tutoriales-ia/como-integrar-datarobot-en-squarespace/\",\"da\":\"\",\"e\":\"2024-01-11T00:00:00.0000000\",\"h\":0,\"i\":\"vivevirtual.es\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"C\\u00f3mo Integrar DataRobot En Squarespace \\ufe0f 2024 - \\u00a9Vive Virtual\",\"u\":\"https://vivevirtual.es/inteligencia-artificial/tutoriales-ia/como-integrar-datarobot-en-squarespace/\"},{\"a\":\"Nomura is a global financial services group with an integrated network spanning over 30 countries. By connecting markets East & West, Nomura services the needs of individuals, institutions, corporates and governments through its four business divisions: Retail, Asset Management, Wholesale (Global Markets and Investment Banking), and Merchant Banking.\",\"ae\":null,\"c\":\"https://www.nomuraholdings.com/top.html\",\"d\":\"www.nomuraholdings.com/top-html\",\"da\":\"\",\"h\":0,\"i\":\"www.nomuraholdings.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Home | NOMURA\",\"u\":\"https://www.nomuraholdings.com/top.html\"},{\"a\":\"Nomura is a global financial services group with an integrated network spanning over 30 countries. By connecting markets East & West, Nomura services the needs of individuals, institutions, corporates and governments through its four business divisions: Retail, Asset Management, Wholesale (Global Markets and Investment Banking), and Merchant ...\",\"ae\":null,\"c\":\"https://www.nomura.com/\",\"d\":\"www.nomura.com\",\"da\":\"\",\"h\":0,\"i\":\"www.nomura.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Home - NOMURA\",\"u\":\"https://www.nomura.com/\"},{\"n\":\"/d.js?q=datarobot&kl=wt-wt&l=wt-wt&p=&s=22&ex=-1&ct=US&sp=0&vqd=4-305438614248843041332096319235456803678\"}]);DDG.duckbar.load('images');DDG.duckbar.load('news');DDG.duckbar.load('videos', {\"ads\":[],\"query\":\"datarobot\",\"queryEncoded\":\"datarobot\",\"response_type\":\"places\",\"results\":[{\"content\":\"https://www.youtube.com/watch?v=vyi_0D-rJ1A\",\"description\":\"Demonstration of how the DataRobot AI Platform works covering both ML Experimentation and ML Production. Request a live, personalized demonstration at https://www.datarobot.com/request-a-demo. This demo shows the workflows in DataRobot data ingest and preparation, model development, insight extraction, model deployment, on-going monitoring and ...\",\"duration\":\"13:17\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/vyi_0D-rJ1A?autoplay=1\",\"image_token\":\"46e31c38546fa6c15f77d2eaca52158f719c396ddbc4e84f07f660a20f050f23\",\"images\":{\"large\":\"https://tse2.mm.bing.net/th?id=OVP.RjJmdFbBiUCndVAZOmjitwEsDh&pid=Api\",\"medium\":\"https://tse2.mm.bing.net/th?id=OVP.RjJmdFbBiUCndVAZOmjitwEsDh&pid=Api\",\"motion\":\"https://tse2.mm.bing.net/th?id=OM2.GlK_gSFTQdYbbg_1695678156&pid=Api\",\"small\":\"https://tse2.mm.bing.net/th?id=OVP.RjJmdFbBiUCndVAZOmjitwEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-06-13T21:14:17.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":5574},\"title\":\"DataRobot AI Platform Demo 2023 | End-to-end Workflow | How DataRobot Works\",\"uploader\":\"DataRobot\"},{\"content\":\"https://www.youtube.com/watch?v=cOV5dss8xo0\",\"description\":\"DataRobot offers a platform and a reusable framework for developing AI applications that have a generative and a predictive component to them. Watch the process of creating a fully functioning joint generative and predictive AI solution for a real-world business problem that delivers tangible value. Use this framework and process to deliver ...\",\"duration\":\"5:06\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/cOV5dss8xo0?autoplay=1\",\"image_token\":\"1f5b660aa742a554ce19c9e8b001eb5d5a89059afab5ab4ef499f34bbef8ae2d\",\"images\":{\"large\":\"https://tse1.mm.bing.net/th?id=OVP.z0cOprAHCz_P_Hpd5bMHWgEsDh&pid=Api\",\"medium\":\"https://tse1.mm.bing.net/th?id=OVP.z0cOprAHCz_P_Hpd5bMHWgEsDh&pid=Api\",\"motion\":\"https://tse1.mm.bing.net/th?id=OM1.pQ6zRQHggxqCCw_1696646714&pid=Api\",\"small\":\"https://tse1.mm.bing.net/th?id=OVP.z0cOprAHCz_P_Hpd5bMHWgEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-08-18T22:22:31.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":7622},\"title\":\"End-to-end Generative AI Applications with DataRobot | Develop, Deploy, Monitor and Maintain\",\"uploader\":\"DataRobot\"},{\"content\":\"https://www.youtube.com/watch?v=IMP0OZC6wPw\",\"description\":\"DataRobot is the leading end-to-end enterprise AI/ML platform that automates the process of building, training and deploying AI models at scale. Download the data and slides here: https://drive.google.com/drive/folders/1Zl1XO24zkbY7fHEh59Ux3NssNPXQD19t?usp=sharing In this video, we will learn how to build, train and deploy a machine learning ...\",\"duration\":\"21:22\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/IMP0OZC6wPw?autoplay=1\",\"image_token\":\"1a5261fccec96060141db7cec373afd6834e80eed802a6ed6c7ee3f67f6228ec\",\"images\":{\"large\":\"https://tse4.mm.bing.net/th?id=OVP.uJ2Vw0goB-5yx8dORs0CPgHgFo&pid=Api\",\"medium\":\"https://tse4.mm.bing.net/th?id=OVP.uJ2Vw0goB-5yx8dORs0CPgHgFo&pid=Api\",\"motion\":\"https://tse4.mm.bing.net/th?id=OM2.21pythA4GgwdwQ&pid=Api\",\"small\":\"https://tse4.mm.bing.net/th?id=OVP.uJ2Vw0goB-5yx8dORs0CPgHgFo&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2021-08-08T15:42:18.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":18696},\"title\":\"DataRobot AI For Absolute Beginners (Part 1) | Build, Train & Deploy an AI in 30 Minutes\",\"uploader\":\"Prof. Ryan Ahmed\"},{\"content\":\"https://www.youtube.com/watch?v=Grip5G1EUgY\",\"description\":\"Machine learning models developed with DataRobot can be deployed to Azure ML with complete service health, drift and accuracy monitoring. In this video Brian Bell demonstrates the Azure ML deployment capability. Users can create a DataRobot-managed AzureML prediction environment to deploy DataRobot Scoring Code in AzureML. With DataRobot ...\",\"duration\":\"3:22\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/Grip5G1EUgY?autoplay=1\",\"image_token\":\"8db48ef7fbe873efad3cfa34f62c3b13b4e77e556767a5d60a215edbb13b3e14\",\"images\":{\"large\":\"https://tse3.mm.bing.net/th?id=OVP.klj-M0G2PlRq58QgzSlIwAEsDh&pid=Api\",\"medium\":\"https://tse3.mm.bing.net/th?id=OVP.klj-M0G2PlRq58QgzSlIwAEsDh&pid=Api\",\"motion\":\"https://tse3.mm.bing.net/th?id=OM.-_UGz9qtcxRXvw_1704026949&pid=Api\",\"small\":\"https://tse3.mm.bing.net/th?id=OVP.klj-M0G2PlRq58QgzSlIwAEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-06-09T20:38:50.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":409},\"title\":\"Deploy DataRobot Models to AzureML | Demonstration of Azure ML Deployment Capability\",\"uploader\":\"DataRobot\"},{\"content\":\"https://www.youtube.com/watch?v=xZUaBvPQfXY\",\"description\":\"Get a high-level introduction to DataRobot's AI Production through a tour of several live Deployments. See the monitoring, alerting, lifecycle management and reporting capabilities in action for both predictive and generative AI via a series of examples. Learn more at: https://www.datarobot.com/platform/ https://docs.datarobot.com/en/docs/mlops ...\",\"duration\":\"7:08\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/xZUaBvPQfXY?autoplay=1\",\"image_token\":\"faecb8d567fd4445c52cd8e60c649952b393188212856a19ce331371c4b4b21b\",\"images\":{\"large\":\"https://tse3.mm.bing.net/th?id=OVP.V95ETtZnbbbKDAA2lzjFHgEsDh&pid=Api\",\"medium\":\"https://tse3.mm.bing.net/th?id=OVP.V95ETtZnbbbKDAA2lzjFHgEsDh&pid=Api\",\"motion\":\"https://tse3.mm.bing.net/th?id=OM1.oMwA1JZw78kwsw_1701730108&pid=Api\",\"small\":\"https://tse3.mm.bing.net/th?id=OVP.V95ETtZnbbbKDAA2lzjFHgEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-11-09T05:05:53.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":329},\"title\":\"Operate AI with DataRobot | Welcome to the AI Operations Console for New DataRobot Users\",\"uploader\":\"DataRobot\"},{\"content\":\"https://www.youtube.com/watch?v=Y00VSO6Uq60\",\"description\":\"Complete all phases of building, operating and governing a Predictive AI solution following the starter Flight Delays Use Case. This starter use case showcases essential DataRobot capabilities, but is not comprehensive. Learn about the complete capabilities of the DataRobot AI Platform at https://www.datarobot.com/platform. Watch the video as ...\",\"duration\":\"14:12\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/Y00VSO6Uq60?autoplay=1\",\"image_token\":\"e687f48c41420bafef42b0d7aec2ea6e0d55ae6c8c51adbb6d76c5d197637430\",\"images\":{\"large\":\"https://tse3.mm.bing.net/th?id=OVP.WMkyYYho3O9V85gpNsutVAEsDh&pid=Api\",\"medium\":\"https://tse3.mm.bing.net/th?id=OVP.WMkyYYho3O9V85gpNsutVAEsDh&pid=Api\",\"motion\":\"https://tse3.mm.bing.net/th?id=OM1.s30yJWAIBWugpw_1704039193&pid=Api\",\"small\":\"https://tse3.mm.bing.net/th?id=OVP.WMkyYYho3O9V85gpNsutVAEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-11-28T02:29:50.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":829},\"title\":\"Predictive AI: Build, Operate, and Govern with the DataRobot AI Platform | Flight Delays Use Case\",\"uploader\":\"DataRobot\"},{\"content\":\"https://www.youtube.com/watch?v=Jj8JovBRflA\",\"description\":\"Quick overview of DataRobot AI Accelerators including how to access them, what topics are covered, and how to get started using them with the data science notebook of your choice. Learn more https://community.datarobot.com/t5/ai-accelerators-library/tkb-p/ai-accelerators-library https://github.com/datarobot-community/ai-accelerators Transcript ...\",\"duration\":\"1:45\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/Jj8JovBRflA?autoplay=1\",\"image_token\":\"2749c14ed9db6fea61b6d86d8b55a56c24a8ac5900035d1ce6b326d0f9d807fe\",\"images\":{\"large\":\"https://tse1.mm.bing.net/th?id=OVP.yk9KOy_GBh3vO51YxcV8NQEsDh&pid=Api\",\"medium\":\"https://tse1.mm.bing.net/th?id=OVP.yk9KOy_GBh3vO51YxcV8NQEsDh&pid=Api\",\"motion\":\"https://tse1.mm.bing.net/th?id=OM2.w0xOKKBTuyvZdw_1699233002&pid=Api\",\"small\":\"https://tse1.mm.bing.net/th?id=OVP.yk9KOy_GBh3vO51YxcV8NQEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-09-28T17:01:32.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":1},\"title\":\"AI Accelerators Overview | Repeatable Workflows using the DataRobot API\",\"uploader\":\"DataRobot\"},{\"content\":\"https://www.youtube.com/watch?v=fm6nxsAo5J0\",\"description\":\"This demo showcases the end-to-end capabilities in the DataRobot Enterprise AI Platform using a house price listings dataset containing diverse feature types including numeric, categorical, raw text, images, and geospatial data. The demo takes us on a journey from raw data to value, and highlights DataRobot's governance, explainability, and ...\",\"duration\":\"4:56\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/fm6nxsAo5J0?autoplay=1\",\"image_token\":\"8aee60e8b0fc16a6e77cd5a9391dd3c6214fa2fab314b27a320055efcb29f75e\",\"images\":{\"large\":\"https://tse4.mm.bing.net/th?id=OVP.chafFyMzRYblao3a5sr79gEsDh&pid=Api\",\"medium\":\"https://tse4.mm.bing.net/th?id=OVP.chafFyMzRYblao3a5sr79gEsDh&pid=Api\",\"motion\":\"https://tse4.mm.bing.net/th?id=OM1.lHW5r59lMDBG5w_1684178452&pid=Api\",\"small\":\"https://tse4.mm.bing.net/th?id=OVP.chafFyMzRYblao3a5sr79gEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2020-08-11T16:00:10.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":17912},\"title\":\"DataRobot AI Vision Demo\",\"uploader\":\"DataRobot\"},{\"content\":\"https://www.youtube.com/watch?v=XhipOG-S1q8\",\"description\":\"This end-to-end demo shows the tight integrations between the DataRobot AI Platform and AWS services. By natively connecting to data in Amazon S3, you can build, test, and evaluate models in DataRobot, and then deploy them in Amazon SageMaker. These models can be monitored for drift and other relevant parameters via DataRobot MLOps. In this ...\",\"duration\":\"13:27\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/XhipOG-S1q8?autoplay=1\",\"image_token\":\"d2a00656028a69312df6f3ae2ff1e484fd23dc4bceff58bc5157813853d6db90\",\"images\":{\"large\":\"https://tse2.mm.bing.net/th?id=OVP.BC-1oDSg7Hw6OofUPNCBFwEsDh&pid=Api\",\"medium\":\"https://tse2.mm.bing.net/th?id=OVP.BC-1oDSg7Hw6OofUPNCBFwEsDh&pid=Api\",\"motion\":\"https://tse2.mm.bing.net/th?id=OM1.gmEOiAfzgn4Y-A_1684167397&pid=Api\",\"small\":\"https://tse2.mm.bing.net/th?id=OVP.BC-1oDSg7Hw6OofUPNCBFwEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-02-08T10:14:17.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":2025},\"title\":\"DataRobot and AWS: Rapidly Prototype and Deploy AI Models | Demo Tutorial\",\"uploader\":\"DataRobot\"},{\"content\":\"https://www.youtube.com/watch?v=RrbJLm6atwc\",\"description\":\"Please visit the 2023 DataRobot Product Demo at https://www.youtube.com/watch?v=vyi_0D-rJ1A This video shows the DataRobot AI Platform in action as of 2018. DataRobot is the category creator of AutoML and MLOps with a rich history of innovation as detailed here - https://www.datarobot.com/innovation. Our Platform is constantly evolving with new ...\",\"duration\":\"1:32\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/RrbJLm6atwc?autoplay=1\",\"image_token\":\"f08d101a4ecaa1ecbc1702f740cddd0e35c504901e82dcfe0276733759319a8e\",\"images\":{\"large\":\"https://tse1.mm.bing.net/th?id=OVP.NzKh9KZZ5OuUbK1j19RfbwHgFo&pid=Api\",\"medium\":\"https://tse1.mm.bing.net/th?id=OVP.NzKh9KZZ5OuUbK1j19RfbwHgFo&pid=Api\",\"motion\":\"https://tse1.mm.bing.net/th?id=OM1.AWDbr86dOILmXQ_1645855537&pid=Api\",\"small\":\"https://tse1.mm.bing.net/th?id=OVP.NzKh9KZZ5OuUbK1j19RfbwHgFo&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2018-04-16T14:01:36.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":49270},\"title\":\"DataRobot AI Platform [2018 Version - Update Available]\",\"uploader\":\"DataRobot\"}],\"vqd\":{\"datarobot\":\"4-305438614248843041332096319235456803678\"}});DDG.duckbar.loadModule('related_searches', {\"ads\":[],\"query\":\"datarobot\",\"queryEncoded\":\"datarobot\",\"response_type\":\"places\",\"results\":[{\"display_text\":\"datarobot login\",\"text\":\"datarobot login\",\"web_search_url\":\"?q=datarobot%20login\"},{\"display_text\":\"datarobot download\",\"text\":\"datarobot download\",\"web_search_url\":\"?q=datarobot%20download\"},{\"display_text\":\"datarobot products\",\"text\":\"datarobot products\",\"web_search_url\":\"?q=datarobot%20products\"},{\"display_text\":\"datarobot company\",\"text\":\"datarobot company\",\"web_search_url\":\"?q=datarobot%20company\"},{\"display_text\":\"datarobot wikipedia\",\"text\":\"datarobot wikipedia\",\"web_search_url\":\"?q=datarobot%20wikipedia\"},{\"display_text\":\"datarobot artificial intelligence\",\"text\":\"datarobot artificial intelligence\",\"web_search_url\":\"?q=datarobot%20artificial%20intelligence\"},{\"display_text\":\"datarobot for your daily life\",\"text\":\"datarobot for your daily life\",\"web_search_url\":\"?q=datarobot%20for%20your%20daily%20life\"},{\"display_text\":\"data robot tool\",\"text\":\"data robot tool\",\"web_search_url\":\"?q=data%20robot%20tool\"}],\"vqd\":{\"datarobot\":\"4-305438614248843041332096319235456803678\"}});if (DDG.pageLayout) DDG.pageLayout.initialize({\"mainline\":{\"items\":[[\"ad\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"videos\"],[\"related_searches\"]]}}, { start: 0 });DDG.deep.emit(\"load:completed\");", + "curl-cffi-GET-https://links.duckduckgo.com/d.js-{\"params\": {\"bing_market\": \"wt-WT\", \"df\": null, \"ex\": \"-1\", \"kl\": \"wt-wt\", \"l\": \"wt-wt\", \"q\": \"dataiku\", \"s\": \"0\", \"sp\": \"0\", \"vqd\": \"4-337400409169293617055811118659485228425\"}}": "DDG.search.altIsNavigational = 1;DDG.search.isNavigational = 1;if (DDG.deep && DDG.deep.setUpstream) DDG.deep.setUpstream(\"bingv7aa\");DDG.deep.bn={'ivc':1};if (DDG.pageLayout) DDG.pageLayout.load('a',[], {\"page_load_url\":\"https://duckduckgo.com/y.js?iurl=%7B2%7DIG%3D4B3C73388F5840C299E628F54B339615%26CID%3D15DC198D7A096819249D0D8B7B3C6947%26Type%3DEvent.CPT%26DATA%3D0\"});DDG.duckbar.future_signal_tab({signal:'medium',from:'deep_answer'});DDG.duckbar.add({\"data\":{\"Abstract\":\"Dataiku is an artificial intelligence and machine learning company which was founded in 2013. In December 2019, Dataiku announced that CapitalG\\u2014the late-stage growth venture capital fund financed by Alphabet Inc.\\u2014joined Dataiku as an investor and that it had achieved unicorn status. As of 2021, Dataiku is valued at $4.6 billion. Dataiku currently employs more than 1,000 people worldwide between offices in New York, Denver, Washington DC, Los Angeles, Paris, London, Munich, Frankfurt, Sydney, Singapore, Tokyo, and Dubai.\",\"AbstractSource\":\"Wikipedia\",\"AbstractText\":\"Dataiku is an artificial intelligence and machine learning company which was founded in 2013. In December 2019, Dataiku announced that CapitalG\\u2014the late-stage growth venture capital fund financed by Alphabet Inc.\\u2014joined Dataiku as an investor and that it had achieved unicorn status. As of 2021, Dataiku is valued at $4.6 billion. Dataiku currently employs more than 1,000 people worldwide between offices in New York, Denver, Washington DC, Los Angeles, Paris, London, Munich, Frankfurt, Sydney, Singapore, Tokyo, and Dubai.\",\"AbstractURL\":\"https://en.wikipedia.org/wiki/Dataiku\",\"Answer\":\"\",\"AnswerType\":\"\",\"Definition\":\"\",\"DefinitionSource\":\"\",\"DefinitionURL\":\"\",\"Entity\":\"company\",\"Heading\":\"Dataiku\",\"Image\":\"/i/b50c1c3f.png\",\"ImageHeight\":270,\"ImageIsLogo\":1,\"ImageWidth\":483,\"Infobox\":{\"content\":[{\"data_type\":\"string\",\"label\":\"Type\",\"sort_order\":\"1000\",\"value\":\"Private\",\"wiki_order\":0},{\"data_type\":\"string\",\"label\":\"Industry\",\"sort_order\":\"1001\",\"value\":\"Computer software\",\"wiki_order\":1},{\"data_type\":\"string\",\"label\":\"Founded\",\"sort_order\":\"1\",\"value\":\"February 14, 2013 in Paris, France\",\"wiki_order\":2},{\"data_type\":\"string\",\"label\":\"Founders\",\"sort_order\":\"1002\",\"value\":\"Florian Douetteau, Cl\\u00e9ment Stenac, Marc Batty, Thomas Cabrol\",\"wiki_order\":3},{\"data_type\":\"string\",\"label\":\"Key people\",\"sort_order\":\"2\",\"value\":\"Florian Douetteau (CEO)\",\"wiki_order\":4},{\"data_type\":\"string\",\"label\":\"Products\",\"sort_order\":\"1003\",\"value\":\"Dataiku Data Science Studio\",\"wiki_order\":5},{\"data_type\":\"string\",\"label\":\"Revenue\",\"sort_order\":\"3\",\"value\":\"US$ 150 million (2021)\",\"wiki_order\":6},{\"data_type\":\"string\",\"label\":\"Number of employees\",\"sort_order\":\"1004\",\"value\":\"1,000+ (2022)\",\"wiki_order\":7},{\"data_type\":\"string\",\"label\":\"Website\",\"sort_order\":\"1005\",\"value\":\"[dataiku.com]\",\"wiki_order\":8},{\"data_type\":\"twitter_profile\",\"label\":\"Twitter profile\",\"value\":\"dataiku\",\"wiki_order\":\"102\"},{\"data_type\":\"instance\",\"label\":\"Instance of\",\"value\":{\"entity-type\":\"item\",\"id\":\"Q4830453\",\"numeric-id\":4830453},\"wiki_order\":\"207\"},{\"data_type\":\"official_website\",\"label\":\"Official Website\",\"value\":\"http://www.dataiku.com/\",\"wiki_order\":\"208\"}],\"meta\":[{\"data_type\":\"string\",\"label\":\"article_title\",\"value\":\"Dataiku\"},{\"data_type\":\"string\",\"label\":\"template_name\",\"value\":\"infobox company\"},{\"data_type\":\"string\",\"label\":\"formatting_rules\",\"value\":\"company\"}]},\"OfficialDomain\":\"dataiku.com\",\"OfficialWebsite\":\"https://dataiku.com\",\"Redirect\":\"\",\"RelatedTopics\":[{\"FirstURL\":\"https://duckduckgo.com/c/Big_data_companies\",\"Icon\":{\"Height\":\"\",\"URL\":\"\",\"Width\":\"\"},\"Result\":\"Big data companies\",\"Text\":\"Big data companies\"},{\"FirstURL\":\"https://duckduckgo.com/c/Data_analysis_software\",\"Icon\":{\"Height\":\"\",\"URL\":\"\",\"Width\":\"\"},\"Result\":\"Data analysis software\",\"Text\":\"Data analysis software\"},{\"FirstURL\":\"https://duckduckgo.com/c/Privately_held_companies_based_in_New_York_City\",\"Icon\":{\"Height\":\"\",\"URL\":\"\",\"Width\":\"\"},\"Result\":\"Privately held companies based in New York City\",\"Text\":\"Privately held companies based in New York City\"},{\"FirstURL\":\"https://duckduckgo.com/c/Proprietary_software\",\"Icon\":{\"Height\":\"\",\"URL\":\"\",\"Width\":\"\"},\"Result\":\"Proprietary software\",\"Text\":\"Proprietary software\"}],\"Results\":[{\"FirstURL\":\"https://dataiku.com\",\"Icon\":{\"Height\":16,\"URL\":\"/i/dataiku.com.ico\",\"Width\":16},\"Result\":\"Official site\",\"Text\":\"Official site\"},{\"FirstURL\":\"http://www.dataiku.com/\",\"Icon\":{\"Height\":16,\"URL\":\"/i/dataiku.com.ico\",\"Width\":16},\"Result\":\"Official site - Dataiku\",\"Text\":\"Official site - Dataiku\"}],\"Type\":\"A\",\"meta\":{\"attribution\":null,\"blockgroup\":null,\"created_date\":null,\"description\":\"Wikipedia\",\"designer\":null,\"dev_date\":null,\"dev_milestone\":\"live\",\"developer\":[{\"name\":\"DDG Team\",\"type\":\"ddg\",\"url\":\"http://www.duckduckhack.com\"}],\"example_query\":\"nikola tesla\",\"id\":\"wikipedia_fathead\",\"is_stackexchange\":null,\"js_callback_name\":\"wikipedia\",\"live_date\":null,\"maintainer\":{\"github\":\"duckduckgo\"},\"name\":\"Wikipedia\",\"perl_module\":\"DDG::Fathead::Wikipedia\",\"producer\":null,\"production_state\":\"online\",\"repo\":\"fathead\",\"signal_from\":\"wikipedia_fathead\",\"src_domain\":\"en.wikipedia.org\",\"src_id\":1,\"src_name\":\"Wikipedia\",\"src_options\":{\"directory\":\"\",\"is_fanon\":0,\"is_mediawiki\":1,\"is_wikipedia\":1,\"language\":\"en\",\"min_abstract_length\":\"20\",\"skip_abstract\":0,\"skip_abstract_paren\":0,\"skip_end\":\"0\",\"skip_icon\":0,\"skip_image_name\":0,\"skip_qr\":\"\",\"source_skip\":\"\",\"src_info\":\"\"},\"src_url\":null,\"status\":\"live\",\"tab\":\"About\",\"topic\":[\"productivity\"],\"unsafe\":0}},\"duckbar_topic\":\"About\",\"from\":\"deep_answer\",\"meta\":{\"attribution\":null,\"blockgroup\":null,\"created_date\":null,\"description\":\"Wikipedia\",\"designer\":null,\"dev_date\":null,\"dev_milestone\":\"live\",\"developer\":[{\"name\":\"DDG Team\",\"type\":\"ddg\",\"url\":\"http://www.duckduckhack.com\"}],\"example_query\":\"nikola tesla\",\"id\":\"wikipedia_fathead\",\"is_stackexchange\":null,\"js_callback_name\":\"wikipedia\",\"live_date\":null,\"maintainer\":{\"github\":\"duckduckgo\"},\"name\":\"Wikipedia\",\"perl_module\":\"DDG::Fathead::Wikipedia\",\"producer\":null,\"production_state\":\"online\",\"repo\":\"fathead\",\"signal_from\":\"wikipedia_fathead\",\"src_domain\":\"en.wikipedia.org\",\"src_id\":1,\"src_name\":\"Wikipedia\",\"src_options\":{\"directory\":\"\",\"is_fanon\":0,\"is_mediawiki\":1,\"is_wikipedia\":1,\"language\":\"en\",\"min_abstract_length\":\"20\",\"skip_abstract\":0,\"skip_abstract_paren\":0,\"skip_end\":\"0\",\"skip_icon\":0,\"skip_image_name\":0,\"skip_qr\":\"\",\"source_skip\":\"\",\"src_info\":\"\"},\"src_url\":null,\"status\":\"live\",\"tab\":\"About\",\"topic\":[\"productivity\"],\"unsafe\":0},\"model\":\"FatheadArticle\",\"official_site\":\"https://dataiku.com\",\"pixel_id\":\"wikipedia_fathead_deep\",\"signal\":\"medium\",\"templates\":{\"detail\":\"info_detail\"}});DDG.deep.signalSummary = \"about:m\";DDG.inject('DDG.Data.languages.resultLanguages', {\"en\":[\"https://www.dataiku.com/\",\"https://en.wikipedia.org/wiki/Dataiku\",\"https://www.dataiku.com/product/get-started/\",\"https://academy.dataiku.com/basics-101\",\"https://www.dataiku.com/product/dataiku-as-a-managed-service/\",\"https://knowledge.dataiku.com/latest/getting-started/about-dataiku/index.html\",\"https://discover.dataiku.com/data-quality/\",\"https://discover.dataiku.com/dataiku-12/\",\"https://www.linkedin.com/company/dataiku\",\"https://pages.dataiku.com/interactive-data-sheet\",\"https://developer.dataiku.com/latest/tutorials/index.html\",\"https://discover.dataiku.com/dataiku-for-generative-ai/\",\"https://blog.dataiku.com/dataiku-12\",\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/dataiku/product/dataiku\",\"https://blog.dataiku.com/why-users-love-dataiku\",\"https://knowledge.dataiku.com/latest/getting-started/about-dataiku/concept-value-proposition.html\",\"https://twitter.com/dataiku\",\"https://in.linkedin.com/company/persistent-systems\",\"https://www.freelancer.com/projects/database-administration/snowflake-dbt-dataiku-support\"],\"ja\":[\"https://jp.linkedin.com/in/tadashi-mishima-32638445\",\"https://www.intellilink.co.jp/topics/seminar_event/2024/dataiku-fy2023-2h.aspx\"]});DDG.deep.pageLayoutSummary = \"w5v1w16r1,e1\";DDG.inject('DDG.Data.languages.adLanguages', {});if (DDG.pageLayout) DDG.pageLayout.load('d',[{\"a\":\"Dataiku is a platform that lets you build, deploy, and manage data and AI projects all in one place. Whether you need to prepare data, train models, deploy models, or monitor and govern models, Dataiku has the tools and features to help you achieve your goals.\",\"ae\":null,\"c\":\"https://www.dataiku.com/\",\"d\":\"www.dataiku.com\",\"da\":\"\",\"h\":0,\"i\":\"www.dataiku.com\",\"k\":null,\"l\":[{\"snippet\":\"Dataiku saves time with quick visual analysis of columns, including the distribution of values, top values, outliers, invalids, and overall good statistics. For categorical data, the visual analysis includes the distribution by value. Martin Leijen . Business Data Architect at Action ...\",\"targetUrl\":\"https://www.dataiku.com/product/\",\"text\":\"Product\"},{\"snippet\":\"At Dataiku, we believe that diversity of people and thought is inherent to creating not only a top-quality product but an environment of inclusivity and belonging in which everyone can bring their full selves to work. It is the responsibility of every Dataiker to build a community of tolerance and open-mindedness.\",\"targetUrl\":\"https://www.dataiku.com/careers/\",\"text\":\"Careers\"},{\"snippet\":\"Dataiku lets you access and process data using the coding language of your choice, and lets you use code notebooks to prototype your recipes. Check out the use cases! Learn more Dataiku APIs . APIs in Dataiku allow coders to programmatically interact with various Dataiku objects and with the instance itself to accomplish a wide variety of tasks\",\"targetUrl\":\"https://www.dataiku.com/learn/\",\"text\":\"Learn\"},{\"snippet\":\"Thrive SPC Uses Dataiku, Snowflake, and Snow Fox Data to Improve Clinical Home Care. By moving to Dataiku and working with Dataiku partners, Snowflake and Snow Fox Data, Thrive Skilled Pediatric Care (Thrive SPC) has been able to advance from complicated spreadsheets to a central platform that provides clear insights and metrics to fuel their data-driven healthcare solutions.\",\"targetUrl\":\"https://www.dataiku.com/stories/\",\"text\":\"Stories\"},{\"snippet\":\"Dataiku was founded on the principle that in order to succeed in the world's rapidly evolving ecosystem, companies \\u2014 no matter what their industry or size \\u2014 must elevate their people to continuously innovate. Since 2013, Dataiku has been the leader in democratizing data and empowering organization-wide collaboration. We've been a part ...\",\"targetUrl\":\"https://www.dataiku.com/company/\",\"text\":\"Company\"},{\"snippet\":\"Join the Dataiku Partner Ecosystem. Become a part of the growing Dataiku service partner ecosystem or get in touch to talk integrations and technical partnerships.\",\"targetUrl\":\"https://www.dataiku.com/partners/\",\"text\":\"Partners\"}],\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku | Everyday AI, Extraordinary People\",\"u\":\"https://www.dataiku.com/\"},{\"a\":\"Dataiku is an artificial intelligence (AI) and machine learning company which was founded in 2013. In December 2019, Dataiku announced that CapitalG\\u2014the late-stage growth venture capital fund financed by Alphabet Inc.\\u2014joined Dataiku as an investor and that it had achieved unicorn status.\",\"ae\":null,\"b\":\"w\\tWikipedia\\ten.wikipedia.org\",\"c\":\"https://en.wikipedia.org/wiki/Dataiku\",\"d\":\"en.wikipedia.org/wiki/Dataiku\",\"da\":\"en_wikipedia_queries,nlp_fathead,nlp_wiki\",\"h\":0,\"i\":\"en.wikipedia.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku - Wikipedia\",\"u\":\"https://en.wikipedia.org/wiki/Dataiku\"},{\"a\":\"Dataiku is a fully managed online and installed platform that helps you build and deploy AI projects with data preparation, pipelines, AutoML, and automation. Start a 14-day free trial or download the free edition for up to 3 users and explore the features and benefits of Dataiku.\",\"ae\":null,\"c\":\"https://www.dataiku.com/product/get-started/\",\"d\":\"www.dataiku.com/product/get-started/\",\"da\":\"\",\"e\":\"2022-03-01T00:00:00.0000000\",\"h\":0,\"i\":\"www.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Get Started With Dataiku - Start an Online Trial or Download for Free\",\"u\":\"https://www.dataiku.com/product/get-started/\"},{\"a\":\"Dataiku Academy is a free online course that introduces the basics of Dataiku, a data analysis platform that allows you to create and explore data projects. You will learn how to create a project, a dataset, a connection, and a chart using Dataiku's interface and tools.\",\"ae\":null,\"c\":\"https://academy.dataiku.com/basics-101\",\"d\":\"academy.dataiku.com/basics-101\",\"da\":\"translations\",\"h\":0,\"i\":\"academy.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Basics 101 - Dataiku\",\"u\":\"https://academy.dataiku.com/basics-101\"},{\"a\":\"Dataiku Cloud is a fully managed service that lets you create AI and analytics insights from your data using modern cloud platforms and easy-to-use tools. Learn how to use Dataiku Cloud with built-in data connectors, AutoML, and online learning and support.\",\"ae\":null,\"c\":\"https://www.dataiku.com/product/dataiku-as-a-managed-service/\",\"d\":\"www.dataiku.com/product/dataiku-as-a-managed-service/\",\"da\":\"\",\"h\":0,\"i\":\"www.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku Cloud | Dataiku\",\"u\":\"https://www.dataiku.com/product/dataiku-as-a-managed-service/\"},{\"a\":\"About Dataiku# The following resources walk you through the main principles of the platform and how those core concepts can be applied to build an end-to-end solution. Concepts# Concept | The value proposition of Dataiku; Concept | Dataiku project walkthrough; Next.\",\"ae\":null,\"c\":\"https://knowledge.dataiku.com/latest/getting-started/about-dataiku/index.html\",\"d\":\"knowledge.dataiku.com/latest/getting-started/about-dataiku/index.html\",\"da\":\"\",\"h\":0,\"i\":\"knowledge.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"About Dataiku - Dataiku Knowledge Base\",\"u\":\"https://knowledge.dataiku.com/latest/getting-started/about-dataiku/index.html\"},{\"a\":\"Dataiku is a universal data, analytics, and AI platform that helps you detect and improve data quality challenges. Learn how to use Dataiku's discovery capabilities, data catalog, and data quality rules to deliver trusted data at speed across your organization.\",\"ae\":null,\"c\":\"https://discover.dataiku.com/data-quality/\",\"d\":\"discover.dataiku.com/data-quality/\",\"da\":\"\",\"h\":0,\"i\":\"discover.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Data Quality - Discover Dataiku\",\"u\":\"https://discover.dataiku.com/data-quality/\"},{\"a\":\"Dataiku 12 is a platform that connects data experts with generative AI models like OpenAI GPT and ChatGPT to create data projects. Learn how to use Dataiku 12 to do more with generative AI and data using a visual interface and natural language prompts.\",\"ae\":null,\"c\":\"https://discover.dataiku.com/dataiku-12/\",\"d\":\"discover.dataiku.com/dataiku-12/\",\"da\":\"\",\"h\":0,\"i\":\"discover.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku 12 - Discover Dataiku\",\"u\":\"https://discover.dataiku.com/dataiku-12/\"},{\"a\":\"Dataiku is the platform for Everyday AI, systemizing the use of data for exceptional business results. Organizations that use Dataiku elevate their people (whether technical and working in code or ...\",\"ae\":null,\"b\":\"li\\tLinkedIn\\twww.linkedin.com\",\"c\":\"https://www.linkedin.com/company/dataiku\",\"d\":\"www.linkedin.com/company/dataiku\",\"da\":\"\",\"h\":0,\"i\":\"www.linkedin.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku | LinkedIn\",\"u\":\"https://www.linkedin.com/company/dataiku\"},{\"a\":\"Dataiku is a platform that enables data experts and domain experts to work together to build AI into their daily operations. Learn about the key benefits and features of Dataiku, such as a visual lineage, governance, and collaboration, with this interactive data sheet.\",\"ae\":null,\"c\":\"https://pages.dataiku.com/interactive-data-sheet\",\"d\":\"pages.dataiku.com/interactive-data-sheet\",\"da\":\"\",\"h\":0,\"i\":\"pages.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku Interactive Data Sheet\",\"u\":\"https://pages.dataiku.com/interactive-data-sheet\"},{\"a\":\"Plugin development. Extend the native capabilities of Dataiku with custom-built components. This section contains tutorials which will help you learn how to use and combine programmatic features of Dataiku through step-by-step exercises. Developer tools Tooling and guidance to write code ...\",\"ae\":null,\"c\":\"https://developer.dataiku.com/latest/tutorials/index.html\",\"d\":\"developer.dataiku.com/latest/tutorials/index-html\",\"da\":\"\",\"h\":0,\"i\":\"developer.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Tutorials - Dataiku Developer Guide\",\"u\":\"https://developer.dataiku.com/latest/tutorials/index.html\"},{\"a\":\"With Dataiku's Prompt Studios. Prompt engineering is the key to developing robust interactions and reliable outputs from Generative AI services. With Dataiku's Prompt Studios, data scientists and engineers can design and operationalize high-performing, reusable prompts, complete with cost estimates across different LLM providers and models.\",\"ae\":null,\"c\":\"https://discover.dataiku.com/dataiku-for-generative-ai/\",\"d\":\"discover.dataiku.com/dataiku-for-generative-ai/\",\"da\":\"\",\"h\":0,\"i\":\"discover.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku for Generative AI - Discover Dataiku\",\"u\":\"https://discover.dataiku.com/dataiku-for-generative-ai/\"},{\"a\":\"Dataiku 12 includes new capabilities for data and IT teams to streamline MLOps and governance processes to deploy models faster, better manage production models, and improve model governance. A core principle of AI safety is keeping a human in the loop. Models don't always have the best or safest answer.\",\"ae\":null,\"c\":\"https://blog.dataiku.com/dataiku-12\",\"d\":\"blog.dataiku.com/dataiku-12\",\"da\":\"\",\"e\":\"2023-05-31T00:00:00.0000000\",\"h\":0,\"i\":\"blog.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Keep AI Under Control With Dataiku 12\",\"u\":\"https://blog.dataiku.com/dataiku-12\"},{\"a\":\"Dataiku is a platform for data science and machine learning, enabling data experts and domain experts to work together to build data into their daily operations. Read customer reviews, ratings, features, and alternatives of Dataiku from Gartner Peer Insights.\",\"ae\":null,\"c\":\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/dataiku/product/dataiku\",\"d\":\"www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/dataiku/product/dataiku\",\"da\":\"\",\"h\":0,\"i\":\"www.gartner.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku Reviews, Ratings & Features 2024 | Gartner Peer Insights\",\"u\":\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/dataiku/product/dataiku\"},{\"a\":\"" Dataiku has greatly impacted my day-to-day work by streamlining and automating many of the data processing and analysis tasks that were previously time consuming and labor intensive. Overall, Dataiku has significantly increased the efficiency and effectiveness of an organization's data-driven decision-making processes.\",\"ae\":null,\"c\":\"https://blog.dataiku.com/why-users-love-dataiku\",\"d\":\"blog.dataiku.com/why-users-love-dataiku\",\"da\":\"\",\"e\":\"2023-02-22T00:00:00.0000000\",\"h\":0,\"i\":\"blog.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Why Users Love Dataiku: Stories From the Community\",\"u\":\"https://blog.dataiku.com/why-users-love-dataiku\"},{\"a\":\"Dataiku is the platform for Everyday AI, systemizing the use of data for exceptional business results. In the same way that computers or the internet have become embedded in the everyday activities of organizations, AI can help organizations transform processes and help make better and wiser decisions.\",\"ae\":null,\"c\":\"https://knowledge.dataiku.com/latest/getting-started/about-dataiku/concept-value-proposition.html\",\"d\":\"knowledge.dataiku.com/latest/getting-started/about-dataiku/concept-value-proposition.html\",\"da\":\"\",\"h\":0,\"i\":\"knowledge.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Concept | The value proposition of Dataiku\",\"u\":\"https://knowledge.dataiku.com/latest/getting-started/about-dataiku/concept-value-proposition.html\"},{\"a\":\"Dataiku. @dataiku. Dataiku is the only AI platform that connects data and doers, enabling anyone across organizations to transform business data into real business impact. Software Company New York, NY dataiku.com Joined September 2012. 690 Following.\",\"ae\":null,\"b\":\"@\\tTwitter User Page\\ttwitter.com\",\"c\":\"https://twitter.com/dataiku\",\"d\":\"twitter.com/dataiku\",\"da\":\"\",\"h\":0,\"i\":\"twitter.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku (@dataiku) | Twitter\",\"u\":\"https://twitter.com/dataiku\"},{\"a\":\"Persistent Systems. 1,142,095 followers. 2mo. We're delighted to share that we continued our growth momentum as we reported $291.71M in revenue in Q2 FY24, delivering 14.1% year-over-year revenue growth. Our focus on client-centricity has enabled us to register the highest-ever TCV with more than $475M in the current quarter.\",\"ae\":null,\"b\":\"li\\tLinkedIn\\twww.linkedin.com\",\"c\":\"https://in.linkedin.com/company/persistent-systems\",\"d\":\"in.linkedin.com/company/persistent-systems\",\"da\":\"\",\"h\":0,\"i\":\"in.linkedin.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Persistent Systems | LinkedIn\",\"u\":\"https://in.linkedin.com/company/persistent-systems\"},{\"a\":\"Dataiku\\u306f\\u3001\\u30ad\\u30fc\\u30a6\\u30a9\\u30fc\\u30ab\\u30fc\\u69d8\\u3068\\u30b3\\u30f3\\u30b5\\u30eb\\u30c6\\u30a3\\u30f3\\u30b0\\u30d1\\u30fc\\u30c8\\u30ca\\u30fc\\u5951\\u7d04\\u3092\\u7de0\\u7d50\\u3057\\u307e\\u3057\\u305f\\u3002\\u30ad\\u30fc\\u30a6\\u30a9\\u30fc\\u30ab\\u30fc\\u69d8\\u306fDataiku\\u306e\\u6d3b\\u7528\\u3092\\u901a\\u3058\\u3066\\u304a\\u5ba2\\u69d8\\u306e\\u30c7\\u30fc\\u30bf\\u6d3b\\u7528\\u652f\\u63f4\\u3084\\u5185\\u88fd\\u5316\\u652f\\u63f4\\u3092\\u884c\\u3044\\u3001\\u30b5\\u30a4\\u30ed\\u5316\\u3055\\u308c\\u305f\\u30b7\\u30b9\\u30c6\\u30e0\\u304b\\u3089\\u306e\\u8131\\u5374\\u3084\\u30c7\\u30b8\\u30bf\\u30eb\\u4eba\\u6750\\u306e\\u78ba\\u4fdd\\u3068\\u3044\\u3063\\u305f\\u8ab2\\u984c\\u306b\\u5bfe\\u5fdc\\u3057\\u307e\\u3059\\u3002\",\"ae\":null,\"b\":\"li\\tLinkedIn\\twww.linkedin.com\",\"c\":\"https://jp.linkedin.com/in/tadashi-mishima-32638445\",\"d\":\"jp.linkedin.com/in/tadashi-mishima-32638445\",\"da\":\"\",\"h\":0,\"i\":\"jp.linkedin.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Tadashi Mishima - Growth Sales Specialist - UiPath Japan | LinkedIn\",\"u\":\"https://jp.linkedin.com/in/tadashi-mishima-32638445\"},{\"a\":\"Snowflake, DBT and Dataiku Support. Open Posted 10 minutes ago \\u2022 Ends in 6 days. $10-30 USD. Paid on delivery. Project Title: Snowflake, DBT, Dataiku project Support - Urgent. I am in need of a data engineer freelancer who can provide urgent support for the project. The ideal candidate should have experience and expertise in working with ...\",\"ae\":null,\"c\":\"https://www.freelancer.com/projects/database-administration/snowflake-dbt-dataiku-support\",\"d\":\"www.freelancer.com/projects/database-administration/snowflake-dbt-dataiku-support\",\"da\":\"\",\"e\":\"2024-01-14T00:00:00.0000000\",\"h\":0,\"i\":\"www.freelancer.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Snowflake, DBT and Dataiku Support | Freelancer\",\"u\":\"https://www.freelancer.com/projects/database-administration/snowflake-dbt-dataiku-support\"},{\"a\":\"Dataiku\\u30bb\\u30df\\u30ca\\u30fc\\u300c\\u30c7\\u30fc\\u30bf\\u99c6\\u52d5\\u578b\\u30d3\\u30b8\\u30cd\\u30b9\\u306e\\u8ab2\\u984c\\u3068\\u89e3\\u6c7a\\u6cd5\\u300d\\u3092\\u5f53\\u793e\\u4e3b\\u50ac\\u3067\\u958b\\u50ac\\u3044\\u305f\\u3057\\u307e\\u3059\\u3002 \\u696d\\u52d9\\u5909\\u9769\\u306e\\u63a8\\u9032\\u3092\\u76ee\\u7684\\u3068\\u3057\\u305f\\u3001\\u30c7\\u30fc\\u30bf\\u306e\\u96c6\\u7d04\\u30fb\\u7ba1\\u7406\\u30fb\\u904b\\u7528\\u30fb\\u6d3b\\u7528\\u30d7\\u30ed\\u30bb\\u30b9\\u306e\\u5b9a\\u7740\\u3092\\u3069\\u3046\\u5b9f\\u73fe\\u3059\\u308b\\u304b\\u306f\\u3001\\u30c7\\u30fc\\u30bf\\u30c9\\u30ea\\u30d6\\u30f3\\u7d4c\\u55b6\\u306e\\u5b9f\\u73fe\\u306b\\u304a\\u3051\\u308b\\u3001\\u3088\\u304f\\u3042\\u308b\\u8ab2\\u984c\\u306e1\\u3064\\u3067\\u3059\\u3002\",\"ae\":null,\"c\":\"https://www.intellilink.co.jp/topics/seminar_event/2024/dataiku-fy2023-2h.aspx\",\"d\":\"www.intellilink.co.jp/topics/seminar_event/2024/dataiku-fy2023-2h.aspx\",\"da\":\"translations\",\"e\":\"2024-01-12T00:00:00.0000000\",\"h\":0,\"i\":\"www.intellilink.co.jp\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"\\u5f53\\u793e\\u4e3b\\u50ac\\u30bb\\u30df\\u30ca\\u30fc\\u300c\\u30c7\\u30fc\\u30bf\\u99c6\\u52d5\\u578b\\u30d3\\u30b8\\u30cd\\u30b9\\u306e\\u8ab2\\u984c\\u3068\\u89e3\\u6c7a\\u6cd5\\u300d\\u3092\\u958b\\u50ac | Ntt\\u30c7\\u30fc\\u30bf\\u5148\\u7aef\\u6280\\u8853\\u682a\\u5f0f\\u4f1a\\u793e\",\"u\":\"https://www.intellilink.co.jp/topics/seminar_event/2024/dataiku-fy2023-2h.aspx\"},{\"n\":\"/d.js?q=dataiku&kl=wt-wt&l=wt-wt&p=&s=21&ex=-1&ct=US&sp=0&vqd=4-337400409169293617055811118659485228425\"}]);DDG.duckbar.load('images');DDG.duckbar.load('news');DDG.duckbar.load('videos', {\"ads\":[],\"query\":\"dataiku\",\"queryEncoded\":\"dataiku\",\"response_type\":\"places\",\"results\":[{\"content\":\"https://www.youtube.com/watch?v=Mmt7IluxE0M\",\"description\":\"Learn more about Dataiku and how to better use your enterprise data in this 3 minute demo. CHECK OUT DATAIKU: https://bit.ly/36XBlpK BRIGHTTALK WEBINARS: htt...\",\"duration\":\"3:35\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/Mmt7IluxE0M?autoplay=1\",\"image_token\":\"9fbd7f08d3fe9c5ae80966f30b030c5de63ce962d42298ce9455edc487f55f54\",\"images\":{\"large\":\"https://tse3.mm.bing.net/th?id=OVP.TMea5u6ptVU_48mCeCUUlQEsDh&pid=Api\",\"medium\":\"https://tse3.mm.bing.net/th?id=OVP.TMea5u6ptVU_48mCeCUUlQEsDh&pid=Api\",\"motion\":\"https://tse3.mm.bing.net/th?id=OM.2qQ9AZ4EgLzxPA_1699065185&pid=Api\",\"small\":\"https://tse3.mm.bing.net/th?id=OVP.TMea5u6ptVU_48mCeCUUlQEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2021-09-23T18:53:57.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":45758},\"title\":\"Dataiku 3-Minute Demo\",\"uploader\":\"Dataiku\"},{\"content\":\"https://www.youtube.com/watch?v=ryZRRIjQ5Z8\",\"description\":\"If you're a code-first data practitioner, Dataiku helps you efficiently build high quality data pipelines and models in a number of ways. CHECK OUT DATAIKU: https://bit.ly/36XBlpK EGG ON AIR: https://bit.ly/37GhXMY BRIGHTTALK WEBINARS: https://bit.ly/33TIRjn DATA SCIENCE PIONEERS DOCUMENTARY: https://bit.ly/36V3rBF PARTNER ECOSYSTEM: https ...\",\"duration\":\"10:43\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/ryZRRIjQ5Z8?autoplay=1\",\"image_token\":\"8a4abca8613c6680a108591849e5d7b13b86111004ae004898a7f059b64c8355\",\"images\":{\"large\":\"https://tse4.mm.bing.net/th?id=OVP.WoendyuZJ9qxql-n6jit5AEsDh&pid=Api\",\"medium\":\"https://tse4.mm.bing.net/th?id=OVP.WoendyuZJ9qxql-n6jit5AEsDh&pid=Api\",\"motion\":\"https://tse4.mm.bing.net/th?id=OM1.cmvppfhHVUeE4Q_1684256861&pid=Api\",\"small\":\"https://tse4.mm.bing.net/th?id=OVP.WoendyuZJ9qxql-n6jit5AEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2021-06-08T21:15:02.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":12391},\"title\":\"Dataiku Demo for Data Scientists and Coders\",\"uploader\":\"Dataiku\"},{\"content\":\"https://www.youtube.com/watch?v=1IgcAAPW4fQ\",\"description\":\"Dataiku 11 is now out! This release is jam-packed with features designed to help organizations deliver on the promise of Everyday AI. Check out this video to get an introduction to V11. CHECK OUT DATAIKU: https://bit.ly/36XBlpK DATAIKU ACADEMY: https://bit.ly/2LjsEgZ DATAIKU COMMUNITY: https://bit.ly/2K8lOtV DATA SCIENCE AND ANALYTICS MEETUPS ...\",\"duration\":\"30:52\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/1IgcAAPW4fQ?autoplay=1\",\"image_token\":\"885849fd3f3285ae15a77b9c7e40acc1fbe9d37fa39ac789ecf88e4b889aa796\",\"images\":{\"large\":\"https://tse1.mm.bing.net/th?id=OVP.mQkABdeGdBHH8E6VyTt64AEsDh&pid=Api\",\"medium\":\"https://tse1.mm.bing.net/th?id=OVP.mQkABdeGdBHH8E6VyTt64AEsDh&pid=Api\",\"motion\":\"https://tse1.mm.bing.net/th?id=OM1.mCQeogW3-u_iVw_1684181286&pid=Api\",\"small\":\"https://tse1.mm.bing.net/th?id=OVP.mQkABdeGdBHH8E6VyTt64AEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2022-07-15T15:35:31.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":3314},\"title\":\"Introduction to Dataiku 11!\",\"uploader\":\"Dataiku\"},{\"content\":\"https://www.youtube.com/watch?v=FluiuHuaU8A\",\"description\":\"In this breakout session of Dataiku's Product Days 2021, you will see a demo of Dataiku's Data Science Studio, the centralized, collaborative, and end-to-end platform for data science in the enterprise. CHECK OUT DATAIKU: https://bit.ly/36XBlpK EGG ON AIR: https://bit.ly/37GhXMY BRIGHTTALK WEBINARS: https://bit.ly/33TIRjn DATA SCIENCE PIONEERS ...\",\"duration\":\"13:50\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/FluiuHuaU8A?autoplay=1\",\"image_token\":\"2943fa8c1580f2936fc11667d670c0b827b94ff3d16b897f8b5ef2e2426487b3\",\"images\":{\"large\":\"https://tse2.mm.bing.net/th?id=OVP.RIM-ftwDZjYP58RimJfgwwEsDh&pid=Api\",\"medium\":\"https://tse2.mm.bing.net/th?id=OVP.RIM-ftwDZjYP58RimJfgwwEsDh&pid=Api\",\"motion\":\"https://tse2.mm.bing.net/th?id=OM1.MIQ7BoQz1MVkNw_1662248868&pid=Api\",\"small\":\"https://tse2.mm.bing.net/th?id=OVP.RIM-ftwDZjYP58RimJfgwwEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2021-07-08T15:56:22.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":3844},\"title\":\"Introduction to Dataiku Data Science | Product Days 2021\",\"uploader\":\"Dataiku\"},{\"content\":\"https://www.youtube.com/watch?v=gp8QeJJ4KuE\",\"description\":\"This tutorial is to quickly help users become familiar with the Dataiku platform (DSS). Links for setting up the tutorial. Step 1: https://www.dataiku.com/ Step 2: https://www.dataiku.com/product/get-started/virtualbox/ Step 3: http://127.0.0.1:10000/ Step 4: https://github.com/ageron/handson-ml Step 5: https://raw.githubusercontent.com/ageron ...\",\"duration\":\"10:24\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/gp8QeJJ4KuE?autoplay=1\",\"image_token\":\"36a6e666bdcb9e34aacad09f504181152667f35deab62b50ee48da1a76c38303\",\"images\":{\"large\":\"https://tse1.mm.bing.net/th?id=OVP.htgX0HRO9l8nlfoFzmlA5AHgFo&pid=Api\",\"medium\":\"https://tse1.mm.bing.net/th?id=OVP.htgX0HRO9l8nlfoFzmlA5AHgFo&pid=Api\",\"motion\":\"https://tse1.mm.bing.net/th?id=OM2.SsfUJx35-DP9OA_1632775689&pid=Api\",\"small\":\"https://tse1.mm.bing.net/th?id=OVP.htgX0HRO9l8nlfoFzmlA5AHgFo&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2020-06-09T17:48:44.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":37512},\"title\":\"Get started with Dataiku | From data to machine learning predictions in 10 minutes\",\"uploader\":\"Jose RazGuzman\"},{\"content\":\"https://www.youtube.com/watch?v=S6AY-q_5Bd0\",\"description\":\"Learn more about Dataiku's demand forecast solution to optimize your sorting and production planning, inventory management, and much more. Without you, it's just data. Learn more at https://www.dataiku.com/without-you/ CHECK OUT DATAIKU: https://bit.ly/36XBlpK BRIGHTTALK WEBINARS: https://bit.ly/33TIRjn DATA SCIENCE PIONEERS DOCUMENTARY: https ...\",\"duration\":\"2:00\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/S6AY-q_5Bd0?autoplay=1\",\"image_token\":\"456216e1949ad91527408a297243af22822d03602cc51e3fb1c7324680e27e90\",\"images\":{\"large\":\"https://tse2.mm.bing.net/th?id=OVP.B1QN6Sk8tATAQDmEZgvp8wEsDh&pid=Api\",\"medium\":\"https://tse2.mm.bing.net/th?id=OVP.B1QN6Sk8tATAQDmEZgvp8wEsDh&pid=Api\",\"motion\":\"https://tse2.mm.bing.net/th?id=OM2.BiTArkALFRqLyQ_1684244069&pid=Api\",\"small\":\"https://tse2.mm.bing.net/th?id=OVP.B1QN6Sk8tATAQDmEZgvp8wEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2022-08-09T20:43:58.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":569},\"title\":\"Improve Your Demand Forecasting with Dataiku\",\"uploader\":\"Dataiku\"},{\"content\":\"https://www.youtube.com/watch?v=tyd262JRo9g\",\"description\":\"In this session from Everyday AI Tech Day 2023, hear from Jacqueline Kuo, one of our solutions engineers, on how to build sustainable pipelines. Optimizing and automating data pipelines to transform, prepare, and analyze data on an ongoing basis is critical for production-ready AI projects. In this session, learn how Dataiku supports the ...\",\"duration\":\"20:35\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/tyd262JRo9g?autoplay=1\",\"image_token\":\"e946d16ff6600b2df00b4cddf4f97fa56d41df90c0a1608bcd33fedf69af63bd\",\"images\":{\"large\":\"https://tse2.mm.bing.net/th?id=OVP.xmckza2EYvyQk4nTqTsfZgEsDh&pid=Api\",\"medium\":\"https://tse2.mm.bing.net/th?id=OVP.xmckza2EYvyQk4nTqTsfZgEsDh&pid=Api\",\"motion\":\"https://tse2.mm.bing.net/th?id=OM1.VfASmCqEmyb4NA_1700289203&pid=Api\",\"small\":\"https://tse2.mm.bing.net/th?id=OVP.xmckza2EYvyQk4nTqTsfZgEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-10-20T12:35:12.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":98},\"title\":\"A Well-Oiled Machine: Create Sustainable Data Pipelines With Dataiku\",\"uploader\":\"Dataiku\"},{\"content\":\"https://www.youtube.com/watch?v=-amc9iVauuE\",\"description\":\"Dataiku is the leading platform for Everyday AI, systemizing the use of data for exceptional business results. In today's video we will take a tour of Dataiku's end to end capabilities by exploring a real life use case around environmental impact. Let's take a look at how a data science team with different skills can work together to turn ...\",\"duration\":\"12:35\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/-amc9iVauuE?autoplay=1\",\"image_token\":\"2a05a65ad8a2727aa5c48b8daa7f9ec363a24d4336a3509016d4b200c9d003cd\",\"images\":{\"large\":\"https://tse1.mm.bing.net/th?id=OVP.Az9RhdSVwpXe56mGcs6FqQEsDh&pid=Api\",\"medium\":\"https://tse1.mm.bing.net/th?id=OVP.Az9RhdSVwpXe56mGcs6FqQEsDh&pid=Api\",\"motion\":\"https://tse1.mm.bing.net/th?id=OM1.Q2OhN9DzfowU6A_1685345657&pid=Api\",\"small\":\"https://tse1.mm.bing.net/th?id=OVP.Az9RhdSVwpXe56mGcs6FqQEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-01-09T21:12:27.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":9768},\"title\":\"End to End Demo 2023\",\"uploader\":\"Dataiku\"},{\"content\":\"https://www.youtube.com/watch?v=MxKNdVNyLJY\",\"description\":\"Simply collecting vast amounts of data isn't enough to unlock its potentially massive value to your business. In this video, we will explore Dataiku's data preparation capabilities that will help you access, cleanse, transform, and enrich data faster than ever before. Timestamps: 0:00 Introduction 0:53 The Flow 1:18 Accessing and Exploring ...\",\"duration\":\"6:12\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/MxKNdVNyLJY?autoplay=1\",\"image_token\":\"652157f0560bd0c12f1731a0c4876335e712bb986dda679c0157545e9083ab50\",\"images\":{\"large\":\"https://tse4.mm.bing.net/th?id=OVP.xh9SlNNNrKuJLXO5kEtsLgEsDh&pid=Api\",\"medium\":\"https://tse4.mm.bing.net/th?id=OVP.xh9SlNNNrKuJLXO5kEtsLgEsDh&pid=Api\",\"motion\":\"https://tse4.mm.bing.net/th?id=OM2.u85J8gWHLhZhSA_1680067810&pid=Api\",\"small\":\"https://tse4.mm.bing.net/th?id=OVP.xh9SlNNNrKuJLXO5kEtsLgEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-01-09T20:46:27.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":2948},\"title\":\"Key Capabilities: Data Preparation\",\"uploader\":\"Dataiku\"},{\"content\":\"https://www.youtube.com/watch?v=GJA_PAnqGY8\",\"description\":\"In this video we walk through a series of real-world data analysis tasks using a Netflix movie & TV show dataset. We start by solving the tasks using the Python Pandas library. We then complete the same problems using the Dataiku Data Science Studio. Being knowledgeable about various tools in the data science space is very important to becoming ...\",\"duration\":\"58:15\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/GJA_PAnqGY8?autoplay=1\",\"image_token\":\"e00b743925487c5466b2bf1a024869f528876917c05bdd1e7cce1d78ad3d9a3a\",\"images\":{\"large\":\"https://tse2.mm.bing.net/th?id=OVP.D7OGs04gyoaehil3qvxX7gEsDh&pid=Api\",\"medium\":\"https://tse2.mm.bing.net/th?id=OVP.D7OGs04gyoaehil3qvxX7gEsDh&pid=Api\",\"motion\":\"https://tse2.mm.bing.net/th?id=OM1.3R2zQEuONEzSww_1664997993&pid=Api\",\"small\":\"https://tse2.mm.bing.net/th?id=OVP.D7OGs04gyoaehil3qvxX7gEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2022-08-03T15:00:11.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":7769},\"title\":\"Solving Real-World Data Analysis Tasks with Python Pandas & Dataiku DSS (Movie Analysis)\",\"uploader\":\"Recall by Dataiku\"}],\"vqd\":{\"dataiku\":\"4-337400409169293617055811118659485228425\"}});DDG.duckbar.loadModule('related_searches', {\"ads\":[],\"query\":\"dataiku\",\"queryEncoded\":\"dataiku\",\"response_type\":\"places\",\"results\":[{\"display_text\":\"dataiku login\",\"text\":\"dataiku login\",\"web_search_url\":\"?q=dataiku%20login\"},{\"display_text\":\"dataiku japan\",\"text\":\"dataiku japan\",\"web_search_url\":\"?q=dataiku%20japan\"},{\"display_text\":\"dataiku vs tableau\",\"text\":\"dataiku vs tableau\",\"web_search_url\":\"?q=dataiku%20vs%20tableau\"},{\"display_text\":\"dataiku vs alteryx\",\"text\":\"dataiku vs alteryx\",\"web_search_url\":\"?q=dataiku%20vs%20alteryx\"},{\"display_text\":\"dataiku vs databricks\",\"text\":\"dataiku vs databricks\",\"web_search_url\":\"?q=dataiku%20vs%20databricks\"},{\"display_text\":\"what is dataiku used for\",\"text\":\"what is dataiku used for\",\"web_search_url\":\"?q=what%20is%20dataiku%20used%20for\"},{\"display_text\":\"how to pronounce dataiku\",\"text\":\"how to pronounce dataiku\",\"web_search_url\":\"?q=how%20to%20pronounce%20dataiku\"},{\"display_text\":\"dataiku products\",\"text\":\"dataiku products\",\"web_search_url\":\"?q=dataiku%20products\"}],\"vqd\":{\"dataiku\":\"4-337400409169293617055811118659485228425\"}});if (DDG.pageLayout) DDG.pageLayout.initialize({\"mainline\":{\"items\":[[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"videos\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"related_searches\"]]},\"sidebar\":{\"items\":[[\"wikipedia_fathead\"]]}}, { start: 0 });DDG.deep.emit(\"load:completed\");", + "curl-cffi-POST-https://duckduckgo.com-{\"data\": {\"q\": \"Dataiku machine learning platform\"}}": "Dataiku machine learning platform at DuckDuckGo
", + "curl-cffi-GET-https://links.duckduckgo.com/d.js-{\"params\": {\"bing_market\": \"wt-WT\", \"df\": null, \"ex\": \"-1\", \"kl\": \"wt-wt\", \"l\": \"wt-wt\", \"q\": \"Dataiku machine learning platform\", \"s\": \"0\", \"sp\": \"0\", \"vqd\": \"4-135627661863888098952136153695171249901\"}}": "if (DDG.deep && DDG.deep.setUpstream) DDG.deep.setUpstream(\"bingv7aa\");DDG.deep.bn={'ivc':1};if (DDG.pageLayout) DDG.pageLayout.load('a',[], {\"page_load_url\":\"https://duckduckgo.com/y.js?iurl=%7B2%7DIG%3DD926DA28CBA14E6D9DA10C17531E1D6B%26CID%3D37369241D5576E2736ED8647D4286F5F%26Type%3DEvent.CPT%26DATA%3D0\"});DDG.deep.signalSummary = \"\";DDG.inject('DDG.Data.languages.resultLanguages', {\"en\":[\"https://www.dataiku.com/\",\"https://www.dataiku.com/product/key-capabilities/machine-learning/\",\"https://www.dataiku.com/product/key-capabilities/mlops/\",\"https://www.dataiku.com/product/key-capabilities/\",\"https://developer.dataiku.com/latest/tutorials/machine-learning/index.html\",\"https://knowledge.dataiku.com/latest/ml-analytics/index.html\",\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/dataiku/product/dataiku\",\"https://www.dataiku.com/product/dataiku-as-a-managed-service/\",\"https://pages.dataiku.com/dataiku-enterprise-ai-info\",\"https://blog.dataiku.com/dataiku-ml-key-capabilities\",\"https://discover.dataiku.com/data-scientists/\",\"https://academy.dataiku.com/\",\"https://academy.dataiku.com/machine-learning-basics\",\"https://www.infoworld.com/article/3618837/dataiku-review-data-science-fit-for-the-enterprise.html\",\"https://pages.dataiku.com/gartner-2021-pr\",\"https://blog.dataiku.com/what-is-machine-learning-model-deployment\",\"https://datascientest.com/en/dataiku-a-must-have-tool-for-data-science-and-ai\",\"https://blog.dataiku.com/gartner-2020\",\"https://www.infoq.com/news/2024/01/instacart-machine-learning/\",\"https://seekingalpha.com/article/4662201-c3ai-missing-the-boat\",\"https://jobs.apple.com/en-us/details/200516237/aiml-software-engineer-machine-learning-platform-infrastructure\",\"https://www.tokyodev.com/companies/rapyuta-robotics\",\"https://www.servicenow.com/partners/partner-finder/ntt-data-corp.html\",\"https://www.linkedin.com/company/maruha-nichiro-corporation\",\"https://www.gatech.edu/event/2024/01/12/access-big-data-machine-learning-workshop\",\"https://apply.workable.com/rapyuta-robotics/j/3ADFC72B04/\"]});DDG.deep.pageLayoutSummary = \"w26r1,e1\";DDG.inject('DDG.Data.languages.adLanguages', {});if (DDG.pageLayout) DDG.pageLayout.load('d',[{\"a\":\"MLOps Deploy, monitor, and maintain machine learning models, all in a single platform. Explore the Capability Collaboration With Dataiku, teams can move beyond the lab and build real and safe Generative AI applications at enterprise scale. Explore the Capability Governance\",\"ae\":null,\"c\":\"https://www.dataiku.com/\",\"d\":\"www.dataiku.com\",\"da\":\"\",\"h\":0,\"i\":\"www.dataiku.com\",\"k\":null,\"l\":[{\"snippet\":\"The Platform for Everyday AI. Empower people across your business to do more with data and AI, build projects faster, and work together, all in a shared and safe environment. With Dataiku, everyone can get involved in data and AI projects on a single platform for design and production that delivers use cases in days, not months.\",\"targetUrl\":\"https://www.dataiku.com/product/\",\"text\":\"Product\"},{\"snippet\":\"Check out Dataiku's job openings and apply to help companies transform raw data into business impacting predictions and products.\",\"targetUrl\":\"https://www.dataiku.com/careers/\",\"text\":\"Careers\"},{\"snippet\":\"A Single Platform For. Generative AI; Data Preparation; Visualization; Machine Learning ... Get started with the basics, quickly move on to advanced courses, and become a certified user on Dataiku's online learning and certification platform. Master Dataiku ... Learn how Dataiku makes it easy to build machine learning models, deploy them to a ...\",\"targetUrl\":\"https://www.dataiku.com/learn/\",\"text\":\"Learn\"},{\"snippet\":\"Thrive SPC Uses Dataiku, Snowflake, and Snow Fox Data to Improve Clinical Home Care. By moving to Dataiku and working with Dataiku partners, Snowflake and Snow Fox Data, Thrive Skilled Pediatric Care (Thrive SPC) has been able to advance from complicated spreadsheets to a central platform that provides clear insights and metrics to fuel their data-driven healthcare solutions.\",\"targetUrl\":\"https://www.dataiku.com/stories/\",\"text\":\"Stories\"},{\"snippet\":\"Dataiku was founded on the principle that in order to succeed in the world's rapidly evolving ecosystem, companies \\u2014 no matter what their industry or size \\u2014 must elevate their people to continuously innovate. Since 2013, Dataiku has been the leader in democratizing data and empowering organization-wide collaboration.\",\"targetUrl\":\"https://www.dataiku.com/company/\",\"text\":\"Company\"},{\"snippet\":\"Join the Dataiku Partner Ecosystem. Become a part of the growing Dataiku service partner ecosystem or get in touch to talk integrations and technical partnerships.\",\"targetUrl\":\"https://www.dataiku.com/partners/\",\"text\":\"Partners\"}],\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku | Everyday AI, Extraordinary People\",\"u\":\"https://www.dataiku.com/\"},{\"a\":\"AI and Machine Learning with Dataiku Build and evaluate advanced machine learning models using AutoML and the latest AI techniques. See Dataiku ML in Action START FOR FREE Visualization DataOps Feature Engineering\",\"ae\":null,\"c\":\"https://www.dataiku.com/product/key-capabilities/machine-learning/\",\"d\":\"www.dataiku.com/product/key-capabilities/machine-learning/\",\"da\":\"\",\"h\":0,\"i\":\"www.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"AI and Machine Learning with Dataiku | Dataiku\",\"u\":\"https://www.dataiku.com/product/key-capabilities/machine-learning/\"},{\"a\":\"Product Dataiku Key Capabilities MLOps with Dataiku MLOps with Dataiku Deploy, monitor, and manage machine learning models and projects in production. See MLOps in Action START FOR FREE DataOps Analytic Apps Deploying Projects to Production\",\"ae\":null,\"c\":\"https://www.dataiku.com/product/key-capabilities/mlops/\",\"d\":\"www.dataiku.com/product/key-capabilities/mlops/\",\"da\":\"\",\"h\":0,\"i\":\"www.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MLOps with Dataiku | Dataiku\",\"u\":\"https://www.dataiku.com/product/key-capabilities/mlops/\"},{\"a\":\"AI & Machine Learning Dataiku AutoML accelerates the model development process with a guided framework for AI and machine learning including prompt engineering, prediction, clustering, time series forecasting, computer vision tasks, causal ML, and more.\",\"ae\":null,\"c\":\"https://www.dataiku.com/product/key-capabilities/\",\"d\":\"www.dataiku.com/product/key-capabilities/\",\"da\":\"\",\"h\":0,\"i\":\"www.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku Key Capabilities | Dataiku\",\"u\":\"https://www.dataiku.com/product/key-capabilities/\"},{\"a\":\"This tutorial section contains learning material on programmatically training, managing and deploying machine learning models in Dataiku. Generative AI - NLP Programmatic RAG with Dataiku's LLM Mesh and Langchain Using LLM Mesh to benchmark zero-shot classification models GPT-based zero-shot text classification with the OpenAI API\",\"ae\":null,\"c\":\"https://developer.dataiku.com/latest/tutorials/machine-learning/index.html\",\"d\":\"developer.dataiku.com/latest/tutorials/machine-learning/index.html\",\"da\":\"\",\"h\":0,\"i\":\"developer.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Machine Learning - Dataiku Developer Guide\",\"u\":\"https://developer.dataiku.com/latest/tutorials/machine-learning/index.html\"},{\"a\":\"Dataiku supports a wide range of machine learning and analytic tasks, such as prediction, clustering, time series, image classification and much more. Explore the resources here for improving your building machine learning models and analytics tasks. Tip\",\"ae\":null,\"c\":\"https://knowledge.dataiku.com/latest/ml-analytics/index.html\",\"d\":\"knowledge.dataiku.com/latest/ml-analytics/index.html\",\"da\":\"\",\"h\":0,\"i\":\"knowledge.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Machine Learning & Analytics - Dataiku Knowledge Base\",\"u\":\"https://knowledge.dataiku.com/latest/ml-analytics/index.html\"},{\"a\":\"by Dataiku in Data Science and Machine Learning Platforms 4.8 504 Ratings compare_arrows Compare rate_review Write a Review download_2 Download PDF Related markets: Dataiku in Data Preparation Tools (18 Reviews), Dataiku in Cloud AI Developer Services (14 Reviews) Overview Reviews Alternatives Likes and Dislikes Dataiku Ratings Overview\",\"ae\":null,\"c\":\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/dataiku/product/dataiku\",\"d\":\"www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/dataiku/product/dataiku\",\"da\":\"\",\"h\":0,\"i\":\"www.gartner.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku Reviews, Ratings & Features 2024 | Gartner Peer Insights\",\"u\":\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/dataiku/product/dataiku\"},{\"a\":\"The fully managed data science and machine learning platform for your team to create AI and analytics insights. "Dataiku Cloud allows us to focus on analysis, not server administration. Data insights fuel our growth, and Dataiku Cloud enables us to develop insights faster than our competitors." Scott Walker, Managing Partner Sarissa Partners\",\"ae\":null,\"c\":\"https://www.dataiku.com/product/dataiku-as-a-managed-service/\",\"d\":\"www.dataiku.com/product/dataiku-as-a-managed-service/\",\"da\":\"\",\"h\":0,\"i\":\"www.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku Cloud | Dataiku\",\"u\":\"https://www.dataiku.com/product/dataiku-as-a-managed-service/\"},{\"a\":\"Dataiku is the end-to-end platform democratizing access to data. Manage the entire data science workflow from data prep to auto ML to model maintenance. ... Dataiku offers the latest machine learning technologies all in one place, including: Automated machine learning (AutoML) - choose between several ML backends to train models. ...\",\"ae\":null,\"c\":\"https://pages.dataiku.com/dataiku-enterprise-ai-info\",\"d\":\"pages.dataiku.com/dataiku-enterprise-ai-info\",\"da\":\"\",\"h\":0,\"i\":\"pages.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku: Your Path to Enterprise AI\",\"u\":\"https://pages.dataiku.com/dataiku-enterprise-ai-info\"},{\"a\":\"Dataiku Makes Machine Learning Customizable, Accessible, & Transparent May 17, 2023 Dataiku Product Lauren Anderson It only takes a quick look around to see that the use of machine learning (ML) is more prevalent across industries than ever before!\",\"ae\":null,\"c\":\"https://blog.dataiku.com/dataiku-ml-key-capabilities\",\"d\":\"blog.dataiku.com/dataiku-ml-key-capabilities\",\"da\":\"\",\"e\":\"2023-05-17T00:00:00.0000000\",\"h\":0,\"i\":\"blog.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku Makes Machine Learning Customizable, Accessible, & Transparent\",\"u\":\"https://blog.dataiku.com/dataiku-ml-key-capabilities\"},{\"a\":\"Jump Right In Dataiku is an end-to-end data and machine learning platform. Build and maintain predictive models throughout their entire lifecycles while pushing computation to the most efficient engines.\",\"ae\":null,\"c\":\"https://discover.dataiku.com/data-scientists/\",\"d\":\"discover.dataiku.com/data-scientists/\",\"da\":\"\",\"h\":0,\"i\":\"discover.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Data Scientists - Discover Dataiku\",\"u\":\"https://discover.dataiku.com/data-scientists/\"},{\"a\":\"Academy Your Path to Dataiku Mastery Quick Starts Follow our quick starts that introduce using Dataiku for different tasks View More Learning Paths From novice to expert, follow guided sets of curriculums to master Dataiku View More Certifications Test your Dataiku knowledge in key thematic areas View More Crash Course\",\"ae\":null,\"c\":\"https://academy.dataiku.com/\",\"d\":\"academy.dataiku.com\",\"da\":\"\",\"h\":0,\"i\":\"academy.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku Academy\",\"u\":\"https://academy.dataiku.com/\"},{\"a\":\"The Machine Learning Course is designed to provide a first hands-on overview of basic Dataiku DSS machine learning concepts so that you can easily create and evaluate your first models in DSS. Completion of this course will enable you to move on to more advanced courses. In this course, we'll work with two use cases.\",\"ae\":null,\"c\":\"https://academy.dataiku.com/machine-learning-basics\",\"d\":\"academy.dataiku.com/machine-learning-basics\",\"da\":\"\",\"h\":0,\"i\":\"academy.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Machine Learning Basics - Dataiku\",\"u\":\"https://academy.dataiku.com/machine-learning-basics\"},{\"a\":\"Dataiku Data Science Studio (DSS) is a platform that tries to span the needs of data scientists, data engineers, business analysts, and AI consumers. It mostly succeeds. In addition, Dataiku...\",\"ae\":null,\"c\":\"https://www.infoworld.com/article/3618837/dataiku-review-data-science-fit-for-the-enterprise.html\",\"d\":\"www.infoworld.com/article/3618837/dataiku-review-data-science-fit-for-the-enterprise.html\",\"da\":\"\",\"h\":0,\"i\":\"www.infoworld.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku review: Data science fit for the enterprise | InfoWorld\",\"u\":\"https://www.infoworld.com/article/3618837/dataiku-review-data-science-fit-for-the-enterprise.html\"},{\"a\":\"NEW YORK - March 4, 2021 - Today Dataiku, the world's most advanced Enterprise AI platform, was named a Leader in the Gartner 2021 Magic Quadrant for Data Science and Machine-Learning Platforms, marking its second consecutive year in the Leaders quadrant.Dataiku believes the placement amid the fast-moving market for AI tools cements its position as the driving force behind breakthroughs in ...\",\"ae\":null,\"c\":\"https://pages.dataiku.com/gartner-2021-pr\",\"d\":\"pages.dataiku.com/gartner-2021-pr\",\"da\":\"translations\",\"h\":0,\"i\":\"pages.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku Again Named a Leader in the Gartner 2021 Magic Quadrant for ...\",\"u\":\"https://pages.dataiku.com/gartner-2021-pr\"},{\"a\":\"An ML model is considered in production once it's been successfully deployed and being used by end users to realize business value. This article will shed more light on what exactly model deployment means and how Dataiku's end-to-end platform makes the model deployment process seamless. Why Is Model Deployment So Important (and So Hard)?\",\"ae\":null,\"c\":\"https://blog.dataiku.com/what-is-machine-learning-model-deployment\",\"d\":\"blog.dataiku.com/what-is-machine-learning-model-deployment\",\"da\":\"\",\"e\":\"2023-04-10T00:00:00.0000000\",\"h\":0,\"i\":\"blog.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"What Is Machine Learning Model Deployment? - Dataiku\",\"u\":\"https://blog.dataiku.com/what-is-machine-learning-model-deployment\"},{\"a\":\"A visual interface makes it very easy to apply Machine Learning models. Additionally, the platform-as-a-service approach eliminates the need for infrastructure. Furthermore, Dataiku is also compatible with Bayesian search. This allows running a second AI model in a loop to test different settings and parameters until the optimal configuration ...\",\"ae\":null,\"c\":\"https://datascientest.com/en/dataiku-a-must-have-tool-for-data-science-and-ai\",\"d\":\"datascientest.com/en/dataiku-a-must-have-tool-for-data-science-and-ai\",\"da\":\"\",\"e\":\"2023-11-28T00:00:00.0000000\",\"h\":0,\"i\":\"datascientest.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku: A must-have tool for Data Science and AI\",\"u\":\"https://datascientest.com/en/dataiku-a-must-have-tool-for-data-science-and-ai\"},{\"a\":\"Dataiku: A Gartner Magic Quadrant Leader in Data Science and Machine-Learning Platforms February 17, 2020 Dataiku Company, Dataiku Product Lynn Heidmann Our 2019 ended with a bang with the announcement that Dataiku became a unicorn valued at $1.4 billion and gained a new investor (CapitalG).\",\"ae\":null,\"c\":\"https://blog.dataiku.com/gartner-2020\",\"d\":\"blog.dataiku.com/gartner-2020\",\"da\":\"translations\",\"h\":0,\"i\":\"blog.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku: A Gartner Magic Quadrant Leader in Data Science and Machine ...\",\"u\":\"https://blog.dataiku.com/gartner-2020\"},{\"a\":\"Instacart introduced its original Griffin platform in 2022 to support its journey toward leveraging machine learning for product development. Using a unified platform helped triple the number of ...\",\"ae\":null,\"c\":\"https://www.infoq.com/news/2024/01/instacart-machine-learning/\",\"d\":\"www.infoq.com/news/2024/01/instacart-machine-learning/\",\"da\":\"translations\",\"e\":\"2024-01-01T08:02:59.0000000\",\"h\":0,\"i\":\"www.infoq.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Griffin 2.0: Instacart Revamps Its Machine Learning Platform - InfoQ\",\"u\":\"https://www.infoq.com/news/2024/01/instacart-machine-learning/\"},{\"a\":\"Their core product is Data Science Studio, which is focused on cross-discipline collaboration and ease of use and enables users to start machine-learning projects rapidly. Dataiku is focused on ...\",\"ae\":null,\"c\":\"https://seekingalpha.com/article/4662201-c3ai-missing-the-boat\",\"d\":\"seekingalpha.com/article/4662201-c3ai-missing-the-boat\",\"da\":\"translations\",\"e\":\"2024-01-10T19:38:00.0000000\",\"h\":0,\"i\":\"seekingalpha.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"C3.ai: Missing The Boat (NYSE:AI) | Seeking Alpha\",\"u\":\"https://seekingalpha.com/article/4662201-c3ai-missing-the-boat\"},{\"a\":\"The Information Intelligence teams are building groundbreaking technology for algorithmic search, machine learning, natural language processing, and artificial intelligence. The features we build are redefining how hundreds of millions of people use their computers and mobile devices to search and find what they are looking for.\",\"ae\":null,\"b\":\"apple\\tApple\\twww.apple.com\",\"c\":\"https://jobs.apple.com/en-us/details/200516237/aiml-software-engineer-machine-learning-platform-infrastructure\",\"d\":\"jobs.apple.com/en-us/details/200516237/aiml-software-engineer-machine-learning-platform-infrastructure\",\"da\":\"translations\",\"e\":\"2024-01-09T00:00:00.0000000\",\"h\":0,\"i\":\"jobs.apple.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Software Engineer, Machine Learning Platform & Infrastructure - Apple\",\"u\":\"https://jobs.apple.com/en-us/details/200516237/aiml-software-engineer-machine-learning-platform-infrastructure\"},{\"a\":\"One platform for all your robotics needs. Rapyuta Robotics aims at building low\\u00ad cost, lightweight autonomous mobile robots with high-level intelligence distributed in the cloud, enabling such robots to offload some of their heavy computation and seamlessly learn and share experiences with one another. Live your best life - at work and outside\",\"ae\":null,\"c\":\"https://www.tokyodev.com/companies/rapyuta-robotics\",\"d\":\"www.tokyodev.com/companies/rapyuta-robotics\",\"da\":\"\",\"h\":0,\"i\":\"www.tokyodev.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Rapyuta Robotics | TokyoDev\",\"u\":\"https://www.tokyodev.com/companies/rapyuta-robotics\"},{\"a\":\"NTT DATA - a part of NTT Group - is a trusted global innovator of IT and business services headquartered in Tokyo. We help clients transform through consulting, industry solutions, business process services, IT modernization and managed services. NTT DATA enables clients, as well as society, to move confidently into the digital future. We are committed to our clients' long-term success ...\",\"ae\":null,\"c\":\"https://www.servicenow.com/partners/partner-finder/ntt-data-corp.html\",\"d\":\"www.servicenow.com/partners/partner-finder/ntt-data-corp.html\",\"da\":\"\",\"h\":0,\"i\":\"www.servicenow.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"NTT DATA Corporation - ServiceNow\",\"u\":\"https://www.servicenow.com/partners/partner-finder/ntt-data-corp.html\"},{\"a\":\"Maruha Nichiro Corporation Food Production Toyosu, Koto-ku, Tokyo 1,941 followers "Bringing Delicious Delight to the World."\",\"ae\":null,\"b\":\"li\\tLinkedIn\\twww.linkedin.com\",\"c\":\"https://www.linkedin.com/company/maruha-nichiro-corporation\",\"d\":\"www.linkedin.com/company/maruha-nichiro-corporation\",\"da\":\"\",\"h\":0,\"i\":\"www.linkedin.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Maruha Nichiro Corporation | LinkedIn\",\"u\":\"https://www.linkedin.com/company/maruha-nichiro-corporation\"},{\"a\":\"PACE and SoX in collaboration with ACCESS and the Pittsburgh Supercomputing Center are pleased to host an HPC workshop on Big Data & Machine Learning, to be held on January 29 and 31, 2024. This workshop will focus on topics including big data analytics and machine learning with Spark, and deep learning using Tensorflow. It will have a hands-on component using the Bridges-2 computing platform ...\",\"ae\":null,\"c\":\"https://www.gatech.edu/event/2024/01/12/access-big-data-machine-learning-workshop\",\"d\":\"www.gatech.edu/event/2024/01/12/access-big-data-machine-learning-workshop\",\"da\":\"translations\",\"e\":\"2024-01-12T00:00:00.0000000\",\"h\":0,\"i\":\"www.gatech.edu\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"ACCESS Big Data & Machine Learning Workshop - Georgia Institute of ...\",\"u\":\"https://www.gatech.edu/event/2024/01/12/access-big-data-machine-learning-workshop\"},{\"a\":\"Start date: January 2024 or later. Our Tokyo engineering team is looking for robotics software interns for a minimum duration of six months capable of supporting the team to build state-of-the-art, scalable, autonomous mobile robots. The team works closely with some of the leading Japanese companies to build pioneering robotics solutions by leveraging rapyuta.io, our cloud robotics platform.\",\"ae\":null,\"c\":\"https://apply.workable.com/rapyuta-robotics/j/3ADFC72B04/\",\"d\":\"apply.workable.com/rapyuta-robotics/j/3ADFC72B04/\",\"da\":\"translations\",\"h\":0,\"i\":\"apply.workable.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Robotics Software Intern 2024 - Rapyuta Robotics\",\"u\":\"https://apply.workable.com/rapyuta-robotics/j/3ADFC72B04/\"},{\"n\":\"/d.js?q=Dataiku%20machine%20learning%20platform&kl=wt-wt&l=wt-wt&p=&s=26&ex=-1&ct=US&sp=0&vqd=4-135627661863888098952136153695171249901\"}]);DDG.duckbar.load('images');DDG.duckbar.load('news');DDG.duckbar.load('videos');DDG.duckbar.loadModule('related_searches', {\"ads\":[],\"query\":\"Dataiku machine learning platform\",\"queryEncoded\":\"Dataiku%20machine%20learning%20platform\",\"response_type\":\"places\",\"results\":[{\"display_text\":\"dataiku online machine learning\",\"text\":\"dataiku online machine learning\",\"web_search_url\":\"?q=dataiku%20online%20machine%20learning\"},{\"display_text\":\"dataiku machine learning plugin\",\"text\":\"dataiku machine learning plugin\",\"web_search_url\":\"?q=dataiku%20machine%20learning%20plugin\"},{\"display_text\":\"dataiku machine learning model\",\"text\":\"dataiku machine learning model\",\"web_search_url\":\"?q=dataiku%20machine%20learning%20model\"},{\"display_text\":\"dataiku machine learning extension\",\"text\":\"dataiku machine learning extension\",\"web_search_url\":\"?q=dataiku%20machine%20learning%20extension\"},{\"display_text\":\"automated machine learning dataiku\",\"text\":\"automated machine learning dataiku\",\"web_search_url\":\"?q=automated%20machine%20learning%20dataiku\"},{\"display_text\":\"dataiku production server\",\"text\":\"dataiku production server\",\"web_search_url\":\"?q=dataiku%20production%20server\"},{\"display_text\":\"dataiku automated automation\",\"text\":\"dataiku automated automation\",\"web_search_url\":\"?q=dataiku%20automated%20automation\"},{\"display_text\":\"dataiku key capabilities\",\"text\":\"dataiku key capabilities\",\"web_search_url\":\"?q=dataiku%20key%20capabilities\"}],\"vqd\":{\"Dataiku%20machine%20learning%20platform\":\"4-135627661863888098952136153695171249901\"}});if (DDG.pageLayout) DDG.pageLayout.initialize({\"mainline\":{\"items\":[[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"related_searches\"]]},\"sidebar\":{\"items\":[[\"wikipedia_fathead\"]]}}, { start: 0 });DDG.deep.emit(\"load:completed\");", + "curl-cffi-POST-https://duckduckgo.com-{\"data\": {\"q\": \"DataRobot AI platform comparison\"}}": "DataRobot AI platform comparison at DuckDuckGo
", + "curl-cffi-GET-https://links.duckduckgo.com/d.js-{\"params\": {\"bing_market\": \"wt-WT\", \"df\": null, \"ex\": \"-1\", \"kl\": \"wt-wt\", \"l\": \"wt-wt\", \"q\": \"DataRobot AI platform comparison\", \"s\": \"0\", \"sp\": \"0\", \"vqd\": \"4-58620545585474558767320902709740831322\"}}": "if (DDG.deep && DDG.deep.setUpstream) DDG.deep.setUpstream(\"bingv7aa\");DDG.deep.bn={'ivc':1};if (DDG.pageLayout) DDG.pageLayout.load('a',[{\"a\":\"\\u9ad8\\u7cbe\\u5ea6\\u306a\\u6a5f\\u68b0\\u5b66\\u7fd2\\u30e2\\u30c7\\u30eb\\u3092\\u69cb\\u7bc9\\u3001\\u5b9f\\u88c5\\u3001\\u904b\\u7528\\u3002DataRobot\\u306f\\u793e\\u5185\\u30c7\\u30fc\\u30bf\\u304b\\u3089\\u65b0\\u3057\\u3044\\u4fa1\\u5024\\u3092\\u5275\\u9020\\u3057\\u307e\\u3059. DataRobot\\u306f\\u7c21\\u5358\\u306a\\u64cd\\u4f5c\\u3067\\u30c7\\u30fc\\u30bf\\u304b\\u3089\\u4fa1\\u5024\\u3092\\u5275\\u51fa\",\"adext\":{\"filterlinks\":{\"l\":[],\"tid\":\"\"},\"sitelinks\":{\"l\":[{\"snippet\":\"Explore the DataRobot AI Platform Get Started With a 30-Day Trial\",\"targetUrl\":\"https://duckduckgo.com/y.js?ad_domain=datarobot.com&ad_provider=bingv7aa&ad_type=txad&eddgt=TADjssf2xUPiUP4Z%2DpL%2Djw%3D%3D&rut=458d772e8fcb4e4324d9389389d604bf4a83994e8046495904402de9aadd8689&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De8wZWcT1GcY01mZtMuvlFYRDVUCUwR59UrUKaRMOBuANnCWi%2D8iLso5PuAoywij1cMeNLxieP5AAeMQUBqIlxZTsdWNA7YZoBicc1jtvLW_ZEmp6X0lumVtbFw9IJCDFojWdEcfYE0O0SvTWErWCBN9jCLQwEegRDrKirFobgUYBqEYLfhMVAeD3x862y3EMIPxcrug60dWItb0_gnTv06GzMdT9Q%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuZGF0YXJvYm90LmNvbSUyZnRyaWFsJTJmJTNmdXRtX21lZGl1bSUzZHNlYXJjaCUyNnV0bV9zb3VyY2UlM2RiaW5nJTI2dXRtX2NhbXBhaWduJTNkRnJlZVRyaWFsMjAyM1dXMDgxNkdQU2FkZXh0JTI2Y2FtcGFpZ25pZCUzZDUzMDcwODA5OSUyNmFkZ3JvdXBpZCUzZDEzNTAyMDI3NzQyMTc2OTglMjZhZGlkJTNkJTI2bXNjbGtpZCUzZDgwNjM0Y2YzMGRlMDE3MTAyNjVmMjk1MGYyZTYxNzM1%26rlid%3D80634cf30de01710265f2950f2e61735&vqd=4-266956621161894169747609047079670999441&iurl=%7B1%7DIG%3D4E234A2A70BD47BA8D9437C2FE20102A%26CID%3D0DBF2147217B62AE10B83541204E63FB%26ID%3DDevEx%2C5065.1\",\"text\":\"DataRobot Free Trial\"},{\"snippet\":\"Unlock Your AI Success in 2023 Tips on the Path of Value-Driven AI\",\"targetUrl\":\"https://duckduckgo.com/y.js?ad_domain=datarobot.com&ad_provider=bingv7aa&ad_type=txad&eddgt=TADjssf2xUPiUP4Z%2DpL%2Djw%3D%3D&rut=d527db3345d9139883b504b6a057be69be3347eeb5acd23b8e64f7449bb73b50&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De8UTprN2eQMdFB5SJFU5fAVjVUCUxyH9ey14F3ix7IMGUN9R8j4XI%2DxHFXyG6wW8QyDclA1ah53V6Dl1LRU3JgQHXtprRWsm0zG%2DDqcpZf1i6kJFAmi315DCmvKoT6C3z97QkhAnr4yX%2Dv3glHLhN9uc3yL9wfU5U7nv5YTzG9UKxaD_%2DK2eubvLD7ldJaBI4tjPKQUhliUQqV4yr72OBpGoew774%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuZGF0YXJvYm90LmNvbSUyZnJlc291cmNlcyUyZmFpc3VjY2VzczIwMjMlMmYlM2Z1dG1fbWVkaXVtJTNkc2VhcmNoJTI2dXRtX3NvdXJjZSUzZGJpbmclMjZ1dG1fY2FtcGFpZ24lM2RDb250ZW50MTBLZXlzdG9BSVN1Y2Nlc3MyMDIzV1cwNTIyR1BTYWRleHQlMjZ1dG1fdGVybSUzZGRhdGFyb2JvdCUyNnV0bV9jb250ZW50JTNkYWRfZXh0JTI2Y2FtcGFpZ25pZCUzZDUzMDcwODA5OSUyNmFkZ3JvdXBpZCUzZDEzNTAyMDI3NzQyMTc2OTglMjZhZGlkJTNkJTI2bXNjbGtpZCUzZDRiOWQwYTViOGQ4YTE4YTNkOGI1NmZmODcyOGM1Yzg4%26rlid%3D4b9d0a5b8d8a18a3d8b56ff8728c5c88&vqd=4-186891061025271488176703891649000666566&iurl=%7B1%7DIG%3D4E234A2A70BD47BA8D9437C2FE20102A%26CID%3D0DBF2147217B62AE10B83541204E63FB%26ID%3DDevEx%2C5067.1\",\"text\":\"10 Keys to AI Success\"},{\"snippet\":\"Our Platform Includes Four Fully Integrated Products. Read More.\",\"targetUrl\":\"https://duckduckgo.com/y.js?ad_domain=datarobot.com&ad_provider=bingv7aa&ad_type=txad&eddgt=TADjssf2xUPiUP4Z%2DpL%2Djw%3D%3D&rut=9fb136d3bde9c28bb5c474789d38421625142c03b6cd89976b66ce5517d4b669&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De8ouxsAouiEsi%2DDc9BM8u61DVUCUxad20UaEBujK70scJAQPZlPSbbKQxos2Uiw4fxi21%2DVgvVutJWxTj1GAp5dA40ea3WyEU8c7sfEzgUyRqLe5kCWLFg_dSdKKL5y1cUUcQ8Vz7ZK25elf6NLz9GTXdqOHZ9m5%2D1nK%2DKxXXXJEwnP9Hq6S9AujOSKuU63ixP2CmCTMdq9C64CzxUWU_R17nOAG0%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuZGF0YXJvYm90LmNvbSUyZnByb2R1Y3QlMmYlM2ZjYW1wYWlnbmlkJTNkNTMwNzA4MDk5JTI2YWRncm91cGlkJTNkMTM1MDIwMjc3NDIxNzY5OCUyNmFkaWQlM2QlMjZtc2Nsa2lkJTNkM2E2NzBlMjdhODcxMThhODkzNDQ2Yjg0NTc3Nzk1YmQ%26rlid%3D3a670e27a87118a893446b84577795bd&vqd=4-73641448877716277028608780106696479949&iurl=%7B1%7DIG%3D4E234A2A70BD47BA8D9437C2FE20102A%26CID%3D0DBF2147217B62AE10B83541204E63FB%26ID%3DDevEx%2C5069.1\",\"text\":\"Product Overview\"}],\"tid\":\"6\\t8[7]\\t10[9]\\t12[11]\",\"type\":\"EnhancedSiteLink\"},\"tid\":\"1\"},\"ae\":null,\"c\":\"https://duckduckgo.com/y.js?ad_domain=datarobot.com&ad_provider=bingv7aa&ad_type=txad&eddgt=TADjssf2xUPiUP4Z%2DpL%2Djw%3D%3D&rut=489d897f192221406105931c23952ee9ddfcf9255a24130474c891f263e96e00&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De839_LMH5pa8tva7WIdtsEfDVUCUwdFj2%2D%2DKtKdG6HDyt85Ce7V%2DiiQ2w5qD19CAl57L1dYymA6REaydrRBR2k46ZVmaiPv9HjtdlGliBcpsrqORKeHvMrkxqdZFZpqnPhGXg22zPoUr7K1CebeDBNfuES4v6ILDlFk4%2DMyDHiYvcYYoW2JyhgHVssKxbFEZG7OHwmGw%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuZGF0YXJvYm90LmNvbSUyZmpwJTJmbHAlMmZhaS1mb3ItYnVzaW5lc3MlMmYlM2Z1dG1fbWVkaXVtJTNkc2VhcmNoJTI2dXRtX3NvdXJjZSUzZGJpbmclMjZ1dG1fY2FtcGFpZ24lM2RERU1PMjAyM0FsbFByb2R1Y3RzSlAwNjI2QlBTJTI2dXRtX3Rlcm0lM2RkYXRhcm9ib3QlMjZ1dG1fY29udGVudCUzZERSX2JyYW5kZWRfcnNhJTI2Y2FtcGFpZ25pZCUzZDUzMDcwODA5OSUyNmFkZ3JvdXBpZCUzZDEzNTAyMDI3NzQyMTc2OTglMjZhZGlkJTNkJTI2bXNjbGtpZCUzZDQ3ZTc1ZGU1Y2FjYzFlODRlOTUzMzg0NjM1Y2FjNTc2%26rlid%3D47e75de5cacc1e84e953384635cac576&vqd=4-261134921179772841597192846410308597281&iurl=%7B1%7DIG%3D4E234A2A70BD47BA8D9437C2FE20102A%26CID%3D0DBF2147217B62AE10B83541204E63FB%26ID%3DDevEx%2C5061.1\",\"d\":\"datarobot.com\",\"h\":0,\"i\":\"\",\"k\":0,\"m\":0,\"o\":\"\",\"p\":1,\"relevancy\":{\"abstract\":\"%E9%AB%98%E7%B2%BE%E5%BA%A6%E3%81%AA%E6%A9%9F%E6%A2%B0%E5%AD%A6%E7%BF%92%E3%83%A2%E3%83%87%E3%83%AB%E3%82%92%E6%A7%8B%E7%AF%89%E3%80%81%E5%AE%9F%E8%A3%85%E3%80%81%E9%81%8B%E7%94%A8%E3%80%82%3Cb%3EDataRobot%3C%2Fb%3E%E3%81%AF%E7%A4%BE%E5%86%85%E3%83%87%E3%83%BC%E3%82%BF%E3%81%8B%E3%82%89%E6%96%B0%E3%81%97%E3%81%84%E4%BE%A1%E5%80%A4%E3%82%92%E5%89%B5%E9%80%A0%E3%81%97%E3%81%BE%E3%81%99.%20%3Cb%3EDataRobot%3C%2Fb%3E%E3%81%AF%E7%B0%A1%E5%8D%98%E3%81%AA%E6%93%8D%E4%BD%9C%E3%81%A7%E3%83%87%E3%83%BC%E3%82%BF%E3%81%8B%E3%82%89%E4%BE%A1%E5%80%A4%E3%82%92%E5%89%B5%E5%87%BA\",\"adx_name\":\"none\",\"is_good_v10\":1,\"organic_ranks\":[\"0\",12,15,20,23],\"q\":\"DataRobot%20AI%20platform%20comparison\",\"q_words\":4,\"q_words_fuzzy\":0.25,\"q_words_in_ad\":1,\"root_domain\":\"datarobot.com\",\"start\":\"0\",\"title\":\"%E3%83%93%E3%83%83%E3%82%B0%E3%83%87%E3%83%BC%E3%82%BF%E5%88%86%E6%9E%90%E3%82%92%E9%AB%98%E9%80%9F%E5%8C%96%20%2D%20%E3%83%87%E3%83%BC%E3%82%BF%E3%81%8B%E3%82%89%E6%96%B0%E3%81%97%E3%81%84%E4%BE%A1%E5%80%A4%E3%82%92\"},\"s\":\"bingv7aa\",\"t\":\"\\u30d3\\u30c3\\u30b0\\u30c7\\u30fc\\u30bf\\u5206\\u6790\\u3092\\u9ad8\\u901f\\u5316 - \\u30c7\\u30fc\\u30bf\\u304b\\u3089\\u65b0\\u3057\\u3044\\u4fa1\\u5024\\u3092\",\"tid\":\"1,6,8[7],10[9],12[11]\",\"u\":\"https://duckduckgo.com/y.js?ad_domain=datarobot.com&ad_provider=bingv7aa&ad_type=txad&eddgt=TADjssf2xUPiUP4Z%2DpL%2Djw%3D%3D&rut=489d897f192221406105931c23952ee9ddfcf9255a24130474c891f263e96e00&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De839_LMH5pa8tva7WIdtsEfDVUCUwdFj2%2D%2DKtKdG6HDyt85Ce7V%2DiiQ2w5qD19CAl57L1dYymA6REaydrRBR2k46ZVmaiPv9HjtdlGliBcpsrqORKeHvMrkxqdZFZpqnPhGXg22zPoUr7K1CebeDBNfuES4v6ILDlFk4%2DMyDHiYvcYYoW2JyhgHVssKxbFEZG7OHwmGw%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuZGF0YXJvYm90LmNvbSUyZmpwJTJmbHAlMmZhaS1mb3ItYnVzaW5lc3MlMmYlM2Z1dG1fbWVkaXVtJTNkc2VhcmNoJTI2dXRtX3NvdXJjZSUzZGJpbmclMjZ1dG1fY2FtcGFpZ24lM2RERU1PMjAyM0FsbFByb2R1Y3RzSlAwNjI2QlBTJTI2dXRtX3Rlcm0lM2RkYXRhcm9ib3QlMjZ1dG1fY29udGVudCUzZERSX2JyYW5kZWRfcnNhJTI2Y2FtcGFpZ25pZCUzZDUzMDcwODA5OSUyNmFkZ3JvdXBpZCUzZDEzNTAyMDI3NzQyMTc2OTglMjZhZGlkJTNkJTI2bXNjbGtpZCUzZDQ3ZTc1ZGU1Y2FjYzFlODRlOTUzMzg0NjM1Y2FjNTc2%26rlid%3D47e75de5cacc1e84e953384635cac576&vqd=4-261134921179772841597192846410308597281&iurl=%7B1%7DIG%3D4E234A2A70BD47BA8D9437C2FE20102A%26CID%3D0DBF2147217B62AE10B83541204E63FB%26ID%3DDevEx%2C5061.1\"}], {\"page_load_url\":\"https://duckduckgo.com/y.js?ifu=%7B3%7Dappid%3D055AAD1BA669BEB8B048128DC89A107C678B527B%26rguid%3D0716358df9934510b6d5d49119d2d6d3&iurl=%7B2%7DIG%3D4E234A2A70BD47BA8D9437C2FE20102A%26CID%3D0DBF2147217B62AE10B83541204E63FB%26Type%3DEvent.CPT%26DATA%3D0\",\"visibility_url\":\"https://duckduckgo.com/y.js?ivu=%7B4%7Dtype%3Dmv%26reqver%3D1.0%26rg%3D0716358df9934510b6d5d49119d2d6d3\"});DDG.deep.signalSummary = \"\";DDG.inject('DDG.Data.languages.resultLanguages', {\"en\":[\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/datarobot/product/datarobot-ai-platform\",\"https://www.gartner.com/reviews/market/data-preparation-tools/vendor/datarobot/product/datarobot-ai-platform\",\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/compare/datarobot-vs-h2o-ai\",\"https://www.trustradius.com/compare-products/datarobot-vs-google-cloud-ai\",\"https://www.trustradius.com/compare-products/dataiku-dss-vs-datarobot\",\"https://www.trustradius.com/compare-products/datarobot-vs-h2o\",\"https://www.eweek.com/big-data-and-analytics/c3-ai-vs-datarobot/\",\"https://research.aimultiple.com/automl-comparison/\",\"https://valohai.com/mlops-platforms-compared/\",\"https://www.gartner.com/reviews/market/augmented-data-quality-solutions/vendor/datarobot/product/datarobot-ai-platform\",\"https://internetstack.com/comparison/datarobot/vs/h2o-ai/\",\"https://www.datarobot.com/\",\"https://www.datarevenue.com/en-blog/ml-platforms-dataiku-vs-alteryx-vs-sagemaker\",\"https://solutionsreview.com/business-intelligence/the-best-ai-tools-for-data-science/\",\"https://docs.datarobot.com/en/docs/modeling/analyze-models/other/model-compare.html\",\"https://www.g2.com/products/datarobot/competitors/alternatives\",\"https://openai.com/blog/introducing-chatgpt-team\",\"https://www.datarobot.com/blog/big-data-and-artificial-intelligence-a-quick-comparison/\",\"https://www.gartner.com/reviews/market/data-preparation-tools/vendor/datarobot/product/datarobot-ai-platform/alternatives\",\"https://nvidianews.nvidia.com/news/geforce-rtx-40-super-series\",\"https://www.datarobot.com/blog/2023-a-year-of-innovation-and-impact/\"]});DDG.deep.pageLayoutSummary = \"a1w24r1\";DDG.inject('DDG.Data.languages.adLanguages', {});if (DDG.pageLayout) DDG.pageLayout.load('d',[{\"a\":\"AI APIs AND FRAMEWORKS DATA PLATFORMS Custom Chat APPLICATIONS Compose and Compare Compose and Compare Train and Tune Train and Tune Analyze and Transform Analyze and Transform BUILD BUILD Document and Comply Document and Comply Audit and Approve Audit and Approve Register and Manage Register and Manage GOVERN GOVERN Learn and Optimize Learn and...\",\"ae\":null,\"c\":\"https://www.datarobot.com/platform/\",\"d\":\"www.datarobot.com/platform/\",\"da\":\"\",\"h\":0,\"i\":\"www.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Platform Overview | DataRobot AI Platform\",\"u\":\"https://www.datarobot.com/platform/\"},{\"a\":\"Reviewed in Last 12 Months mail_outline Email Page 4.6 508 Ratings (All Time) Rating Distribution 5 Star 63% 4 Star 33% 3 Star 3% 2 Star 0% 1 Star 0% Distribution based on 508 ratings Customer Experience Evaluation & Contracting 4.5 Integration & Deployment 4.5 Service & Support 4.7 Product Capabilities 4.6 FREE\",\"ae\":null,\"c\":\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/datarobot/product/datarobot-ai-platform\",\"d\":\"www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/datarobot/product/datarobot-ai-platform\",\"da\":\"\",\"h\":0,\"i\":\"www.gartner.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot AI Platform Reviews - Gartner\",\"u\":\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/datarobot/product/datarobot-ai-platform\"},{\"a\":\"4 Star 42% 3 Star 3% 2 Star 0% 1 Star 0% Distribution based on 36 ratings Customer Experience Evaluation & Contracting 4.6 Integration & Deployment 4.4 Service & Support 4.6 Product Capabilities 4.3 FREE View and Download Peer Insights About DataRobot AI Platform\",\"ae\":null,\"c\":\"https://www.gartner.com/reviews/market/data-preparation-tools/vendor/datarobot/product/datarobot-ai-platform\",\"d\":\"www.gartner.com/reviews/market/data-preparation-tools/vendor/datarobot/product/datarobot-ai-platform\",\"da\":\"\",\"h\":0,\"i\":\"www.gartner.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot AI Platform Reviews - Gartner\",\"u\":\"https://www.gartner.com/reviews/market/data-preparation-tools/vendor/datarobot/product/datarobot-ai-platform\"},{\"a\":\"DataRobot vs H2O.ai Based on verified reviews from real users in the Data Science and Machine Learning Platforms market. DataRobot has a rating of 4.6 stars with 508 reviews. H2O.ai has a rating of 4.4 stars with 108 reviews.\",\"ae\":null,\"c\":\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/compare/datarobot-vs-h2o-ai\",\"d\":\"www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/compare/datarobot-vs-h2o-ai\",\"da\":\"\",\"h\":0,\"i\":\"www.gartner.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot vs H2O.ai 2024 | Gartner Peer Insights\",\"u\":\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/compare/datarobot-vs-h2o-ai\"},{\"a\":\"84 Reviews and Ratings Path to AI Success Google Cloud AI 84 Reviews and Ratings Have you used any of these products before? No, I use something else Compare DataRobot vs Google Cloud AI. 168 verified user reviews and ratings of features, pros, cons, pricing, support and more.\",\"ae\":null,\"c\":\"https://www.trustradius.com/compare-products/datarobot-vs-google-cloud-ai\",\"d\":\"www.trustradius.com/compare-products/datarobot-vs-google-cloud-ai\",\"da\":\"\",\"h\":0,\"i\":\"www.trustradius.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot vs Google Cloud AI | TrustRadius\",\"u\":\"https://www.trustradius.com/compare-products/datarobot-vs-google-cloud-ai\"},{\"a\":\"DataRobot 84 Reviews and Ratings Path to AI Success Compare Dataiku DSS vs DataRobot. 103 verified user reviews and ratings of features, pros, cons, pricing, support and more.\",\"ae\":null,\"c\":\"https://www.trustradius.com/compare-products/dataiku-dss-vs-datarobot\",\"d\":\"www.trustradius.com/compare-products/dataiku-dss-vs-datarobot\",\"da\":\"\",\"h\":0,\"i\":\"www.trustradius.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku DSS vs DataRobot | TrustRadius\",\"u\":\"https://www.trustradius.com/compare-products/dataiku-dss-vs-datarobot\"},{\"a\":\"The DataRobot AI Platform is presented as a solution that accelerates and democratizes data science by automating the end-to-end journey from data to value and allows users to deploy AI applications at scale. DataRobot provides a centrally governed platform that gives users AI to drive business outcomes, that is available on the user's cloud ...\",\"ae\":null,\"c\":\"https://www.trustradius.com/compare-products/datarobot-vs-h2o\",\"d\":\"www.trustradius.com/compare-products/datarobot-vs-h2o\",\"da\":\"\",\"h\":0,\"i\":\"www.trustradius.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot vs H2O | TrustRadius\",\"u\":\"https://www.trustradius.com/compare-products/datarobot-vs-h2o\"},{\"a\":\"C3 AI and DataRobot are two of the leading AI cloud platforms. As such, this is a close comparison. Each has an extensive set of artificial intelligence features. Which is best for your...\",\"ae\":null,\"c\":\"https://www.eweek.com/big-data-and-analytics/c3-ai-vs-datarobot/\",\"d\":\"www.eweek.com/big-data-and-analytics/c3-ai-vs-datarobot/\",\"da\":\"\",\"e\":\"2022-12-02T00:00:00.0000000\",\"h\":0,\"i\":\"www.eweek.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"C3 AI vs. DataRobot: Top AI CloudPlatforms | eWEEK\",\"u\":\"https://www.eweek.com/big-data-and-analytics/c3-ai-vs-datarobot/\"},{\"a\":\"Performance: H2O.ai has greater performance measures in classification and regression tasks. Automation: Tazi.ai and DataRobot offer greater automation rates. Popularity: Along with the Google Cloud AutoML platform, H2O.ai is also the most searched autoML vendor. DataRobot, H2O.ai, and Google Cloud AutoML are the leading vendors. However, you ...\",\"ae\":null,\"c\":\"https://research.aimultiple.com/automl-comparison/\",\"d\":\"research.aimultiple.com/automl-comparison/\",\"da\":\"\",\"e\":\"2023-12-14T00:00:00.0000000\",\"h\":0,\"i\":\"research.aimultiple.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"AutoML Tech / Products Comparison & Market Landscape in 2024 - AIMultiple\",\"u\":\"https://research.aimultiple.com/automl-comparison/\"},{\"a\":\"The platforms we've chosen for our analysis are ClearML, cnvrg.io, Dataiku, Datarobot, Iguazio, Sagemaker, Seldon and Valohai from the managed side, and Flyte, Kubeflow, MLflow and Metaflow from the open-source side. This is by no means an exhaustive list of all the MLOps tools out there. Most of these are tools that describe themselves as ...\",\"ae\":null,\"c\":\"https://valohai.com/mlops-platforms-compared/\",\"d\":\"valohai.com/mlops-platforms-compared/\",\"da\":\"\",\"h\":0,\"i\":\"valohai.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MLOps Platforms Compared - Valohai\",\"u\":\"https://valohai.com/mlops-platforms-compared/\"},{\"a\":\"5 Star 33% 4 Star 67% 3 Star 0% 2 Star 0% 1 Star 0% Distribution based on 3 ratings Customer Experience Evaluation & Contracting 4.5 Integration & Deployment 4.7 Service & Support 4.7 Product Capabilities 4.7 FREE View and Download Peer Insights About DataRobot AI Platform\",\"ae\":null,\"c\":\"https://www.gartner.com/reviews/market/augmented-data-quality-solutions/vendor/datarobot/product/datarobot-ai-platform\",\"d\":\"www.gartner.com/reviews/market/augmented-data-quality-solutions/vendor/datarobot/product/datarobot-ai-platform\",\"da\":\"\",\"h\":0,\"i\":\"www.gartner.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot AI Platform Reviews - Gartner\",\"u\":\"https://www.gartner.com/reviews/market/augmented-data-quality-solutions/vendor/datarobot/product/datarobot-ai-platform\"},{\"a\":\"H2O.ai. H2O.ai is an open source platform for machine learning and predictive analytics. It is designed to help businesses and organizations make better decisions by leveraging the power of data. H2O.ai is used by data scientists, engineers, and business analysts to build and deploy machine learning models quickly and easily.\",\"ae\":null,\"c\":\"https://internetstack.com/comparison/datarobot/vs/h2o-ai/\",\"d\":\"internetstack.com/comparison/datarobot/vs/h2o-ai/\",\"da\":\"\",\"h\":0,\"i\":\"internetstack.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot vs H2O.ai - Which is better? (Comparison)\",\"u\":\"https://internetstack.com/comparison/datarobot/vs/h2o-ai/\"},{\"a\":\"75% faster from start to implementation with AI automation 18X greater likelihood to buy for the highest-scored leads See the Story\",\"ae\":null,\"c\":\"https://www.datarobot.com/\",\"d\":\"www.datarobot.com\",\"da\":\"\",\"h\":0,\"i\":\"www.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot AI Platform | Deliver Value from AI\",\"u\":\"https://www.datarobot.com/\"},{\"a\":\"Dataiku vs. Alteryx. Dataiku and Alteryx are both managed machine learning platforms, but Dataiku focuses on the engineering aspects, while Alteryx focuses on analytics and presentation. Dataiku provides Data Science Studio (DSS), a cross-platform desktop application that includes a notebook (similar to Jupyter Notebook) for engineers to write ...\",\"ae\":null,\"c\":\"https://www.datarevenue.com/en-blog/ml-platforms-dataiku-vs-alteryx-vs-sagemaker\",\"d\":\"www.datarevenue.com/en-blog/ml-platforms-dataiku-vs-alteryx-vs-sagemaker\",\"da\":\"\",\"h\":0,\"i\":\"www.datarevenue.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"ML Platforms: Dataiku vs. Alteryx vs. Sagemaker vs. Datarobot\",\"u\":\"https://www.datarevenue.com/en-blog/ml-platforms-dataiku-vs-alteryx-vs-sagemaker\"},{\"a\":\"DataRobot. Platform: DataRobot Enterprise AI Platform Related products: Paxata Data Preparation, Automated Machine Learning, Automated Time Series, MLOps Description: DataRobot offers an enterprise AI platform that automates the end-to-end process for building, deploying, and maintaining AI. The product is powered by open-source algorithms and can be leveraged on-prem, in the cloud or as a ...\",\"ae\":null,\"c\":\"https://solutionsreview.com/business-intelligence/the-best-ai-tools-for-data-science/\",\"d\":\"solutionsreview.com/business-intelligence/the-best-ai-tools-for-data-science/\",\"da\":\"\",\"e\":\"2024-01-11T00:00:00.0000000\",\"h\":0,\"i\":\"solutionsreview.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"The 11 Best AI Tools for Data Science to Consider in 2024\",\"u\":\"https://solutionsreview.com/business-intelligence/the-best-ai-tools-for-data-science/\"},{\"a\":\"Compare models. To compare models in a project with at least two models built, either: Select the Model Comparison tab. Select two models from the Leaderboard and use the Leaderboard menu's Compare Selected option. Once on the page, select models from the dropdown. The associated model statistics update to reflect the currently selected model ...\",\"ae\":null,\"c\":\"https://docs.datarobot.com/en/docs/modeling/analyze-models/other/model-compare.html\",\"d\":\"docs.datarobot.com/en/docs/modeling/analyze-models/other/model-compare.html\",\"da\":\"\",\"e\":\"2022-11-01T00:00:00.0000000\",\"h\":0,\"i\":\"docs.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Model Comparison: DataRobot docs - DataRobot AI Platform\",\"u\":\"https://docs.datarobot.com/en/docs/modeling/analyze-models/other/model-compare.html\"},{\"a\":\"Top DataRobot AI Platform Alternatives (All Time) How alternatives are selected Dataiku MATLAB Alteryx Designer IBM SPSS Statistics RapidMiner Studio Base SAS Anaconda Enterprise Databricks Data Intelligence Platform Considering alternatives to DataRobot AI Platform?\",\"ae\":null,\"c\":\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/datarobot/product/datarobot-ai-platform/alternatives\",\"d\":\"www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/datarobot/product/datarobot-ai-platform/alternatives\",\"da\":\"\",\"h\":0,\"i\":\"www.gartner.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot AI Platform Alternatives - Gartner\",\"u\":\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/datarobot/product/datarobot-ai-platform/alternatives\"},{\"a\":\"Top Alternatives to DataRobot AI Platform MathWorks Matlab Databricks Lakehouse Platform Dataiku TensorFlow TFX Google Cloud Vertex AI Alteryx View All Alternatives Best Alternatives and Competitors to DataRobot AI Platform\",\"ae\":null,\"c\":\"https://www.softwarereviews.com/categories/200/products/6813/alternatives\",\"d\":\"www.softwarereviews.com/categories/200/products/6813/alternatives\",\"da\":\"translations\",\"h\":0,\"i\":\"www.softwarereviews.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot AI Platform Alternatives and Competitors | Machine ...\",\"u\":\"https://www.softwarereviews.com/categories/200/products/6813/alternatives\"},{\"a\":\"#1 Alteryx (458) 4.6 out of 5 Alteryx drives transformational business outcomes through unified analytics, data science, and process automation. Categories in common with DataRobot: Data Science and Machine Learning Platforms Predictive Analytics Try for free Reviewers say compared to DataRobot, Alteryx is: Easier to set up More expensive\",\"ae\":null,\"c\":\"https://www.g2.com/products/datarobot/competitors/alternatives\",\"d\":\"www.g2.com/products/datarobot/competitors/alternatives\",\"da\":\"\",\"h\":0,\"i\":\"www.g2.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Top 10 DataRobot Alternatives & Competitors (Free/Paid) - G2\",\"u\":\"https://www.g2.com/products/datarobot/competitors/alternatives\"},{\"a\":\"Today, we're adding a new self-serve plan: ChatGPT Team. ChatGPT Team offers access to our advanced models like GPT-4 and DALL\\u00b7E 3, and tools like Advanced Data Analysis. It additionally includes a dedicated collaborative workspace for your team and admin tools for team management. As with ChatGPT Enterprise, you own and control your ...\",\"ae\":null,\"c\":\"https://openai.com/blog/introducing-chatgpt-team\",\"d\":\"openai.com/blog/introducing-chatgpt-team\",\"da\":\"\",\"e\":\"2024-01-10T00:00:00.0000000\",\"h\":0,\"i\":\"openai.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Introducing ChatGPT Team - OpenAI\",\"u\":\"https://openai.com/blog/introducing-chatgpt-team\"},{\"a\":\"Big data and artificial intelligence: a quick comparison | DataRobot AI Platform Blog Big data and artificial intelligence: a quick comparison Big data and artificial intelligence: a quick comparison March 3, 2020 by DataRobot \\u00b7 3 min read This article was originally published at Algorithimia's website.\",\"ae\":null,\"c\":\"https://www.datarobot.com/blog/big-data-and-artificial-intelligence-a-quick-comparison/\",\"d\":\"www.datarobot.com/blog/big-data-and-artificial-intelligence-a-quick-comparison/\",\"da\":\"\",\"h\":0,\"i\":\"www.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Big data and artificial intelligence: a quick comparison | DataRobot AI ...\",\"u\":\"https://www.datarobot.com/blog/big-data-and-artificial-intelligence-a-quick-comparison/\"},{\"a\":\"36 Ratings compare_arrows Compare rate_review Write a Review download_2 Download PDF Related markets: DataRobot AI Platform in Data Science and Machine Learning Platforms (508 Reviews), DataRobot AI Platform in Augmented Data Quality Solutions (3 Reviews), DataRobot AI Platform in Predictive Analytics Software (1 Review)\",\"ae\":null,\"c\":\"https://www.gartner.com/reviews/market/data-preparation-tools/vendor/datarobot/product/datarobot-ai-platform/alternatives\",\"d\":\"www.gartner.com/reviews/market/data-preparation-tools/vendor/datarobot/product/datarobot-ai-platform/alternatives\",\"da\":\"\",\"h\":0,\"i\":\"www.gartner.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot AI Platform Alternatives - Gartner\",\"u\":\"https://www.gartner.com/reviews/market/data-preparation-tools/vendor/datarobot/product/datarobot-ai-platform/alternatives\"},{\"a\":\"An AI-Powered Leap in PC Computing The new GeForce RTX SUPER GPUs are the ultimate way to experience AI on PCs. Specialized AI Tensor Cores deliver up to 836 AI TOPS to deliver transformative capabilities for AI in gaming, creating and everyday productivity. The rich software stack built on top of RTX GPUs further accelerates AI.\",\"ae\":null,\"c\":\"https://nvidianews.nvidia.com/news/geforce-rtx-40-super-series\",\"d\":\"nvidianews.nvidia.com/news/geforce-rtx-40-super-series\",\"da\":\"translations\",\"e\":\"2024-01-08T00:00:00.0000000\",\"h\":0,\"i\":\"nvidianews.nvidia.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"GeForce RTX 40 SUPER Series: New Heroes Debut in the Gaming and ...\",\"u\":\"https://nvidianews.nvidia.com/news/geforce-rtx-40-super-series\"},{\"a\":\"We unveiled new enterprise-grade generative AI functionality to close the confidence gap and accelerate adoption, including generative AI application cost and performance monitoring, a unified observability console and registry for governance, multi-provider comparison playground and other enhancements to deliver greater transparency and governa...\",\"ae\":null,\"c\":\"https://www.datarobot.com/blog/2023-a-year-of-innovation-and-impact/\",\"d\":\"www.datarobot.com/blog/2023-a-year-of-innovation-and-impact/\",\"da\":\"translations\",\"e\":\"2023-12-21T00:00:00.0000000\",\"h\":0,\"i\":\"www.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"2023: A Year of Innovation and Impact | DataRobot AI Platform\",\"u\":\"https://www.datarobot.com/blog/2023-a-year-of-innovation-and-impact/\"},{\"n\":\"/d.js?q=DataRobot%20AI%20platform%20comparison&kl=wt-wt&l=wt-wt&p=&s=24&ex=-1&ct=US&sp=0&vqd=4-58620545585474558767320902709740831322\"}]);DDG.duckbar.load('images');DDG.duckbar.load('news');DDG.duckbar.load('videos');DDG.duckbar.loadModule('related_searches', {\"ads\":[],\"query\":\"DataRobot AI platform comparison\",\"queryEncoded\":\"DataRobot%20AI%20platform%20comparison\",\"response_type\":\"places\",\"results\":[{\"display_text\":\"data robot ai platform\",\"text\":\"data robot ai platform\",\"web_search_url\":\"?q=data%20robot%20ai%20platform\"},{\"display_text\":\"datarobot ai partners\",\"text\":\"datarobot ai partners\",\"web_search_url\":\"?q=datarobot%20ai%20partners\"},{\"display_text\":\"data robot platforms\",\"text\":\"data robot platforms\",\"web_search_url\":\"?q=data%20robot%20platforms\"},{\"display_text\":\"datarobot partner portal\",\"text\":\"datarobot partner portal\",\"web_search_url\":\"?q=datarobot%20partner%20portal\"},{\"display_text\":\"data robot partners\",\"text\":\"data robot partners\",\"web_search_url\":\"?q=data%20robot%20partners\"},{\"display_text\":\"data robot data warehouse\",\"text\":\"data robot data warehouse\",\"web_search_url\":\"?q=data%20robot%20data%20warehouse\"}],\"vqd\":{\"DataRobot%20AI%20platform%20comparison\":\"4-58620545585474558767320902709740831322\"}});if (DDG.pageLayout) DDG.pageLayout.initialize({\"mainline\":{\"items\":[[\"ad\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"related_searches\"]]}}, { start: 0 });DDG.deep.emit(\"load:completed\");", + "curl-cffi-POST-https://duckduckgo.com-{\"data\": {\"q\": \"Dataiku vs DataRobot features\"}}": "Dataiku vs DataRobot features at DuckDuckGo
", + "curl-cffi-GET-https://links.duckduckgo.com/d.js-{\"params\": {\"bing_market\": \"wt-WT\", \"df\": null, \"ex\": \"-1\", \"kl\": \"wt-wt\", \"l\": \"wt-wt\", \"q\": \"Dataiku vs DataRobot features\", \"s\": \"0\", \"sp\": \"0\", \"vqd\": \"4-334935250614046875026454141242803242982\"}}": "if (DDG.deep && DDG.deep.setUpstream) DDG.deep.setUpstream(\"bingv7aa\");DDG.deep.bn={'ivc':1};if (DDG.pageLayout) DDG.pageLayout.load('a',[{\"a\":\"\\u9ad8\\u7cbe\\u5ea6\\u306a\\u6a5f\\u68b0\\u5b66\\u7fd2\\u30e2\\u30c7\\u30eb\\u3092\\u69cb\\u7bc9\\u3001\\u5b9f\\u88c5\\u3001\\u904b\\u7528\\u3002DataRobot\\u306f\\u793e\\u5185\\u30c7\\u30fc\\u30bf\\u304b\\u3089\\u65b0\\u3057\\u3044\\u4fa1\\u5024\\u3092\\u5275\\u9020\\u3057\\u307e\\u3059. DataRobot\\u306f\\u4f01\\u696d\\u306e\\u8ab2\\u984c\\u89e3\\u6c7a\\u306b\\u7279\\u5316\\u3002\\u610f\\u601d\\u6c7a\\u5b9a\\u306e\\u81ea\\u52d5\\u5316\\u304b\\u3089\\u9700\\u8981\\u4e88\\u6e2c\\u3001\\u8981\\u56e0\\u5206\\u6790\\u307e\\u3067\\u3053\\u306a\\u3059AI\\u30c4\\u30fc\\u30eb\",\"adext\":{\"callout\":{\"t\":\"Data Science Guardrails \\u00b7 Applied AI Expertise \\u00b7 Trusted by Fortune 50\",\"tid\":\"6\"},\"filterlinks\":{\"l\":[],\"tid\":\"\"},\"sitelinks\":{\"l\":[{\"snippet\":\"Explore the DataRobot AI Platform Get Started With a 30-Day Trial\",\"targetUrl\":\"https://duckduckgo.com/y.js?ad_domain=datarobot.com&ad_provider=bingv7aa&ad_type=txad&eddgt=uchwI3Eul8XsE%2DSUlPqxXg%3D%3D&rut=2381550f96a087800d427735905717264a1a708643136f2736a970e740068621&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De8BGV0WifLHqlNArHdJt3WDTVUCUzDyrVI_ULomBTgn_xk1MKGRFElGY7vQ8fpE4l__S3CnH6%2D2cXlBQayeIz9CbLU7C4XEu8BgG6oZNQ6EtjG6vrYe5hjw1GZN7VBIkj6nn%2DsoUXy14mVbvkM5ojXVf8oeoz8pwdOc4ANH2TiL9vqJe6Lud2IZXvxJf1I%2DA935XcPQobPZKQaFNFMyygI3Y4TW8k%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuZGF0YXJvYm90LmNvbSUyZnRyaWFsJTJmJTNmdXRtX21lZGl1bSUzZHNlYXJjaCUyNnV0bV9zb3VyY2UlM2RiaW5nJTI2dXRtX2NhbXBhaWduJTNkRnJlZVRyaWFsMjAyM1dXMDgxNkdQU2FkZXh0JTI2Y2FtcGFpZ25pZCUzZDUzMDcwODA5OSUyNmFkZ3JvdXBpZCUzZDEzNTAyMDI3NzQyMTc2OTglMjZhZGlkJTNkJTI2bXNjbGtpZCUzZDFmMzU0ODE0ODNmMTEyM2Y5NGMzMmRiNzdjZjk5OWFm%26rlid%3D1f35481483f1123f94c32db77cf999af&vqd=4-25671318592048362755712261648304518289&iurl=%7B1%7DIG%3D3EB403B8C4EA42F4B7FF0CE90CB46EF0%26CID%3D2F20CB6F269D6DD02331DF69279D6C12%26ID%3DDevEx%2C5064.1\",\"text\":\"DataRobot Free Trial\"},{\"snippet\":\"Unlock Your AI Success in 2023 Tips on the Path of Value-Driven AI\",\"targetUrl\":\"https://duckduckgo.com/y.js?ad_domain=datarobot.com&ad_provider=bingv7aa&ad_type=txad&eddgt=uchwI3Eul8XsE%2DSUlPqxXg%3D%3D&rut=08def2477dd7311fbcffe4c409d28fcdbe68925a50cd2894a7502f8a11785352&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De8lYdQlfjG0%2Dh77MMyzT0CuDVUCUyAuuZDH6K8NWyD2XSLoABvrUNVChVbIVOVgzl4xdT3EEUvHgd9P_FWLUDT2My42qKUP3iV87B7hLXXHLdGf7yjst8tWjp%2DcaQz3uiI0c5oom%2DRo8D7A4nohZAtS9199RQLYbNcbOpJnrNMCFmz6EiWk7JqMQ9DE1t9AjaMUWEkEV%2D3W2e8XmBq5bKtRsWnT0E%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuZGF0YXJvYm90LmNvbSUyZnJlc291cmNlcyUyZmFpc3VjY2VzczIwMjMlMmYlM2Z1dG1fbWVkaXVtJTNkc2VhcmNoJTI2dXRtX3NvdXJjZSUzZGJpbmclMjZ1dG1fY2FtcGFpZ24lM2RDb250ZW50MTBLZXlzdG9BSVN1Y2Nlc3MyMDIzV1cwNTIyR1BTYWRleHQlMjZ1dG1fdGVybSUzZGRhdGFyb2JvdCUyNnV0bV9jb250ZW50JTNkYWRfZXh0JTI2Y2FtcGFpZ25pZCUzZDUzMDcwODA5OSUyNmFkZ3JvdXBpZCUzZDEzNTAyMDI3NzQyMTc2OTglMjZhZGlkJTNkJTI2bXNjbGtpZCUzZDI5Zjc0NWY1MzNiNzE2NDU5ZGY0MjA1NmNjYmYyYWU0%26rlid%3D29f745f533b716459df42056ccbf2ae4&vqd=4-333465595216651803104351585568313334233&iurl=%7B1%7DIG%3D3EB403B8C4EA42F4B7FF0CE90CB46EF0%26CID%3D2F20CB6F269D6DD02331DF69279D6C12%26ID%3DDevEx%2C5066.1\",\"text\":\"10 Keys to AI Success\"},{\"snippet\":\"Our Platform Includes Four Fully Integrated Products. Read More.\",\"targetUrl\":\"https://duckduckgo.com/y.js?ad_domain=datarobot.com&ad_provider=bingv7aa&ad_type=txad&eddgt=uchwI3Eul8XsE%2DSUlPqxXg%3D%3D&rut=fbe7591a97a4b400635f8cfafd71893553c70fc90218355b7d5622310d9567db&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De8cB2vIW6%2D5rxeC5vl08jFZjVUCUw2oN7vfXdo8rlxVmZIfw2bF94_ya9lvPQwUYXJFtTGXBslf_XCcVTiFtj2KJzp9yzLPOdWafvxxwBzn2iwextOSL%2Daq20iQ8nZNktMLYBD1xp3WjThLdejbBCFrR_RvD1YZcHcKf5y5auyV04F_V6x_D6nUwdRYFDmdyciLcpT7JO12EZkmM%2D1buahlzuiBmw%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuZGF0YXJvYm90LmNvbSUyZnByb2R1Y3QlMmYlM2ZjYW1wYWlnbmlkJTNkNTMwNzA4MDk5JTI2YWRncm91cGlkJTNkMTM1MDIwMjc3NDIxNzY5OCUyNmFkaWQlM2QlMjZtc2Nsa2lkJTNkMGZhOTg4ZjJkYWU2MWE3MGJhOTVlZDUxMjVlZWFlNDA%26rlid%3D0fa988f2dae61a70ba95ed5125eeae40&vqd=4-211419575679328898707892660118042825990&iurl=%7B1%7DIG%3D3EB403B8C4EA42F4B7FF0CE90CB46EF0%26CID%3D2F20CB6F269D6DD02331DF69279D6C12%26ID%3DDevEx%2C5068.1\",\"text\":\"Product Overview\"}],\"tid\":\"7\\t9[8]\\t11[10]\\t13[12]\",\"type\":\"EnhancedSiteLink\"},\"tid\":\"1\"},\"ae\":{\"callout\":[\"Data Science Guardrails \\u00b7 Applied AI Expertise \\u00b7 Trusted by Fortune 50\"]},\"c\":\"https://duckduckgo.com/y.js?ad_domain=datarobot.com&ad_provider=bingv7aa&ad_type=txad&eddgt=uchwI3Eul8XsE%2DSUlPqxXg%3D%3D&rut=94a279ed1549c0107c5c13f21161fd5aaa0d3f08d19e7afd2ed4a19463b69d7d&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De8XX6qufLbIkEZRIFo_zgmlDVUCUwOnCSpTtxK0dn2QInSfOGU5eU24GjiRwhmSr89Qa92PcEtK2h6KVoghC%2DNwNrkANG4L6sVirCfv5kl7GPWO9gqgcdw8x5ELjGH7N2HWgbdtH%2D7TWKtxZVdVIFwYJUQDUgM_ODwTspzwBbKKLHD4EPAO5U3RDO3R_igFUlsxkeFXA%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuZGF0YXJvYm90LmNvbSUyZmpwJTJmbHAlMmZhaS1mb3ItYnVzaW5lc3MlMmYlM2Z1dG1fbWVkaXVtJTNkc2VhcmNoJTI2dXRtX3NvdXJjZSUzZGJpbmclMjZ1dG1fY2FtcGFpZ24lM2RERU1PMjAyM0FsbFByb2R1Y3RzSlAwNjI2QlBTJTI2dXRtX3Rlcm0lM2RkYXRhcm9ib3QlMjZ1dG1fY29udGVudCUzZERSX2JyYW5kZWRfcnNhJTI2Y2FtcGFpZ25pZCUzZDUzMDcwODA5OSUyNmFkZ3JvdXBpZCUzZDEzNTAyMDI3NzQyMTc2OTglMjZhZGlkJTNkJTI2bXNjbGtpZCUzZGQxMGY4ZjY4ZDYxZjFiOTg2NTc1ZWFjYjI5MTczYTQ1%26rlid%3Dd10f8f68d61f1b986575eacb29173a45&vqd=4-152568096679810917558416500867559274982&iurl=%7B1%7DIG%3D3EB403B8C4EA42F4B7FF0CE90CB46EF0%26CID%3D2F20CB6F269D6DD02331DF69279D6C12%26ID%3DDevEx%2C5059.1\",\"d\":\"datarobot.com\",\"h\":0,\"i\":\"\",\"k\":0,\"m\":0,\"o\":\"\",\"p\":1,\"relevancy\":{\"abstract\":\"%E9%AB%98%E7%B2%BE%E5%BA%A6%E3%81%AA%E6%A9%9F%E6%A2%B0%E5%AD%A6%E7%BF%92%E3%83%A2%E3%83%87%E3%83%AB%E3%82%92%E6%A7%8B%E7%AF%89%E3%80%81%E5%AE%9F%E8%A3%85%E3%80%81%E9%81%8B%E7%94%A8%E3%80%82%3Cb%3EDataRobot%3C%2Fb%3E%E3%81%AF%E7%A4%BE%E5%86%85%E3%83%87%E3%83%BC%E3%82%BF%E3%81%8B%E3%82%89%E6%96%B0%E3%81%97%E3%81%84%E4%BE%A1%E5%80%A4%E3%82%92%E5%89%B5%E9%80%A0%E3%81%97%E3%81%BE%E3%81%99.%20DataRobot%E3%81%AF%E4%BC%81%E6%A5%AD%E3%81%AE%E8%AA%B2%E9%A1%8C%E8%A7%A3%E6%B1%BA%E3%81%AB%E7%89%B9%E5%8C%96%E3%80%82%E6%84%8F%E6%80%9D%E6%B1%BA%E5%AE%9A%E3%81%AE%E8%87%AA%E5%8B%95%E5%8C%96%E3%81%8B%E3%82%89%E9%9C%80%E8%A6%81%E4%BA%88%E6%B8%AC%E3%80%81%E8%A6%81%E5%9B%A0%E5%88%86%E6%9E%90%E3%81%BE%E3%81%A7%E3%81%93%E3%81%AA%E3%81%99AI%E3%83%84%E3%83%BC%E3%83%AB\",\"adx_name\":\"none\",\"is_good_v10\":0,\"q\":\"Dataiku%20vs%20DataRobot%20features\",\"q_words\":4,\"q_words_fuzzy\":0.25,\"q_words_in_ad\":1,\"root_domain\":\"datarobot.com\",\"start\":\"0\",\"title\":\"%E3%83%87%E3%83%BC%E3%82%BF%E3%81%8B%E3%82%89%E6%96%B0%E3%81%97%E3%81%84%E4%BE%A1%E5%80%A4%E3%82%92%20%2D%20%E7%A4%BE%E5%86%85%E3%83%87%E3%83%BC%E3%82%BF%E3%81%8B%E3%82%89%E4%BE%A1%E5%80%A4%E5%89%B5%E5%87%BA\"},\"s\":\"bingv7aa\",\"t\":\"\\u30c7\\u30fc\\u30bf\\u304b\\u3089\\u65b0\\u3057\\u3044\\u4fa1\\u5024\\u3092 - \\u793e\\u5185\\u30c7\\u30fc\\u30bf\\u304b\\u3089\\u4fa1\\u5024\\u5275\\u51fa\",\"tid\":\"1,6,7,9[8],11[10],13[12]\",\"u\":\"https://duckduckgo.com/y.js?ad_domain=datarobot.com&ad_provider=bingv7aa&ad_type=txad&eddgt=uchwI3Eul8XsE%2DSUlPqxXg%3D%3D&rut=94a279ed1549c0107c5c13f21161fd5aaa0d3f08d19e7afd2ed4a19463b69d7d&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De8XX6qufLbIkEZRIFo_zgmlDVUCUwOnCSpTtxK0dn2QInSfOGU5eU24GjiRwhmSr89Qa92PcEtK2h6KVoghC%2DNwNrkANG4L6sVirCfv5kl7GPWO9gqgcdw8x5ELjGH7N2HWgbdtH%2D7TWKtxZVdVIFwYJUQDUgM_ODwTspzwBbKKLHD4EPAO5U3RDO3R_igFUlsxkeFXA%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuZGF0YXJvYm90LmNvbSUyZmpwJTJmbHAlMmZhaS1mb3ItYnVzaW5lc3MlMmYlM2Z1dG1fbWVkaXVtJTNkc2VhcmNoJTI2dXRtX3NvdXJjZSUzZGJpbmclMjZ1dG1fY2FtcGFpZ24lM2RERU1PMjAyM0FsbFByb2R1Y3RzSlAwNjI2QlBTJTI2dXRtX3Rlcm0lM2RkYXRhcm9ib3QlMjZ1dG1fY29udGVudCUzZERSX2JyYW5kZWRfcnNhJTI2Y2FtcGFpZ25pZCUzZDUzMDcwODA5OSUyNmFkZ3JvdXBpZCUzZDEzNTAyMDI3NzQyMTc2OTglMjZhZGlkJTNkJTI2bXNjbGtpZCUzZGQxMGY4ZjY4ZDYxZjFiOTg2NTc1ZWFjYjI5MTczYTQ1%26rlid%3Dd10f8f68d61f1b986575eacb29173a45&vqd=4-152568096679810917558416500867559274982&iurl=%7B1%7DIG%3D3EB403B8C4EA42F4B7FF0CE90CB46EF0%26CID%3D2F20CB6F269D6DD02331DF69279D6C12%26ID%3DDevEx%2C5059.1\"}], {\"page_load_url\":\"https://duckduckgo.com/y.js?ifu=%7B3%7Dappid%3D055AAD1BA669BEB8B048128DC89A107C678B527B%26rguid%3D280881b97b9245e6a74bddebc1a6cbda&iurl=%7B2%7DIG%3D3EB403B8C4EA42F4B7FF0CE90CB46EF0%26CID%3D2F20CB6F269D6DD02331DF69279D6C12%26Type%3DEvent.CPT%26DATA%3D0\",\"visibility_url\":\"https://duckduckgo.com/y.js?ivu=%7B4%7Dtype%3Dmv%26reqver%3D1.0%26rg%3D280881b97b9245e6a74bddebc1a6cbda\"});DDG.deep.signalSummary = \"\";DDG.inject('DDG.Data.languages.resultLanguages', {\"en\":[\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/compare/dataiku-vs-datarobot\",\"https://www.trustradius.com/compare-products/dataiku-dss-vs-datarobot\",\"https://community.dataiku.com/t5/General-Discussion/Dataiku-vs-DataRobot/m-p/9315\",\"https://www.g2.com/compare/datarobot-vs-dataiku-dss\",\"https://www.datarevenue.com/en-blog/ml-platforms-dataiku-vs-alteryx-vs-sagemaker\",\"https://comparisons.financesonline.com/datarobot-vs-dataiku-dss\",\"https://slashdot.org/software/comparison/DataRobot-vs-Dataiku-DSS/\",\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/dataiku/product/dataiku\",\"https://www.gartner.com/reviews/market/dsml-engineering-platforms/compare/dataiku-vs-datarobot\",\"https://www.linkedin.com/pulse/managed-machine-learning-platforms-comparative-analysis/\",\"https://www.trustradius.com/products/datarobot/reviews?qs=pros-and-cons\",\"https://www.getapp.com/emerging-technology-software/a/dataiku-dss/compare/datarobot/\",\"https://www.softwarereviews.com/categories/machine-learning-platforms/compare/dataiku-vs-datarobot-ai-platform\",\"https://slashdot.org/software/comparison/Alteryx-vs-DataRobot-vs-Dataiku-DSS/\",\"https://slashdot.org/software/comparison/DataRobot-vs-Databricks-vs-Dataiku-DSS/\",\"https://valohai.com/mlops-platforms-compared/\",\"https://www.dataiku.com/product/plans-and-features/\",\"https://slashdot.org/software/comparison/DataRobot-vs-Databricks-vs-Dataiku-DSS-vs-datagym/\",\"https://sourceforge.net/software/compare/C3-AI-Suite-vs-DataRobot-vs-Dataiku-DSS/\",\"https://www.softwarereviews.com/categories/machine-learning-platforms/compare/datarobot-ai-platform-vs-dataiku\",\"https://slashdot.org/software/comparison/Amazon-SageMaker-vs-DataRobot-vs-Dataiku-DSS/\",\"https://sourceforge.net/software/compare/Analance-vs-DataRobot-vs-Dataiku-DSS/\"]});DDG.deep.pageLayoutSummary = \"a1w23r1,e1\";DDG.inject('DDG.Data.languages.adLanguages', {});if (DDG.pageLayout) DDG.pageLayout.load('d',[{\"a\":\"1 Star 0% Ratings breakdown Overall Capability Score Overall Rating 4.7 ( 504 reviews) 4.7 (20) Data Access and Manipulation 4.5 (224) Data Exploration and Visualization 4.7\",\"ae\":null,\"c\":\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/compare/dataiku-vs-datarobot\",\"d\":\"www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/compare/dataiku-vs-datarobot\",\"da\":\"\",\"h\":0,\"i\":\"www.gartner.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku vs DataRobot 2024 | Gartner Peer Insights\",\"u\":\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/compare/dataiku-vs-datarobot\"},{\"a\":\"Path to AI Success Compare Dataiku DSS vs DataRobot. 103 verified user reviews and ratings of features, pros, cons, pricing, support and more.\",\"ae\":null,\"c\":\"https://www.trustradius.com/compare-products/dataiku-dss-vs-datarobot\",\"d\":\"www.trustradius.com/compare-products/dataiku-dss-vs-datarobot\",\"da\":\"\",\"h\":0,\"i\":\"www.trustradius.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku DSS vs DataRobot | TrustRadius\",\"u\":\"https://www.trustradius.com/compare-products/dataiku-dss-vs-datarobot\"},{\"a\":\"General Discussion Dataiku vs DataRobot Solved! Raja Level 2 08-22-2020 03:16 AM Please enlighten me, What distinguishes Dataiku from tools like DataRobot? They appear to be similar, trying to know how dataiku has an upper hand, would make it easy for placing option to customers. 1 Reply 2 Solutions Solutions shown first - Read whole discussion\",\"ae\":null,\"c\":\"https://community.dataiku.com/t5/General-Discussion/Dataiku-vs-DataRobot/m-p/9315\",\"d\":\"community.dataiku.com/t5/General-Discussion/Dataiku-vs-DataRobot/m-p/9315\",\"da\":\"\",\"h\":0,\"i\":\"community.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Solved: Dataiku vs DataRobot - Dataiku Community\",\"u\":\"https://community.dataiku.com/t5/General-Discussion/Dataiku-vs-DataRobot/m-p/9315\"},{\"a\":\"DataRobot vs Dataiku DSS When assessing the two solutions, reviewers found Dataiku DSS easier to use and administer. However, reviewers preferred the ease of set up, and doing business with DataRobot overall. Reviewers felt that DataRobot meets the needs of their business better than Dataiku DSS.\",\"ae\":null,\"c\":\"https://www.g2.com/compare/datarobot-vs-dataiku-dss\",\"d\":\"www.g2.com/compare/datarobot-vs-dataiku-dss\",\"da\":\"\",\"h\":0,\"i\":\"www.g2.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Compare DataRobot vs. Dataiku DSS | G2\",\"u\":\"https://www.g2.com/compare/datarobot-vs-dataiku-dss\"},{\"a\":\"Quick overview Before we get into a detailed comparison, here's a quick overview of each platform. Dataiku is a cross-platform desktop application that includes a broad range of tools, such as notebooks (similar to Jupyter Notebook), workflow management (similar to Apache Airflow), and automated machine learning.\",\"ae\":null,\"c\":\"https://www.datarevenue.com/en-blog/ml-platforms-dataiku-vs-alteryx-vs-sagemaker\",\"d\":\"www.datarevenue.com/en-blog/ml-platforms-dataiku-vs-alteryx-vs-sagemaker\",\"da\":\"\",\"h\":0,\"i\":\"www.datarevenue.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"ML Platforms: Dataiku vs. Alteryx vs. Sagemaker vs. Datarobot\",\"u\":\"https://www.datarevenue.com/en-blog/ml-platforms-dataiku-vs-alteryx-vs-sagemaker\"},{\"a\":\"Home Predictive Analysis Software DataRobot Dataiku DSS Why is FinancesOnline free Compare DataRobot vs Dataiku DSS What is better DataRobot or Dataiku DSS? Examining products to find the best Predictive Analysis Software does not always have to be tough.\",\"ae\":null,\"c\":\"https://comparisons.financesonline.com/datarobot-vs-dataiku-dss\",\"d\":\"comparisons.financesonline.com/datarobot-vs-dataiku-dss\",\"da\":\"\",\"h\":0,\"i\":\"comparisons.financesonline.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot vs Dataiku DSS 2024 Comparison | FinancesOnline\",\"u\":\"https://comparisons.financesonline.com/datarobot-vs-dataiku-dss\"},{\"a\":\"Machine Learning Software Dataiku vs DataRobot Dataiku vs DataRobot Share How Capterra Verifies Reviews Pricing Best for Screenshots Features Reviews Pros & Cons Deployment & Support Alternatives Company Details Dataiku VISIT PROFILE DataRobot VISIT PROFILE Pricing Starting from $ 0.01 /Year Pricing Model: Not provided by vendor Free Trial\",\"ae\":null,\"c\":\"https://www.capterra.com/machine-learning-software/compare/142192-179303/Data-Science-Studio-DSS-vs-DataRobot\",\"d\":\"www.capterra.com/machine-learning-software/compare/142192-179303/Data-Science-Studio-DSS-vs-DataRobot\",\"da\":\"translations\",\"h\":0,\"i\":\"www.capterra.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Compare Dataiku vs DataRobot 2024 | Capterra\",\"u\":\"https://www.capterra.com/machine-learning-software/compare/142192-179303/Data-Science-Studio-DSS-vs-DataRobot\"},{\"a\":\"What's the difference between DataRobot and Dataiku DSS? Compare DataRobot vs. Dataiku DSS in 2023 by cost, reviews, features, integrations, deployment, target market, support options, trial offers, training options, years in business, region, and more using the chart below.\",\"ae\":null,\"b\":\"/.\\tSlashdot\\tslashdot.org\",\"c\":\"https://slashdot.org/software/comparison/DataRobot-vs-Dataiku-DSS/\",\"d\":\"slashdot.org/software/comparison/DataRobot-vs-Dataiku-DSS/\",\"da\":\"\",\"h\":0,\"i\":\"slashdot.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Compare DataRobot vs. Dataiku DSS in 2023 - Slashdot\",\"u\":\"https://slashdot.org/software/comparison/DataRobot-vs-Dataiku-DSS/\"},{\"a\":\"1 Star 0% Distribution based on 504 ratings Customer Experience Evaluation & Contracting 4.6 Integration & Deployment 4.7 Service & Support 4.8 Product Capabilities 4.8 FREE View and Download Peer Insights About Dataiku\",\"ae\":null,\"c\":\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/dataiku/product/dataiku\",\"d\":\"www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/dataiku/product/dataiku\",\"da\":\"\",\"h\":0,\"i\":\"www.gartner.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku Reviews, Ratings & Features 2024 | Gartner Peer Insights\",\"u\":\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/dataiku/product/dataiku\"},{\"a\":\"1329 reviews on 16 vendors. chevron_right. Yard Management. 25 reviews on 28 vendors. chevron_right. Zero Trust Network Access. 733 reviews on 47 vendors. chevron_right. Read the latest Gartner-verified reviews covering over 500+ software categories and find the best enterprise software or services for your organization.\",\"ae\":null,\"c\":\"https://www.gartner.com/reviews/market/dsml-engineering-platforms/compare/dataiku-vs-datarobot\",\"d\":\"www.gartner.com/reviews/market/dsml-engineering-platforms/compare/dataiku-vs-datarobot\",\"da\":\"\",\"h\":0,\"i\":\"www.gartner.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Explore Enterprise Software Categories | Gartner Peer Insights\",\"u\":\"https://www.gartner.com/reviews/market/dsml-engineering-platforms/compare/dataiku-vs-datarobot\"},{\"a\":\"1. Dataiku is a versatile desktop application comprised of a wide range of tools, including automated machine learning, notebooks, and workflow management. It aims to replace pre-existing tools...\",\"ae\":null,\"b\":\"li\\tLinkedIn\\twww.linkedin.com\",\"c\":\"https://www.linkedin.com/pulse/managed-machine-learning-platforms-comparative-analysis/\",\"d\":\"www.linkedin.com/pulse/managed-machine-learning-platforms-comparative-analysis/\",\"da\":\"\",\"e\":\"2023-08-11T00:00:00.0000000\",\"h\":0,\"i\":\"www.linkedin.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Managed Machine Learning Platforms: A Comparative Analysis - LinkedIn\",\"u\":\"https://www.linkedin.com/pulse/managed-machine-learning-platforms-comparative-analysis/\"},{\"a\":\"Dataiku DSS, H2O, and Google Cloud AI are common alternatives for DataRobot. What is DataRobot's best feature? Reviewers rate Automated Machine Learning highest, with a score of 9.3. Who uses DataRobot? The most common users of DataRobot are from Mid-sized Companies (51-1,000 employees).\",\"ae\":null,\"c\":\"https://www.trustradius.com/products/datarobot/reviews?qs=pros-and-cons\",\"d\":\"www.trustradius.com/products/datarobot/reviews?qs=pros-and-cons\",\"da\":\"\",\"h\":0,\"i\":\"www.trustradius.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Pros and Cons of DataRobot 2024 - TrustRadius\",\"u\":\"https://www.trustradius.com/products/datarobot/reviews?qs=pros-and-cons\"},{\"a\":\"Compare Dataiku and DataRobot based on features, pricing, verified reviews, integrations & more. Find out which software is best for your business today. 0. App comparison. Add up to 4 apps below to see how they compare. You can also use the "Compare" buttons while browsing.\",\"ae\":null,\"c\":\"https://www.getapp.com/emerging-technology-software/a/dataiku-dss/compare/datarobot/\",\"d\":\"www.getapp.com/emerging-technology-software/a/dataiku-dss/compare/datarobot/\",\"da\":\"\",\"h\":0,\"i\":\"www.getapp.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku vs DataRobot Comparison | GetApp\",\"u\":\"https://www.getapp.com/emerging-technology-software/a/dataiku-dss/compare/datarobot/\"},{\"a\":\"Dataiku vs DataRobot AI Platform Compare Dataiku and DataRobot AI Platform using real user data focused on features, satisfaction, business value, and the vendor relationship. What is Machine Learning Platforms (ML) Software?\",\"ae\":null,\"c\":\"https://www.softwarereviews.com/categories/machine-learning-platforms/compare/dataiku-vs-datarobot-ai-platform\",\"d\":\"www.softwarereviews.com/categories/machine-learning-platforms/compare/dataiku-vs-datarobot-ai-platform\",\"da\":\"\",\"h\":0,\"i\":\"www.softwarereviews.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku vs DataRobot AI Platform - Machine Learning Platforms\",\"u\":\"https://www.softwarereviews.com/categories/machine-learning-platforms/compare/dataiku-vs-datarobot-ai-platform\"},{\"a\":\"What's the difference between Alteryx, DataRobot, and Dataiku DSS? Compare Alteryx vs. DataRobot vs. Dataiku DSS in 2024 by cost, reviews, features, integrations, deployment, target market, support options, trial offers, training options, years in business, region, and more using the chart below. Alteryx View Product DataRobot View Product\",\"ae\":null,\"b\":\"/.\\tSlashdot\\tslashdot.org\",\"c\":\"https://slashdot.org/software/comparison/Alteryx-vs-DataRobot-vs-Dataiku-DSS/\",\"d\":\"slashdot.org/software/comparison/Alteryx-vs-DataRobot-vs-Dataiku-DSS/\",\"da\":\"\",\"h\":0,\"i\":\"slashdot.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Compare Alteryx vs. DataRobot vs. Dataiku DSS in 2023 - Slashdot\",\"u\":\"https://slashdot.org/software/comparison/Alteryx-vs-DataRobot-vs-Dataiku-DSS/\"},{\"a\":\"What's the difference between DataRobot, Databricks Lakehouse, and Dataiku DSS? Compare DataRobot vs. Databricks Lakehouse vs. Dataiku DSS in 2023 by cost, reviews, features, integrations, deployment, target market, support options, trial offers, training options, years in business, region, and more using the chart below.\",\"ae\":null,\"b\":\"/.\\tSlashdot\\tslashdot.org\",\"c\":\"https://slashdot.org/software/comparison/DataRobot-vs-Databricks-vs-Dataiku-DSS/\",\"d\":\"slashdot.org/software/comparison/DataRobot-vs-Databricks-vs-Dataiku-DSS/\",\"da\":\"\",\"h\":0,\"i\":\"slashdot.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Compare DataRobot vs. Databricks Lakehouse vs. Dataiku DSS - Slashdot\",\"u\":\"https://slashdot.org/software/comparison/DataRobot-vs-Databricks-vs-Dataiku-DSS/\"},{\"a\":\"The platforms we've chosen for our analysis are ClearML, cnvrg.io, Dataiku, Datarobot, Iguazio, Sagemaker, Seldon and Valohai from the managed side, and Flyte, Kubeflow, MLflow and Metaflow from the open-source side. This is by no means an exhaustive list of all the MLOps tools out there. Most of these are tools that describe themselves as ...\",\"ae\":null,\"c\":\"https://valohai.com/mlops-platforms-compared/\",\"d\":\"valohai.com/mlops-platforms-compared/\",\"da\":\"\",\"h\":0,\"i\":\"valohai.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MLOps Platforms Compared - Valohai\",\"u\":\"https://valohai.com/mlops-platforms-compared/\"},{\"a\":\"Visual Machine Learning and automated features preprocessing: Builtin charts and dashboards: Code notebooks and recipes: Custom web applications and plugins: Collaboration: DEPLOYMENT OPTIONS; ... Dataiku Scores an overall 4.8 out of 5 rating Based on 249 ratings for the DSMLP market, as of March 1, 2022\",\"ae\":null,\"c\":\"https://www.dataiku.com/product/plans-and-features/\",\"d\":\"www.dataiku.com/product/plans-and-features/\",\"da\":\"\",\"h\":0,\"i\":\"www.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Explore Dataiku Plans and Features | Online or Installed\",\"u\":\"https://www.dataiku.com/product/plans-and-features/\"},{\"a\":\"What's the difference between DataRobot, Databricks Lakehouse, Dataiku DSS, and DATAGYM? Compare DataRobot vs. Databricks Lakehouse vs. Dataiku DSS vs. DATAGYM in 2024 by cost, reviews, features, integrations, deployment, target market, support options, trial offers, training options, years in business, region, and more using the chart below.\",\"ae\":null,\"b\":\"/.\\tSlashdot\\tslashdot.org\",\"c\":\"https://slashdot.org/software/comparison/DataRobot-vs-Databricks-vs-Dataiku-DSS-vs-datagym/\",\"d\":\"slashdot.org/software/comparison/DataRobot-vs-Databricks-vs-Dataiku-DSS-vs-datagym/\",\"da\":\"\",\"h\":0,\"i\":\"slashdot.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Compare DataRobot vs. Databricks Lakehouse vs. Dataiku DSS vs. DATAGYM ...\",\"u\":\"https://slashdot.org/software/comparison/DataRobot-vs-Databricks-vs-Dataiku-DSS-vs-datagym/\"},{\"a\":\"Claim Dataiku DSS and update features and information. Compare C3 AI Suite vs. DataRobot vs. Dataiku DSS using this comparison chart. Compare price, features, and reviews of the software side-by-side to make the best choice for your business.\",\"ae\":null,\"b\":\"srcforge\\tSourceForge\\tsourceforge.net\",\"c\":\"https://sourceforge.net/software/compare/C3-AI-Suite-vs-DataRobot-vs-Dataiku-DSS/\",\"d\":\"sourceforge.net/software/compare/C3-AI-Suite-vs-DataRobot-vs-Dataiku-DSS/\",\"da\":\"\",\"h\":0,\"i\":\"sourceforge.net\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"C3 AI Suite vs. DataRobot vs. Dataiku DSS Comparison - SourceForge\",\"u\":\"https://sourceforge.net/software/compare/C3-AI-Suite-vs-DataRobot-vs-Dataiku-DSS/\"},{\"a\":\"Compare DataRobot AI Platform and Dataiku using real user data focused on features, satisfaction, business value, and the vendor relationship. What is Machine Learning Platforms (ML) Software?\",\"ae\":null,\"c\":\"https://www.softwarereviews.com/categories/machine-learning-platforms/compare/datarobot-ai-platform-vs-dataiku\",\"d\":\"www.softwarereviews.com/categories/machine-learning-platforms/compare/datarobot-ai-platform-vs-dataiku\",\"da\":\"\",\"h\":0,\"i\":\"www.softwarereviews.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"DataRobot AI Platform vs Dataiku - Machine Learning Platforms\",\"u\":\"https://www.softwarereviews.com/categories/machine-learning-platforms/compare/datarobot-ai-platform-vs-dataiku\"},{\"a\":\"What's the difference between Amazon SageMaker, DataRobot, and Dataiku DSS? Compare Amazon SageMaker vs. DataRobot vs. Dataiku DSS in 2024 by cost, reviews, features, integrations, deployment, target market, support options, trial offers, training options, years in business, region, and more using the chart below.\",\"ae\":null,\"b\":\"/.\\tSlashdot\\tslashdot.org\",\"c\":\"https://slashdot.org/software/comparison/Amazon-SageMaker-vs-DataRobot-vs-Dataiku-DSS/\",\"d\":\"slashdot.org/software/comparison/Amazon-SageMaker-vs-DataRobot-vs-Dataiku-DSS/\",\"da\":\"\",\"h\":0,\"i\":\"slashdot.org\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Compare Amazon SageMaker vs. DataRobot vs. Dataiku DSS in 2024 - Slashdot\",\"u\":\"https://slashdot.org/software/comparison/Amazon-SageMaker-vs-DataRobot-vs-Dataiku-DSS/\"},{\"a\":\"Compare Analance vs. DataRobot vs. Dataiku DSS using this comparison chart. Compare price, features, and reviews of the software side-by-side to make the best choice for your business.\",\"ae\":null,\"b\":\"srcforge\\tSourceForge\\tsourceforge.net\",\"c\":\"https://sourceforge.net/software/compare/Analance-vs-DataRobot-vs-Dataiku-DSS/\",\"d\":\"sourceforge.net/software/compare/Analance-vs-DataRobot-vs-Dataiku-DSS/\",\"da\":\"\",\"h\":0,\"i\":\"sourceforge.net\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Analance vs. DataRobot vs. Dataiku DSS Comparison - SourceForge\",\"u\":\"https://sourceforge.net/software/compare/Analance-vs-DataRobot-vs-Dataiku-DSS/\"},{\"n\":\"/d.js?q=Dataiku%20vs%20DataRobot%20features&kl=wt-wt&l=wt-wt&p=&s=23&ex=-1&ct=US&sp=0&vqd=4-334935250614046875026454141242803242982\"}]);DDG.duckbar.load('images');DDG.duckbar.load('news');DDG.duckbar.load('videos');DDG.duckbar.loadModule('related_searches', {\"ads\":[],\"query\":\"Dataiku vs DataRobot features\",\"queryEncoded\":\"Dataiku%20vs%20DataRobot%20features\",\"response_type\":\"places\",\"results\":[{\"display_text\":\"dataiku vs datarobot review\",\"text\":\"dataiku vs datarobot review\",\"web_search_url\":\"?q=dataiku%20vs%20datarobot%20review\"},{\"display_text\":\"dataiku vs alteryx\",\"text\":\"dataiku vs alteryx\",\"web_search_url\":\"?q=dataiku%20vs%20alteryx\"},{\"display_text\":\"gartner dataiku reviews\",\"text\":\"gartner dataiku reviews\",\"web_search_url\":\"?q=gartner%20dataiku%20reviews\"},{\"display_text\":\"alteryx vs dataiku knime\",\"text\":\"alteryx vs dataiku knime\",\"web_search_url\":\"?q=alteryx%20vs%20dataiku%20knime\"},{\"display_text\":\"dataiku vs rapidminer\",\"text\":\"dataiku vs rapidminer\",\"web_search_url\":\"?q=dataiku%20vs%20rapidminer\"},{\"display_text\":\"dataiku vs azure ml\",\"text\":\"dataiku vs azure ml\",\"web_search_url\":\"?q=dataiku%20vs%20azure%20ml\"},{\"display_text\":\"sagemaker vs dataiku\",\"text\":\"sagemaker vs dataiku\",\"web_search_url\":\"?q=sagemaker%20vs%20dataiku\"},{\"display_text\":\"dataiku reviews\",\"text\":\"dataiku reviews\",\"web_search_url\":\"?q=dataiku%20reviews\"}],\"vqd\":{\"Dataiku%20vs%20DataRobot%20features\":\"4-334935250614046875026454141242803242982\"}});if (DDG.pageLayout) DDG.pageLayout.initialize({\"mainline\":{\"items\":[[\"ad\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"related_searches\"]]},\"sidebar\":{\"items\":[[\"wikipedia_fathead\"]]}}, { start: 0 });DDG.deep.emit(\"load:completed\");", + "curl-cffi-POST-https://duckduckgo.com-{\"data\": {\"q\": \"Dataiku and DataRobot use cases\"}}": "Dataiku and DataRobot use cases at DuckDuckGo
", + "curl-cffi-GET-https://links.duckduckgo.com/d.js-{\"params\": {\"bing_market\": \"wt-WT\", \"df\": null, \"ex\": \"-1\", \"kl\": \"wt-wt\", \"l\": \"wt-wt\", \"q\": \"Dataiku and DataRobot use cases\", \"s\": \"0\", \"sp\": \"0\", \"vqd\": \"4-60481969350525797892441552954401970387\"}}": "if (DDG.deep && DDG.deep.setUpstream) DDG.deep.setUpstream(\"bingv7aa\");DDG.deep.bn={'ivc':1};if (DDG.pageLayout) DDG.pageLayout.load('a',[{\"a\":\"\\u9ad8\\u7cbe\\u5ea6\\u306a\\u6a5f\\u68b0\\u5b66\\u7fd2\\u30e2\\u30c7\\u30eb\\u3092\\u69cb\\u7bc9\\u3001\\u5b9f\\u88c5\\u3001\\u904b\\u7528\\u3002DataRobot\\u306f\\u793e\\u5185\\u30c7\\u30fc\\u30bf\\u304b\\u3089\\u65b0\\u3057\\u3044\\u4fa1\\u5024\\u3092\\u5275\\u9020\\u3057\\u307e\\u3059. AI\\u3092\\u6d3b\\u7528\\u3057\\u30c7\\u30fc\\u30bf\\u3092\\u5206\\u6790\\u3001\\u5b9f\\u7528\\u7684\\u306a\\u30a4\\u30f3\\u30b5\\u30a4\\u30c8\\u3092\\u660e\\u3089\\u304b\\u306b\\u3002\\u30d3\\u30b8\\u30cd\\u30b9\\u306e\\u8ab2\\u984c\\u3092\\u3088\\u308a\\u65e9\\u304f\\u89e3\\u6c7a\",\"adext\":{\"callout\":{\"t\":\"30-Day Free Trial \\u00b7 Trusted by Fortune 50 \\u00b7 No Vendor Lock-in\",\"tid\":\"6\"},\"filterlinks\":{\"l\":[],\"tid\":\"\"},\"sitelinks\":{\"l\":[{\"snippet\":\"Explore the DataRobot AI Platform Get Started With a 30-Day Trial\",\"targetUrl\":\"https://duckduckgo.com/y.js?ad_domain=datarobot.com&ad_provider=bingv7aa&ad_type=txad&eddgt=2_trBPli7jgDj1WnqJbnww%3D%3D&rut=d0faee2c8c1aae9ac3a012e21d37352a1181970dce9edeba4107839fbfbf097a&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De81Q_rqdj1ZxH5XXGh4PG6pjVUCUzdB7rGpyykWEihNc_sSp5n%2DJ9jIyTjOSnXg0OUazrpKgDJrNvBOdNa5PjBGtyLGt23nrBAabI6opJXrliWQ4o%2DTyxIsqOeCXqzLOOJ3jJb74k6KEx20zilzwKmzSg3nBop2A9JqsasC17VVDPc3_i3EzPbWeRNS4nhxXWJqBKd55GfhuEOg2RZUbmmuAUhWvM%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuZGF0YXJvYm90LmNvbSUyZnRyaWFsJTJmJTNmdXRtX21lZGl1bSUzZHNlYXJjaCUyNnV0bV9zb3VyY2UlM2RiaW5nJTI2dXRtX2NhbXBhaWduJTNkRnJlZVRyaWFsMjAyM1dXMDgxNkdQU2FkZXh0JTI2Y2FtcGFpZ25pZCUzZDUzMDcwODA5OSUyNmFkZ3JvdXBpZCUzZDEzNTAyMDI3NzQyMTc2OTglMjZhZGlkJTNkJTI2bXNjbGtpZCUzZDc2YmMwNmFmNTA0NDFjOGVjOGYxNjMwY2FmNGU4ZTVk%26rlid%3D76bc06af50441c8ec8f1630caf4e8e5d&vqd=4-164177780916400746369660096493208330918&iurl=%7B1%7DIG%3D5E053B32922A4B5781ED405D9621559B%26CID%3D0176CEA622686E9D34D1DAA0235D6F30%26ID%3DDevEx%2C5063.1\",\"text\":\"DataRobot Free Trial\"},{\"snippet\":\"Unlock Your AI Success in 2023 Tips on the Path of Value-Driven AI\",\"targetUrl\":\"https://duckduckgo.com/y.js?ad_domain=datarobot.com&ad_provider=bingv7aa&ad_type=txad&eddgt=2_trBPli7jgDj1WnqJbnww%3D%3D&rut=fdb107a4de6fffdec2bdf43b561b2c63ca700daaef68f0e683547361efbbc2b0&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De8%2DT0j3GTQEgr%2DmtHPM1LzNzVUCUyRxvVKYHe6LbNa2mmCfCZh3Ept1NM%2DP%2DM1AAluh_OL3VQw_FWI0A3YxC3pzzqthf3gpxan_Lv7CjKenge%2DwMYUz3bRFoFyHtQBMdgqv6T7gMGfyYwN3UCj6FNYwVVn9UNN0h1dIQanHNB6Ya9gRrPBACknA8qtsf6A2oUG1xhq7AOF98NzGphnfQ_38fySnRU%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuZGF0YXJvYm90LmNvbSUyZnJlc291cmNlcyUyZmFpc3VjY2VzczIwMjMlMmYlM2Z1dG1fbWVkaXVtJTNkc2VhcmNoJTI2dXRtX3NvdXJjZSUzZGJpbmclMjZ1dG1fY2FtcGFpZ24lM2RDb250ZW50MTBLZXlzdG9BSVN1Y2Nlc3MyMDIzV1cwNTIyR1BTYWRleHQlMjZ1dG1fdGVybSUzZGRhdGFyb2JvdCUyNnV0bV9jb250ZW50JTNkYWRfZXh0JTI2Y2FtcGFpZ25pZCUzZDUzMDcwODA5OSUyNmFkZ3JvdXBpZCUzZDEzNTAyMDI3NzQyMTc2OTglMjZhZGlkJTNkJTI2bXNjbGtpZCUzZGQzNmQ2MzlkMmFlNTEwMTM3ZTIwMDYzZWQ1ZWY3M2Yz%26rlid%3Dd36d639d2ae510137e20063ed5ef73f3&vqd=4-117927704271333462986714580056949079639&iurl=%7B1%7DIG%3D5E053B32922A4B5781ED405D9621559B%26CID%3D0176CEA622686E9D34D1DAA0235D6F30%26ID%3DDevEx%2C5065.1\",\"text\":\"10 Keys to AI Success\"},{\"snippet\":\"Our Platform Includes Four Fully Integrated Products. Read More.\",\"targetUrl\":\"https://duckduckgo.com/y.js?ad_domain=datarobot.com&ad_provider=bingv7aa&ad_type=txad&eddgt=2_trBPli7jgDj1WnqJbnww%3D%3D&rut=4f06bd3312172b8e61d65ee2626dea6e26d941c3a16aa546b4e11b79e8bf027f&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De8885tVmNmhi65Jmp3f2wYSzVUCUyFey1LCmrSNpGfkWzQnoC7QIbU3ztthJ%2DqKpgCmRfxudhbLK927YN84jvZlV2zTKo9DOULVj5wB8mcGXy_F42SnsrO1jZpY9NnMnzqMYPb5xZTTdgrTO1_w3Bgpd0e0VzO81_O3%2Dfo2z4UiLuVETFVqfACqR6NEwz0yfjzJe6ED9tvi_gPDiUL9iWATrNIrsw%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuZGF0YXJvYm90LmNvbSUyZnByb2R1Y3QlMmYlM2ZjYW1wYWlnbmlkJTNkNTMwNzA4MDk5JTI2YWRncm91cGlkJTNkMTM1MDIwMjc3NDIxNzY5OCUyNmFkaWQlM2QlMjZtc2Nsa2lkJTNkY2U4NzQ1ZDViODBlMTJmNjQ2N2QyMDc2NDcwNDY2YjI%26rlid%3Dce8745d5b80e12f6467d2076470466b2&vqd=4-169069202740993895017985472268973083525&iurl=%7B1%7DIG%3D5E053B32922A4B5781ED405D9621559B%26CID%3D0176CEA622686E9D34D1DAA0235D6F30%26ID%3DDevEx%2C5067.1\",\"text\":\"Product Overview\"}],\"tid\":\"7\\t9[8]\\t11[10]\\t13[12]\",\"type\":\"EnhancedSiteLink\"},\"tid\":\"1\"},\"ae\":{\"callout\":[\"30-Day Free Trial \\u00b7 Trusted by Fortune 50 \\u00b7 No Vendor Lock-in\"]},\"c\":\"https://duckduckgo.com/y.js?ad_domain=datarobot.com&ad_provider=bingv7aa&ad_type=txad&eddgt=2_trBPli7jgDj1WnqJbnww%3D%3D&rut=e744d99a8df00b24df71f821ad4d1332080aa03267e50f0e988d284f58d9d2ef&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De8tT9soRYLZabP1ukFkRsgNzVUCUzl89Y8xEqpxoqHqIlCI5wWbydNnN_PoAKHAa2Vsio83mXA_ax16t6rJ7XGkBv0Cg7_D1eg2QAuJgPKEam4VWI3rW40B03r1p11ZXN1Gd1847Vj05bAnJnPfgVyC8ZzFQxLxONmOI0Hg182z2bZUVII26BUAlUHaVZ7O_9FEXLJWw%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuZGF0YXJvYm90LmNvbSUyZmpwJTJmbHAlMmZhaS1mb3ItYnVzaW5lc3MlMmYlM2Z1dG1fbWVkaXVtJTNkc2VhcmNoJTI2dXRtX3NvdXJjZSUzZGJpbmclMjZ1dG1fY2FtcGFpZ24lM2RERU1PMjAyM0FsbFByb2R1Y3RzSlAwNjI2QlBTJTI2dXRtX3Rlcm0lM2RkYXRhcm9ib3QlMjZ1dG1fY29udGVudCUzZERSX2JyYW5kZWRfcnNhJTI2Y2FtcGFpZ25pZCUzZDUzMDcwODA5OSUyNmFkZ3JvdXBpZCUzZDEzNTAyMDI3NzQyMTc2OTglMjZhZGlkJTNkJTI2bXNjbGtpZCUzZDA2MTIwYzhmMTAxNzEwYTZiNmRiNjkyY2VmMWRiOTY1%26rlid%3D06120c8f101710a6b6db692cef1db965&vqd=4-91027509783546726889708070523412001433&iurl=%7B1%7DIG%3D5E053B32922A4B5781ED405D9621559B%26CID%3D0176CEA622686E9D34D1DAA0235D6F30%26ID%3DDevEx%2C5058.1\",\"d\":\"datarobot.com\",\"h\":0,\"i\":\"\",\"k\":0,\"m\":0,\"o\":\"\",\"p\":1,\"relevancy\":{\"abstract\":\"%E9%AB%98%E7%B2%BE%E5%BA%A6%E3%81%AA%E6%A9%9F%E6%A2%B0%E5%AD%A6%E7%BF%92%E3%83%A2%E3%83%87%E3%83%AB%E3%82%92%E6%A7%8B%E7%AF%89%E3%80%81%E5%AE%9F%E8%A3%85%E3%80%81%E9%81%8B%E7%94%A8%E3%80%82%3Cb%3EDataRobot%3C%2Fb%3E%E3%81%AF%E7%A4%BE%E5%86%85%E3%83%87%E3%83%BC%E3%82%BF%E3%81%8B%E3%82%89%E6%96%B0%E3%81%97%E3%81%84%E4%BE%A1%E5%80%A4%E3%82%92%E5%89%B5%E9%80%A0%E3%81%97%E3%81%BE%E3%81%99.%20AI%E3%82%92%E6%B4%BB%E7%94%A8%E3%81%97%E3%83%87%E3%83%BC%E3%82%BF%E3%82%92%E5%88%86%E6%9E%90%E3%80%81%E5%AE%9F%E7%94%A8%E7%9A%84%E3%81%AA%E3%82%A4%E3%83%B3%E3%82%B5%E3%82%A4%E3%83%88%E3%82%92%E6%98%8E%E3%82%89%E3%81%8B%E3%81%AB%E3%80%82%E3%83%93%E3%82%B8%E3%83%8D%E3%82%B9%E3%81%AE%E8%AA%B2%E9%A1%8C%E3%82%92%E3%82%88%E3%82%8A%E6%97%A9%E3%81%8F%E8%A7%A3%E6%B1%BA\",\"adx_name\":\"none\",\"is_good_v10\":0,\"organic_ranks\":[5,11,12,13],\"q\":\"Dataiku%20and%20DataRobot%20use%20cases\",\"q_words\":4,\"q_words_fuzzy\":0.25,\"q_words_in_ad\":1,\"root_domain\":\"datarobot.com\",\"start\":\"0\",\"title\":\"%E3%83%93%E3%83%83%E3%82%B0%E3%83%87%E3%83%BC%E3%82%BF%E5%88%86%E6%9E%90%E3%82%92%E9%AB%98%E9%80%9F%E5%8C%96%20%2D%20%E3%83%87%E3%83%BC%E3%82%BF%E3%81%8B%E3%82%89%E6%96%B0%E3%81%97%E3%81%84%E4%BE%A1%E5%80%A4%E3%82%92\"},\"s\":\"bingv7aa\",\"t\":\"\\u30d3\\u30c3\\u30b0\\u30c7\\u30fc\\u30bf\\u5206\\u6790\\u3092\\u9ad8\\u901f\\u5316 - \\u30c7\\u30fc\\u30bf\\u304b\\u3089\\u65b0\\u3057\\u3044\\u4fa1\\u5024\\u3092\",\"tid\":\"1,6,7,9[8],11[10],13[12]\",\"u\":\"https://duckduckgo.com/y.js?ad_domain=datarobot.com&ad_provider=bingv7aa&ad_type=txad&eddgt=2_trBPli7jgDj1WnqJbnww%3D%3D&rut=e744d99a8df00b24df71f821ad4d1332080aa03267e50f0e988d284f58d9d2ef&u3=https%3A%2F%2Fwww.bing.com%2Faclick%3Fld%3De8tT9soRYLZabP1ukFkRsgNzVUCUzl89Y8xEqpxoqHqIlCI5wWbydNnN_PoAKHAa2Vsio83mXA_ax16t6rJ7XGkBv0Cg7_D1eg2QAuJgPKEam4VWI3rW40B03r1p11ZXN1Gd1847Vj05bAnJnPfgVyC8ZzFQxLxONmOI0Hg182z2bZUVII26BUAlUHaVZ7O_9FEXLJWw%26u%3DaHR0cHMlM2ElMmYlMmZ3d3cuZGF0YXJvYm90LmNvbSUyZmpwJTJmbHAlMmZhaS1mb3ItYnVzaW5lc3MlMmYlM2Z1dG1fbWVkaXVtJTNkc2VhcmNoJTI2dXRtX3NvdXJjZSUzZGJpbmclMjZ1dG1fY2FtcGFpZ24lM2RERU1PMjAyM0FsbFByb2R1Y3RzSlAwNjI2QlBTJTI2dXRtX3Rlcm0lM2RkYXRhcm9ib3QlMjZ1dG1fY29udGVudCUzZERSX2JyYW5kZWRfcnNhJTI2Y2FtcGFpZ25pZCUzZDUzMDcwODA5OSUyNmFkZ3JvdXBpZCUzZDEzNTAyMDI3NzQyMTc2OTglMjZhZGlkJTNkJTI2bXNjbGtpZCUzZDA2MTIwYzhmMTAxNzEwYTZiNmRiNjkyY2VmMWRiOTY1%26rlid%3D06120c8f101710a6b6db692cef1db965&vqd=4-91027509783546726889708070523412001433&iurl=%7B1%7DIG%3D5E053B32922A4B5781ED405D9621559B%26CID%3D0176CEA622686E9D34D1DAA0235D6F30%26ID%3DDevEx%2C5058.1\"}], {\"page_load_url\":\"https://duckduckgo.com/y.js?ifu=%7B3%7Dappid%3D055AAD1BA669BEB8B048128DC89A107C678B527B%26rguid%3D309794dc72f748f6a2b95ce5c34fbcec&iurl=%7B2%7DIG%3D5E053B32922A4B5781ED405D9621559B%26CID%3D0176CEA622686E9D34D1DAA0235D6F30%26Type%3DEvent.CPT%26DATA%3D0\",\"visibility_url\":\"https://duckduckgo.com/y.js?ivu=%7B4%7Dtype%3Dmv%26reqver%3D1.0%26rg%3D309794dc72f748f6a2b95ce5c34fbcec\"});DDG.deep.signalSummary = \"\";DDG.inject('DDG.Data.languages.resultLanguages', {\"en\":[\"https://knowledge.dataiku.com/latest/use-cases/index.html\",\"https://community.dataiku.com/t5/Dataiku-Use-Cases-Success/tkb-p/use-cases\",\"https://www.datarevenue.com/en-blog/ml-platforms-dataiku-vs-alteryx-vs-sagemaker\",\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/compare/dataiku-vs-datarobot\",\"https://community.dataiku.com/t5/General-Discussion/Dataiku-vs-DataRobot/m-p/9315\",\"https://www.datarobot.com/use-cases/\",\"https://academy.dataiku.com/page/use-cases\",\"https://www.trustradius.com/compare-products/dataiku-dss-vs-datarobot\",\"https://www.g2.com/compare/datarobot-vs-dataiku-dss\",\"https://www.linkedin.com/pulse/managed-machine-learning-platforms-comparative-analysis/\",\"https://londondataconsulting.medium.com/dataiku-what-is-it-how-to-use-it-ultimate-guide-2023-47602c85a48b\",\"https://docs.datarobot.com/en/docs/api/guide/common-case/index.html\",\"https://www.datarobot.com/blog/introducing-the-datarobot-use-case-value-tracker/\",\"https://docs.datarobot.com/en/docs/workbench/wb-usecase/wb-build-usecase.html\",\"https://blog.dataiku.com/topic/use-cases-projects\",\"https://valohai.com/mlops-platforms-compared/\",\"https://www.capterra.com/machine-learning-software/compare/142192-179303/Data-Science-Studio-DSS-vs-DataRobot\",\"https://pages.dataiku.com/experience-a-dataiku-demo\",\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/dataiku/product/dataiku\",\"https://www.dataiku.com/stories/\",\"https://www.dataiku.com/\",\"https://techcrunch.com/2022/12/13/ai-and-analytics-platform-dataiku-raises-200m-at-a-reduced-valuation/\",\"https://www.globenewswire.com/news-release/2022/11/17/2558152/0/en/Ben-Taylor-Joins-Dataiku-as-Chief-AI-Strategist.html\"]});DDG.deep.pageLayoutSummary = \"a1w4v1w19,w1\";DDG.inject('DDG.Data.languages.adLanguages', {});if (DDG.pageLayout) DDG.pageLayout.load('d',[{\"a\":\"Use Cases - Dataiku Knowledge Base Use Cases # These use cases allow you to practice what you've learned by building simplified, but complete use cases in Dataiku. Topics # Data Preparation Use Cases Classification Use Cases Clustering Use Cases Plugin Use Cases\",\"ae\":null,\"c\":\"https://knowledge.dataiku.com/latest/use-cases/index.html\",\"d\":\"knowledge.dataiku.com/latest/use-cases/index.html\",\"da\":\"\",\"h\":0,\"i\":\"knowledge.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Use Cases - Dataiku Knowledge Base\",\"u\":\"https://knowledge.dataiku.com/latest/use-cases/index.html\"},{\"a\":\"Community Dataiku Use Cases & Success Stories \\u26a0\\ufe0f Discover pioneering Dataiku use cases and success stories shared by customers, partners, academics, and nonprofits participating in the Dataiku Frontrunner Awards. Use the following labels to filter submissions by industry:\",\"ae\":null,\"c\":\"https://community.dataiku.com/t5/Dataiku-Use-Cases-Success/tkb-p/use-cases\",\"d\":\"community.dataiku.com/t5/Dataiku-Use-Cases-Success/tkb-p/use-cases\",\"da\":\"\",\"h\":0,\"i\":\"community.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku Use Cases & Success Stories - Dataiku Community\",\"u\":\"https://community.dataiku.com/t5/Dataiku-Use-Cases-Success/tkb-p/use-cases\"},{\"a\":\"Dataiku is a cross-platform desktop application that includes a broad range of tools, such as notebooks (similar to Jupyter Notebook), workflow management (similar to Apache Airflow), and automated machine learning. In general, Dataiku aims to replace many of your existing tools rather than to integrate with them.\",\"ae\":null,\"c\":\"https://www.datarevenue.com/en-blog/ml-platforms-dataiku-vs-alteryx-vs-sagemaker\",\"d\":\"www.datarevenue.com/en-blog/ml-platforms-dataiku-vs-alteryx-vs-sagemaker\",\"da\":\"\",\"h\":0,\"i\":\"www.datarevenue.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"ML Platforms: Dataiku vs. Alteryx vs. Sagemaker vs. Datarobot\",\"u\":\"https://www.datarevenue.com/en-blog/ml-platforms-dataiku-vs-alteryx-vs-sagemaker\"},{\"a\":\"Dataiku has a rating of 4.8 stars with 504 reviews. DataRobot has a rating of 4.6 stars with 508 reviews. See side-by-side comparisons of product capabilities, customer experience, pros and cons, and reviewer demographics to find the best fit for your organization. See more companies in the Data Science and Machine Learning Platforms market. PDF.\",\"ae\":null,\"c\":\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/compare/dataiku-vs-datarobot\",\"d\":\"www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/compare/dataiku-vs-datarobot\",\"da\":\"\",\"h\":0,\"i\":\"www.gartner.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku vs DataRobot 2024 | Gartner Peer Insights\",\"u\":\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/compare/dataiku-vs-datarobot\"},{\"a\":\"In my humble opinion DSS is a more a 'toolbox', where as DataRobot is an autoML platform. DataRobot is really good at what it does - if you have non-technical team who want to drop in data and leave everything to autoML then this may be the option for them.\",\"ae\":null,\"c\":\"https://community.dataiku.com/t5/General-Discussion/Dataiku-vs-DataRobot/m-p/9315\",\"d\":\"community.dataiku.com/t5/General-Discussion/Dataiku-vs-DataRobot/m-p/9315\",\"da\":\"\",\"h\":0,\"i\":\"community.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Solved: Dataiku vs DataRobot - Dataiku Community\",\"u\":\"https://community.dataiku.com/t5/General-Discussion/Dataiku-vs-DataRobot/m-p/9315\"},{\"a\":\"Use cases AI Use Cases AI-driven organizations around the world use DataRobot to solve their most pressing business problems. Build with Free Trial Recent Popular Filters Ready to Get Started? See how a value-driven approach to AI can accelerate time to impact. Start Free Trial\",\"ae\":null,\"c\":\"https://www.datarobot.com/use-cases/\",\"d\":\"www.datarobot.com/use-cases/\",\"da\":\"\",\"h\":0,\"i\":\"www.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Machine Learning Use Cases | DataRobot AI Platform\",\"u\":\"https://www.datarobot.com/use-cases/\"},{\"a\":\"With Dataiku's AI Prepare assistant, you can work smarter, not harder. Simply describe the transformation you want to apply in natural language and the AI assistant automatically generates the necessary data preparation steps. The ability to modify both your prompt and the resulting steps means you can prepare data faster than ever, yet still ...\",\"ae\":null,\"c\":\"https://academy.dataiku.com/page/use-cases\",\"d\":\"academy.dataiku.com/page/use-cases\",\"da\":\"\",\"h\":0,\"i\":\"academy.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Use Cases - Dataiku\",\"u\":\"https://academy.dataiku.com/page/use-cases\"},{\"a\":\"84 Reviews and Ratings Path to AI Success Compare Dataiku DSS vs DataRobot. 103 verified user reviews and ratings of features, pros, cons, pricing, support and more.\",\"ae\":null,\"c\":\"https://www.trustradius.com/compare-products/dataiku-dss-vs-datarobot\",\"d\":\"www.trustradius.com/compare-products/dataiku-dss-vs-datarobot\",\"da\":\"\",\"h\":0,\"i\":\"www.trustradius.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku DSS vs DataRobot | TrustRadius\",\"u\":\"https://www.trustradius.com/compare-products/dataiku-dss-vs-datarobot\"},{\"a\":\"side-by-side comparison of DataRobot vs. Dataiku DSS. based on preference data from user reviews. DataRobot rates 4.4/5 stars with 26 reviews. By contrast, Dataiku DSS rates 4.3/5 stars with 36 reviews. Each product's score is calculated with real-time data from verified user reviews, to help you make the best choice between these two options ...\",\"ae\":null,\"c\":\"https://www.g2.com/compare/datarobot-vs-dataiku-dss\",\"d\":\"www.g2.com/compare/datarobot-vs-dataiku-dss\",\"da\":\"\",\"h\":0,\"i\":\"www.g2.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Compare DataRobot vs. Dataiku DSS | G2\",\"u\":\"https://www.g2.com/compare/datarobot-vs-dataiku-dss\"},{\"a\":\"Use case: Choose Datarobot if you have data stored in spreadsheets and are seeking a platform that is the simplest, albeit one with limited flexibility, ... Dataiku vs. Datarobot .\",\"ae\":null,\"b\":\"li\\tLinkedIn\\twww.linkedin.com\",\"c\":\"https://www.linkedin.com/pulse/managed-machine-learning-platforms-comparative-analysis/\",\"d\":\"www.linkedin.com/pulse/managed-machine-learning-platforms-comparative-analysis/\",\"da\":\"\",\"e\":\"2023-08-11T00:00:00.0000000\",\"h\":0,\"i\":\"www.linkedin.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Managed Machine Learning Platforms: A Comparative Analysis - LinkedIn\",\"u\":\"https://www.linkedin.com/pulse/managed-machine-learning-platforms-comparative-analysis/\"},{\"a\":\"Jan 11, 2023 Dataiku is an artificial intelligence platform created in France in 2013. It has since become one of the world's benchmarks for data science and machine learning studios. What is...\",\"ae\":null,\"c\":\"https://londondataconsulting.medium.com/dataiku-what-is-it-how-to-use-it-ultimate-guide-2023-47602c85a48b\",\"d\":\"londondataconsulting.medium.com/dataiku-what-is-it-how-to-use-it-ultimate-guide-2023-47602c85a48b\",\"da\":\"translations\",\"e\":\"2023-01-11T00:00:00.0000000\",\"h\":0,\"i\":\"londondataconsulting.medium.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku: What is it? How to use it? Ultimate Guide 2023\",\"u\":\"https://londondataconsulting.medium.com/dataiku-what-is-it-how-to-use-it-ultimate-guide-2023-47602c85a48b\"},{\"a\":\"Use cases for version 2.x. Notebooks for uses cases that use methods for 2.x versions of DataRobot's Python client. Measure price elasticity of demand. A use case to identify relationships between price and demand, maximize revenue by properly pricing products, and monitor price elasticities for changes in price and demand. Insurance claim triage.\",\"ae\":null,\"c\":\"https://docs.datarobot.com/en/docs/api/guide/common-case/index.html\",\"d\":\"docs.datarobot.com/en/docs/api/guide/common-case/index.html\",\"da\":\"\",\"h\":0,\"i\":\"docs.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Common use cases: DataRobot docs - DataRobot AI Platform\",\"u\":\"https://docs.datarobot.com/en/docs/api/guide/common-case/index.html\"},{\"a\":\"With the Use Case Value Tracker, you can manage the project lifecycle and understand the value associated with each step. It also enables you to associate and organize all your DataRobot artifacts (e.g., datasets, models, deployments, applications, etc.) around a given use case for a holistic view. In addition to the project management aspects ...\",\"ae\":null,\"c\":\"https://www.datarobot.com/blog/introducing-the-datarobot-use-case-value-tracker/\",\"d\":\"www.datarobot.com/blog/introducing-the-datarobot-use-case-value-tracker/\",\"da\":\"\",\"h\":0,\"i\":\"www.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Introducing the DataRobot Use Case Value Tracker\",\"u\":\"https://www.datarobot.com/blog/introducing-the-datarobot-use-case-value-tracker/\"},{\"a\":\"Use Cases are folder-like containers inside of DataRobot Workbench that allow you to group everything related to solving a specific business problem\\u2014datasets, models, experiments, No-Code AI Apps, and notebooks\\u2014inside of a single, manageable entity. You can share whole Use Cases as well as the individual assets they contain.\",\"ae\":null,\"c\":\"https://docs.datarobot.com/en/docs/workbench/wb-usecase/wb-build-usecase.html\",\"d\":\"docs.datarobot.com/en/docs/workbench/wb-usecase/wb-build-usecase.html\",\"da\":\"\",\"e\":\"2023-09-15T00:00:00.0000000\",\"h\":0,\"i\":\"docs.datarobot.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Use Cases: DataRobot docs\",\"u\":\"https://docs.datarobot.com/en/docs/workbench/wb-usecase/wb-build-usecase.html\"},{\"a\":\"January 2, 2024 Use Cases & Projects, Featured Sophie Dionnet Leveraging AI to Cut Costs December 29, 2023 Data Basics, Featured\",\"ae\":null,\"c\":\"https://blog.dataiku.com/topic/use-cases-projects\",\"d\":\"blog.dataiku.com/topic/use-cases-projects\",\"da\":\"\",\"h\":0,\"i\":\"blog.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Blog - Dataiku | Use Cases & Projects\",\"u\":\"https://blog.dataiku.com/topic/use-cases-projects\"},{\"a\":\"The platforms we've chosen for our analysis are ClearML, cnvrg.io, Dataiku, Datarobot, Iguazio, Sagemaker, Seldon and Valohai from the managed side, and Flyte, Kubeflow, MLflow and Metaflow from the open-source side. This is by no means an exhaustive list of all the MLOps tools out there.\",\"ae\":null,\"c\":\"https://valohai.com/mlops-platforms-compared/\",\"d\":\"valohai.com/mlops-platforms-compared/\",\"da\":\"\",\"h\":0,\"i\":\"valohai.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"MLOps Platforms Compared - Valohai\",\"u\":\"https://valohai.com/mlops-platforms-compared/\"},{\"a\":\"DataRobot. DSS is for all companies, whatever their expertise, industry or size, that want to create their own data-driven strategic advantages by transforming their raw data into business impacting predictions. Cloud based machine learning platform which helps enterprises scale data science capabilities through deploying machine learning ...\",\"ae\":null,\"c\":\"https://www.capterra.com/machine-learning-software/compare/142192-179303/Data-Science-Studio-DSS-vs-DataRobot\",\"d\":\"www.capterra.com/machine-learning-software/compare/142192-179303/Data-Science-Studio-DSS-vs-DataRobot\",\"da\":\"translations\",\"h\":0,\"i\":\"www.capterra.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Compare Dataiku vs DataRobot 2024 | Capterra\",\"u\":\"https://www.capterra.com/machine-learning-software/compare/142192-179303/Data-Science-Studio-DSS-vs-DataRobot\"},{\"a\":\"For Every Industry & Use Case. Organizations that use Dataiku elevate their people (whether technical and working in code or on the business side and low- or no-code) to extraordinary, arming them with the ability to make better day-to-day decisions with data across: Banking & Insurance. Pharmaceuticals. Manufacturing. Telecommunications.\",\"ae\":null,\"c\":\"https://pages.dataiku.com/experience-a-dataiku-demo\",\"d\":\"pages.dataiku.com/experience-a-dataiku-demo\",\"da\":\"\",\"h\":0,\"i\":\"pages.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Check Out This Dataiku Demo\",\"u\":\"https://pages.dataiku.com/experience-a-dataiku-demo\"},{\"a\":\"4 Star 24% 3 Star 1% 2 Star 0% 1 Star 0% Distribution based on 504 ratings Customer Experience Evaluation & Contracting 4.6 Integration & Deployment 4.7 Service & Support 4.8 Product Capabilities 4.8 FREE View and Download Peer Insights About Dataiku\",\"ae\":null,\"c\":\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/dataiku/product/dataiku\",\"d\":\"www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/dataiku/product/dataiku\",\"da\":\"\",\"h\":0,\"i\":\"www.gartner.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku Reviews, Ratings & Features 2024 | Gartner Peer Insights\",\"u\":\"https://www.gartner.com/reviews/market/data-science-and-machine-learning-platforms/vendor/dataiku/product/dataiku\"},{\"a\":\"Read The Full Case Study U.S. Venture + Dataiku: Upskilling Analysts to Save Thousands of Hours The Data and Analytics team at U.S. Venture was built to usher the company into the future of data science and AI.\",\"ae\":null,\"c\":\"https://www.dataiku.com/stories/\",\"d\":\"www.dataiku.com/stories/\",\"da\":\"\",\"h\":0,\"i\":\"www.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Stories | Dataiku\",\"u\":\"https://www.dataiku.com/stories/\"},{\"a\":\"MLOps Deploy, monitor, and maintain machine learning models, all in a single platform. Explore the Capability Collaboration With Dataiku, teams can move beyond the lab and build real and safe Generative AI applications at enterprise scale. Explore the Capability Governance\",\"ae\":null,\"c\":\"https://www.dataiku.com/\",\"d\":\"www.dataiku.com\",\"da\":\"\",\"h\":0,\"i\":\"www.dataiku.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Dataiku | Everyday AI, Extraordinary People\",\"u\":\"https://www.dataiku.com/\"},{\"a\":\"The company today announced that it raised $200 million in a Series F round led by Wellington Management at a $3.7 billion valuation, down from the $4.6 billion that Dataiku received in August ...\",\"ae\":null,\"b\":\"tc\\tTechcrunch\\ttechcrunch.com\",\"c\":\"https://techcrunch.com/2022/12/13/ai-and-analytics-platform-dataiku-raises-200m-at-a-reduced-valuation/\",\"d\":\"techcrunch.com/2022/12/13/ai-and-analytics-platform-dataiku-raises-200m-at-a-reduced-valuation/\",\"da\":\"translations\",\"e\":\"2022-12-13T17:10:00.0000000\",\"h\":0,\"i\":\"techcrunch.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"AI and analytics platform Dataiku raises $200M at a reduced valuation\",\"u\":\"https://techcrunch.com/2022/12/13/ai-and-analytics-platform-dataiku-raises-200m-at-a-reduced-valuation/\"},{\"a\":\"Today, more than 500 companies worldwide use Dataiku to integrate and streamline their use of data, analytics, and AI, driving diverse use cases from fraud detection and customer churn prevention ...\",\"ae\":null,\"c\":\"https://www.globenewswire.com/news-release/2022/11/17/2558152/0/en/Ben-Taylor-Joins-Dataiku-as-Chief-AI-Strategist.html\",\"d\":\"www.globenewswire.com/news-release/2022/11/17/2558152/0/en/Ben-Taylor-Joins-Dataiku-as-Chief-AI-Strategist.html\",\"da\":\"translations\",\"e\":\"2022-11-17T13:00:00.0000000\",\"h\":0,\"i\":\"www.globenewswire.com\",\"k\":null,\"m\":0,\"o\":0,\"p\":0,\"s\":\"bingv7aa\",\"t\":\"Ben Taylor Joins Dataiku as Chief AI Strategist - GlobeNewswire\",\"u\":\"https://www.globenewswire.com/news-release/2022/11/17/2558152/0/en/Ben-Taylor-Joins-Dataiku-as-Chief-AI-Strategist.html\"},{\"n\":\"/d.js?q=Dataiku%20and%20DataRobot%20use%20cases&kl=wt-wt&l=wt-wt&p=&s=23&ex=-1&ct=US&sp=0&vqd=4-60481969350525797892441552954401970387\"}]);DDG.duckbar.load('images');DDG.duckbar.load('news');DDG.duckbar.load('videos', {\"ads\":[],\"query\":\"Dataiku and DataRobot use cases\",\"queryEncoded\":\"Dataiku%20and%20DataRobot%20use%20cases\",\"response_type\":\"places\",\"results\":[{\"content\":\"https://www.youtube.com/watch?v=ryZRRIjQ5Z8\",\"description\":\"If you're a code-first data practitioner, Dataiku helps you efficiently build high quality data pipelines and models in a number of ways. CHECK OUT DATAIKU: https://bit.ly/36XBlpK EGG ON AIR: https://bit.ly/37GhXMY BRIGHTTALK WEBINARS: https://bit.ly/33TIRjn DATA SCIENCE PIONEERS DOCUMENTARY: https://bit.ly/36V3rBF PARTNER ECOSYSTEM: https ...\",\"duration\":\"10:43\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/ryZRRIjQ5Z8?autoplay=1\",\"image_token\":\"8a4abca8613c6680a108591849e5d7b13b86111004ae004898a7f059b64c8355\",\"images\":{\"large\":\"https://tse4.mm.bing.net/th?id=OVP.WoendyuZJ9qxql-n6jit5AEsDh&pid=Api\",\"medium\":\"https://tse4.mm.bing.net/th?id=OVP.WoendyuZJ9qxql-n6jit5AEsDh&pid=Api\",\"motion\":\"https://tse4.mm.bing.net/th?id=OM1.cmvppfhHVUeE4Q_1684256861&pid=Api\",\"small\":\"https://tse4.mm.bing.net/th?id=OVP.WoendyuZJ9qxql-n6jit5AEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2021-06-08T21:15:02.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":12391},\"title\":\"Dataiku Demo for Data Scientists and Coders\",\"uploader\":\"Dataiku\"},{\"content\":\"https://www.youtube.com/watch?v=jKL_I0SCl_E\",\"description\":\"This video showcases the Clinical Trial Explorer use case from the Dataiku Generative AI Use Case Collection. Learn more about Dataiku for Generative AI | https://www.dataiku.com/product/generative-ai/ Explore more Generative AI use cases | https://experience.dataiku.com/generative-ai To explore more about Dataiku, check out the rest of our ...\",\"duration\":\"1:50\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/jKL_I0SCl_E?autoplay=1\",\"image_token\":\"7b19602fe6d9b761aa3cc138448cc632ddbed31da3abf2687f36705f5945973d\",\"images\":{\"large\":\"https://tse3.mm.bing.net/th?id=OVP.wfgHp53woiVosZ37-1HtnwEsDh&pid=Api\",\"medium\":\"https://tse3.mm.bing.net/th?id=OVP.wfgHp53woiVosZ37-1HtnwEsDh&pid=Api\",\"motion\":\"https://tse3.mm.bing.net/th?id=OM2.ZX_yq0xmyCGZBg_1696372683&pid=Api\",\"small\":\"https://tse3.mm.bing.net/th?id=OVP.wfgHp53woiVosZ37-1HtnwEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-06-22T12:19:00.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":150},\"title\":\"Clinical Trial Explorer\",\"uploader\":\"Dataiku\"},{\"content\":\"https://www.youtube.com/watch?v=lASesA4gNFI\",\"description\":\"This video showcases the CO2 Forecast Analyzer use case from the Dataiku Generative AI Use Case Collection. Learn more about Dataiku for Generative AI | https://www.dataiku.com/product/generative-ai/ Explore more Generative AI use cases | https://experience.dataiku.com/generative-ai To explore more about Dataiku, check out the rest of our ...\",\"duration\":\"1:50\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/lASesA4gNFI?autoplay=1\",\"image_token\":\"a09328adca01a788783d759561c2f9c9d4d214e5a26f1462d2b6b69f21a2d478\",\"images\":{\"large\":\"https://tse2.mm.bing.net/th?id=OVP.ZhQI7z-IMlcVnyeMJuGlAQEsDh&pid=Api\",\"medium\":\"https://tse2.mm.bing.net/th?id=OVP.ZhQI7z-IMlcVnyeMJuGlAQEsDh&pid=Api\",\"motion\":\"\",\"small\":\"https://tse2.mm.bing.net/th?id=OVP.ZhQI7z-IMlcVnyeMJuGlAQEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-06-22T12:16:20.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":49},\"title\":\"CO2 Forecast Analyzer\",\"uploader\":\"Dataiku\"},{\"content\":\"https://www.youtube.com/watch?v=RecpD6Vtzj4\",\"description\":\"This video showcases the LLM-Enhanced ESG Document Intelligence use case from the Dataiku Generative AI Use Case Collection. Learn more about Dataiku for Generative AI | https://www.dataiku.com/product/generative-ai/ Explore more Generative AI use cases | https://experience.dataiku.com/generative-ai To explore more about Dataiku, check out the ...\",\"duration\":\"1:20\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/RecpD6Vtzj4?autoplay=1\",\"image_token\":\"6f797accb167e2e6ff7265e35116cdeb9f1c641b1df47932d9597b61b0108614\",\"images\":{\"large\":\"https://tse2.mm.bing.net/th?id=OVP.bm4gAiJOOmKV7uup6LU9pgEsDh&pid=Api\",\"medium\":\"https://tse2.mm.bing.net/th?id=OVP.bm4gAiJOOmKV7uup6LU9pgEsDh&pid=Api\",\"motion\":\"https://tse2.mm.bing.net/th?id=OM1.M8WXwCQ79nrqEA_1691502936&pid=Api\",\"small\":\"https://tse2.mm.bing.net/th?id=OVP.bm4gAiJOOmKV7uup6LU9pgEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-06-22T12:30:00.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":382},\"title\":\"LLM-Enhanced ESG Document Intelligence\",\"uploader\":\"Dataiku\"},{\"content\":\"https://www.youtube.com/watch?v=zLW0TkJoHLw\",\"description\":\"This video showcases the Demand Forecast Analyzer use case from the Dataiku Generative AI Use Case Collection. Learn more about Dataiku for Generative AI | https://www.dataiku.com/product/generative-ai/ Explore more Generative AI use cases | https://experience.dataiku.com/generative-ai To explore more about Dataiku, check out the rest of our ...\",\"duration\":\"1:42\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/zLW0TkJoHLw?autoplay=1\",\"image_token\":\"524eb9572bf9342b859509285d39ec4661fc572cb1452307acc5341b56bab921\",\"images\":{\"large\":\"https://tse1.mm.bing.net/th?id=OVP.yIekQ2cMUesOPJTfsYIZzQHgFo&pid=Api\",\"medium\":\"https://tse1.mm.bing.net/th?id=OVP.yIekQ2cMUesOPJTfsYIZzQHgFo&pid=Api\",\"motion\":\"\",\"small\":\"https://tse1.mm.bing.net/th?id=OVP.yIekQ2cMUesOPJTfsYIZzQHgFo&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-06-22T12:15:02.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":4},\"title\":\"LLM-Enhanced Demand Forecast\",\"uploader\":\"Dataiku\"},{\"content\":\"https://www.youtube.com/watch?v=L-Yys0fzuVY\",\"description\":\"This video showcases the Production Quality Data Explorer use case from the Dataiku Generative AI Use Case Collection. Learn more about Dataiku for Generative AI | https://www.dataiku.com/product/generative-ai/ Explore more Generative AI use cases | https://experience.dataiku.com/generative-ai To explore more about Dataiku, check out the rest ...\",\"duration\":\"1:47\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/L-Yys0fzuVY?autoplay=1\",\"image_token\":\"82924713be1dba83d67124fcaa6cc6afd163900a0c40f25fcf6c144ed0e36536\",\"images\":{\"large\":\"https://tse4.mm.bing.net/th?id=OVP.teiC0mX9nKCbH8qeo52udwEsDh&pid=Api\",\"medium\":\"https://tse4.mm.bing.net/th?id=OVP.teiC0mX9nKCbH8qeo52udwEsDh&pid=Api\",\"motion\":\"https://tse4.mm.bing.net/th?id=OM2.IwRaoLRWcQXzag_1691419996&pid=Api\",\"small\":\"https://tse4.mm.bing.net/th?id=OVP.teiC0mX9nKCbH8qeo52udwEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-06-22T12:28:29.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":175},\"title\":\"Production Quality Data Explorer\",\"uploader\":\"Dataiku\"},{\"content\":\"https://www.youtube.com/watch?v=FluiuHuaU8A\",\"description\":\"In this breakout session of Dataiku's Product Days 2021, you will see a demo of Dataiku's Data Science Studio, the centralized, collaborative, and end-to-end platform for data science in the enterprise. CHECK OUT DATAIKU: https://bit.ly/36XBlpK EGG ON AIR: https://bit.ly/37GhXMY BRIGHTTALK WEBINARS: https://bit.ly/33TIRjn DATA SCIENCE PIONEERS ...\",\"duration\":\"13:50\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/FluiuHuaU8A?autoplay=1\",\"image_token\":\"2943fa8c1580f2936fc11667d670c0b827b94ff3d16b897f8b5ef2e2426487b3\",\"images\":{\"large\":\"https://tse2.mm.bing.net/th?id=OVP.RIM-ftwDZjYP58RimJfgwwEsDh&pid=Api\",\"medium\":\"https://tse2.mm.bing.net/th?id=OVP.RIM-ftwDZjYP58RimJfgwwEsDh&pid=Api\",\"motion\":\"https://tse2.mm.bing.net/th?id=OM1.MIQ7BoQz1MVkNw_1662248868&pid=Api\",\"small\":\"https://tse2.mm.bing.net/th?id=OVP.RIM-ftwDZjYP58RimJfgwwEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2021-07-08T15:56:22.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":3844},\"title\":\"Introduction to Dataiku Data Science | Product Days 2021\",\"uploader\":\"Dataiku\"},{\"content\":\"https://www.youtube.com/watch?v=6TEU5JboP7k\",\"description\":\"This video showcases the LLM-Enhanced Next Best Offer use case from the Dataiku Generative AI Use Case Collection. Learn more about Dataiku for Generative AI | https://www.dataiku.com/product/generative-ai/ Explore more Generative AI use cases | https://experience.dataiku.com/generative-ai To explore more about Dataiku, check out the rest of ...\",\"duration\":\"1:47\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/6TEU5JboP7k?autoplay=1\",\"image_token\":\"a3bc327ff2f099462935a8979bb599655c7a88a44e25a64bfea7e5973f773158\",\"images\":{\"large\":\"https://tse2.mm.bing.net/th?id=OVP.Q6SP0MmL89M_TnLvNPG4oQEsDh&pid=Api\",\"medium\":\"https://tse2.mm.bing.net/th?id=OVP.Q6SP0MmL89M_TnLvNPG4oQEsDh&pid=Api\",\"motion\":\"https://tse2.mm.bing.net/th?id=OM2.wXNo1CUgYV4Flg_1694065487&pid=Api\",\"small\":\"https://tse2.mm.bing.net/th?id=OVP.Q6SP0MmL89M_TnLvNPG4oQEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-06-22T12:34:32.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":462},\"title\":\"LLM-Enhanced Next Best Offer\",\"uploader\":\"Dataiku\"},{\"content\":\"https://www.youtube.com/watch?v=UVbrpX8Zkn8\",\"description\":\"Move beyond the lab and build real and safe Generative AI applications at enterprise scale. Dataiku brings enterprise-grade development tools, pre-built use cases, and AI-powered assistants throughout the platform. Learn more about Dataiku for Generative AI | https://www.dataiku.com/product/generative-ai/ Explore Generative AI use cases | https ...\",\"duration\":\"2:07\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/UVbrpX8Zkn8?autoplay=1\",\"image_token\":\"3968d2d01ff722efa156290344ab0b37164e57d7efa50905c346ea1cc1a5d369\",\"images\":{\"large\":\"https://tse1.mm.bing.net/th?id=OVP.F-xH3-wKfTjM3YMshnjWwwEsDh&pid=Api\",\"medium\":\"https://tse1.mm.bing.net/th?id=OVP.F-xH3-wKfTjM3YMshnjWwwEsDh&pid=Api\",\"motion\":\"https://tse1.mm.bing.net/th?id=OM1.z-FRwxQ_NByK_A_1689135755&pid=Api\",\"small\":\"https://tse1.mm.bing.net/th?id=OVP.F-xH3-wKfTjM3YMshnjWwwEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-06-22T12:25:16.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":1100},\"title\":\"Dataiku for Generative AI: Real Applications, Real Safety\",\"uploader\":\"Dataiku\"},{\"content\":\"https://www.youtube.com/watch?v=-amc9iVauuE\",\"description\":\"Dataiku is the leading platform for Everyday AI, systemizing the use of data for exceptional business results. In today's video we will take a tour of Dataiku's end to end capabilities by exploring a real life use case around environmental impact. Let's take a look at how a data science team with different skills can work together to turn ...\",\"duration\":\"12:35\",\"embed_html\":\"\",\"embed_url\":\"http://www.youtube.com/embed/-amc9iVauuE?autoplay=1\",\"image_token\":\"2a05a65ad8a2727aa5c48b8daa7f9ec363a24d4336a3509016d4b200c9d003cd\",\"images\":{\"large\":\"https://tse1.mm.bing.net/th?id=OVP.Az9RhdSVwpXe56mGcs6FqQEsDh&pid=Api\",\"medium\":\"https://tse1.mm.bing.net/th?id=OVP.Az9RhdSVwpXe56mGcs6FqQEsDh&pid=Api\",\"motion\":\"https://tse1.mm.bing.net/th?id=OM1.Q2OhN9DzfowU6A_1685345657&pid=Api\",\"small\":\"https://tse1.mm.bing.net/th?id=OVP.Az9RhdSVwpXe56mGcs6FqQEsDh&pid=Api\"},\"provider\":\"Bing\",\"published\":\"2023-01-09T21:12:27.0000000\",\"publisher\":\"YouTube\",\"statistics\":{\"viewCount\":9768},\"title\":\"End to End Demo 2023\",\"uploader\":\"Dataiku\"}],\"vqd\":{\"Dataiku%20and%20DataRobot%20use%20cases\":\"4-60481969350525797892441552954401970387\"}});DDG.duckbar.loadModule('related_searches');if (DDG.pageLayout) DDG.pageLayout.initialize({\"mainline\":{\"items\":[[\"ad\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"videos\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"],[\"organic\"]]},\"sidebar\":{\"items\":[[\"organic\"]]}}, { start: 0 });DDG.deep.emit(\"load:completed\");" +} \ No newline at end of file diff --git a/tests/metagpt/actions/test_action_node.py b/tests/metagpt/actions/test_action_node.py index 384c4507b..53de9cc75 100644 --- a/tests/metagpt/actions/test_action_node.py +++ b/tests/metagpt/actions/test_action_node.py @@ -11,7 +11,7 @@ import pytest from pydantic import ValidationError from metagpt.actions import Action -from metagpt.actions.action_node import ActionNode +from metagpt.actions.action_node import ActionNode, ReviewMode, ReviseMode from metagpt.environment import Environment from metagpt.llm import LLM from metagpt.roles import Role @@ -23,14 +23,12 @@ from metagpt.team import Team async def test_debate_two_roles(): action1 = Action(name="AlexSay", instruction="Express your opinion with emotion and don't repeat it") action2 = Action(name="BobSay", instruction="Express your opinion with emotion and don't repeat it") - biden = Role( + alex = Role( name="Alex", profile="Democratic candidate", goal="Win the election", actions=[action1], watch=[action2] ) - trump = Role( - name="Bob", profile="Republican candidate", goal="Win the election", actions=[action2], watch=[action1] - ) + bob = Role(name="Bob", profile="Republican candidate", goal="Win the election", actions=[action2], watch=[action1]) env = Environment(desc="US election live broadcast") - team = Team(investment=10.0, env=env, roles=[biden, trump]) + team = Team(investment=10.0, env=env, roles=[alex, bob]) history = await team.run(idea="Topic: climate change. Under 80 words per message.", send_to="Alex", n_round=3) assert "Alex" in history @@ -39,9 +37,9 @@ async def test_debate_two_roles(): @pytest.mark.asyncio async def test_debate_one_role_in_env(): action = Action(name="Debate", instruction="Express your opinion with emotion and don't repeat it") - biden = Role(name="Alex", profile="Democratic candidate", goal="Win the election", actions=[action]) + alex = Role(name="Alex", profile="Democratic candidate", goal="Win the election", actions=[action]) env = Environment(desc="US election live broadcast") - team = Team(investment=10.0, env=env, roles=[biden]) + team = Team(investment=10.0, env=env, roles=[alex]) history = await team.run(idea="Topic: climate change. Under 80 words per message.", send_to="Alex", n_round=3) assert "Alex" in history @@ -49,8 +47,8 @@ async def test_debate_one_role_in_env(): @pytest.mark.asyncio async def test_debate_one_role(): action = Action(name="Debate", instruction="Express your opinion with emotion and don't repeat it") - biden = Role(name="Alex", profile="Democratic candidate", goal="Win the election", actions=[action]) - msg: Message = await biden.run("Topic: climate change. Under 80 words per message.") + alex = Role(name="Alex", profile="Democratic candidate", goal="Win the election", actions=[action]) + msg: Message = await alex.run("Topic: climate change. Under 80 words per message.") assert len(msg.content) > 10 assert msg.sent_from == "metagpt.roles.role.Role" @@ -98,6 +96,83 @@ async def test_action_node_two_layer(): assert "579" in answer2.content +@pytest.mark.asyncio +async def test_action_node_review(): + key = "Project Name" + node_a = ActionNode( + key=key, + expected_type=str, + instruction='According to the content of "Original Requirements," name the project using snake case style ' + "with underline, like 'game_2048' or 'simple_crm.", + example="game_2048", + ) + + with pytest.raises(RuntimeError): + _ = await node_a.review() + + _ = await node_a.fill(context=None, llm=LLM()) + setattr(node_a.instruct_content, key, "game snake") # wrong content to review + + review_comments = await node_a.review(review_mode=ReviewMode.AUTO) + assert len(review_comments) == 1 + assert list(review_comments.keys())[0] == key + + review_comments = await node_a.review(strgy="complex", review_mode=ReviewMode.AUTO) + assert len(review_comments) == 0 + + node = ActionNode.from_children(key="WritePRD", nodes=[node_a]) + with pytest.raises(RuntimeError): + _ = await node.review() + + _ = await node.fill(context=None, llm=LLM()) + + review_comments = await node.review(review_mode=ReviewMode.AUTO) + assert len(review_comments) == 1 + assert list(review_comments.keys())[0] == key + + review_comments = await node.review(strgy="complex", review_mode=ReviewMode.AUTO) + assert len(review_comments) == 1 + assert list(review_comments.keys())[0] == key + + +@pytest.mark.asyncio +async def test_action_node_revise(): + key = "Project Name" + node_a = ActionNode( + key=key, + expected_type=str, + instruction='According to the content of "Original Requirements," name the project using snake case style ' + "with underline, like 'game_2048' or 'simple_crm.", + example="game_2048", + ) + + with pytest.raises(RuntimeError): + _ = await node_a.review() + + _ = await node_a.fill(context=None, llm=LLM()) + setattr(node_a.instruct_content, key, "game snake") # wrong content to revise + revise_contents = await node_a.revise(revise_mode=ReviseMode.AUTO) + assert len(revise_contents) == 1 + assert "game_snake" in getattr(node_a.instruct_content, key) + + revise_contents = await node_a.revise(strgy="complex", revise_mode=ReviseMode.AUTO) + assert len(revise_contents) == 0 + + node = ActionNode.from_children(key="WritePRD", nodes=[node_a]) + with pytest.raises(RuntimeError): + _ = await node.revise() + + _ = await node.fill(context=None, llm=LLM()) + setattr(node.instruct_content, key, "game snake") + revise_contents = await node.revise(revise_mode=ReviseMode.AUTO) + assert len(revise_contents) == 1 + assert "game_snake" in getattr(node.instruct_content, key) + + revise_contents = await node.revise(strgy="complex", revise_mode=ReviseMode.AUTO) + assert len(revise_contents) == 1 + assert "game_snake" in getattr(node.instruct_content, key) + + t_dict = { "Required Python third-party packages": '"""\nflask==1.1.2\npygame==2.0.1\n"""\n', "Required Other language third-party packages": '"""\nNo third-party packages required for other languages.\n"""\n', @@ -138,10 +213,10 @@ def test_create_model_class(): assert test_class.__name__ == "test_class" output = test_class(**t_dict) - print(output.schema()) - assert output.schema()["title"] == "test_class" - assert output.schema()["type"] == "object" - assert output.schema()["properties"]["Full API spec"] + print(output.model_json_schema()) + assert output.model_json_schema()["title"] == "test_class" + assert output.model_json_schema()["type"] == "object" + assert output.model_json_schema()["properties"]["Full API spec"] def test_create_model_class_with_fields_unrecognized(): diff --git a/tests/metagpt/actions/test_action_outcls_registry.py b/tests/metagpt/actions/test_action_outcls_registry.py new file mode 100644 index 000000000..eac0ba4d9 --- /dev/null +++ b/tests/metagpt/actions/test_action_outcls_registry.py @@ -0,0 +1,46 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +# @Desc : unittest of action_outcls_registry + +from typing import List + +from metagpt.actions.action_node import ActionNode + + +def test_action_outcls_registry(): + class_name = "test" + out_mapping = {"field": (list[str], ...), "field1": (str, ...)} + out_data = {"field": ["field value1", "field value2"], "field1": "field1 value1"} + + outcls = ActionNode.create_model_class(class_name, mapping=out_mapping) + outinst = outcls(**out_data) + + outcls1 = ActionNode.create_model_class(class_name=class_name, mapping=out_mapping) + outinst1 = outcls1(**out_data) + assert outinst1 == outinst + + outcls2 = ActionNode(key="", expected_type=str, instruction="", example="").create_model_class( + class_name, out_mapping + ) + outinst2 = outcls2(**out_data) + assert outinst2 == outinst + + out_mapping = {"field1": (str, ...), "field": (list[str], ...)} # different order + outcls3 = ActionNode.create_model_class(class_name=class_name, mapping=out_mapping) + outinst3 = outcls3(**out_data) + assert outinst3 == outinst + + out_mapping2 = {"field1": (str, ...), "field": (List[str], ...)} # typing case + outcls4 = ActionNode.create_model_class(class_name=class_name, mapping=out_mapping2) + outinst4 = outcls4(**out_data) + assert outinst4 == outinst + + out_data2 = {"field2": ["field2 value1", "field2 value2"], "field1": "field1 value1"} + out_mapping = {"field1": (str, ...), "field2": (List[str], ...)} # List first + outcls5 = ActionNode.create_model_class(class_name, out_mapping) + outinst5 = outcls5(**out_data2) + + out_mapping = {"field1": (str, ...), "field2": (list[str], ...)} + outcls6 = ActionNode.create_model_class(class_name, out_mapping) + outinst6 = outcls6(**out_data2) + assert outinst5 == outinst6 diff --git a/tests/metagpt/actions/test_debug_code.py b/tests/metagpt/actions/test_debug_code.py index 262f2e60d..83ce75761 100644 --- a/tests/metagpt/actions/test_debug_code.py +++ b/tests/metagpt/actions/test_debug_code.py @@ -48,7 +48,7 @@ def sort_array(arr): async def test_debug_code(): debug_context = Message(content=DebugContext) new_code = await DebugCode().run(context=debug_context, code=CODE, runtime_result=ErrorStr) - assert "def sort_array(arr)" in new_code + assert "def sort_array(arr)" in new_code["code"] def test_messages_to_str(): diff --git a/tests/metagpt/actions/test_debug_error.py b/tests/metagpt/actions/test_debug_error.py index e512c370a..c88818bbd 100644 --- a/tests/metagpt/actions/test_debug_error.py +++ b/tests/metagpt/actions/test_debug_error.py @@ -11,10 +11,7 @@ import uuid import pytest from metagpt.actions.debug_error import DebugError -from metagpt.config import CONFIG -from metagpt.const import TEST_CODES_FILE_REPO, TEST_OUTPUTS_FILE_REPO from metagpt.schema import RunCodeContext, RunCodeResult -from metagpt.utils.file_repository import FileRepository CODE_CONTENT = ''' from typing import List @@ -117,8 +114,8 @@ if __name__ == '__main__': @pytest.mark.asyncio -async def test_debug_error(): - CONFIG.src_workspace = CONFIG.git_repo.workdir / uuid.uuid4().hex +async def test_debug_error(context): + context.src_workspace = context.git_repo.workdir / uuid.uuid4().hex ctx = RunCodeContext( code_filename="player.py", test_filename="test_player.py", @@ -126,8 +123,8 @@ async def test_debug_error(): output_filename="output.log", ) - await FileRepository.save_file(filename=ctx.code_filename, content=CODE_CONTENT, relative_path=CONFIG.src_workspace) - await FileRepository.save_file(filename=ctx.test_filename, content=TEST_CONTENT, relative_path=TEST_CODES_FILE_REPO) + await context.repo.with_src_path(context.src_workspace).srcs.save(filename=ctx.code_filename, content=CODE_CONTENT) + await context.repo.tests.save(filename=ctx.test_filename, content=TEST_CONTENT) output_data = RunCodeResult( stdout=";", stderr="", @@ -141,24 +138,11 @@ async def test_debug_error(): "----------------------------------------------------------------------\n" "Ran 5 tests in 0.007s\n\nFAILED (failures=1)\n;\n", ) - await FileRepository.save_file( - filename=ctx.output_filename, content=output_data.model_dump_json(), relative_path=TEST_OUTPUTS_FILE_REPO - ) - debug_error = DebugError(context=ctx) + await context.repo.test_outputs.save(filename=ctx.output_filename, content=output_data.model_dump_json()) + debug_error = DebugError(i_context=ctx, context=context) rsp = await debug_error.run() assert "class Player" in rsp # rewrite the same class - # Problematic code: - # ``` - # if self.score > 21 and any(card.rank == 'A' for card in self.hand): - # self.score -= 10 - # ``` - # Should rewrite to (used "gpt-3.5-turbo-1106"): - # ``` - # ace_count = sum(1 for card in self.hand if card.rank == 'A') - # while self.score > 21 and ace_count > 0: - # self.score -= 10 - # ace_count -= 1 - # ``` - assert "while self.score > 21" in rsp + # a key logic to rewrite to (original one is "if self.score > 12") + assert "self.score" in rsp diff --git a/tests/metagpt/actions/test_design_api.py b/tests/metagpt/actions/test_design_api.py index 8d4720570..7d3efa7ff 100644 --- a/tests/metagpt/actions/test_design_api.py +++ b/tests/metagpt/actions/test_design_api.py @@ -9,20 +9,17 @@ import pytest from metagpt.actions.design_api import WriteDesign -from metagpt.const import PRDS_FILE_REPO from metagpt.logs import logger from metagpt.schema import Message -from metagpt.utils.file_repository import FileRepository -from tests.metagpt.actions.mock_markdown import PRD_SAMPLE @pytest.mark.asyncio -async def test_design_api(): - inputs = ["我们需要一个音乐播放器，它应该有播放、暂停、上一曲、下一曲等功能。", PRD_SAMPLE] +async def test_design_api(context): + inputs = ["我们需要一个音乐播放器，它应该有播放、暂停、上一曲、下一曲等功能。"] # PRD_SAMPLE for prd in inputs: - await FileRepository.save_file("new_prd.txt", content=prd, relative_path=PRDS_FILE_REPO) + await context.repo.docs.prd.save(filename="new_prd.txt", content=prd) - design_api = WriteDesign() + design_api = WriteDesign(context=context) result = await design_api.run(Message(content=prd, instruct_content=None)) logger.info(result) diff --git a/tests/metagpt/actions/test_design_api_an.py b/tests/metagpt/actions/test_design_api_an.py new file mode 100644 index 000000000..3d11f200d --- /dev/null +++ b/tests/metagpt/actions/test_design_api_an.py @@ -0,0 +1,46 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/01/03 +@Author : mannaandpoem +@File : test_design_api_an.py +""" +import pytest +from openai._models import BaseModel + +from metagpt.actions.action_node import ActionNode, dict_to_markdown +from metagpt.actions.design_api import NEW_REQ_TEMPLATE +from metagpt.actions.design_api_an import REFINED_DESIGN_NODE +from metagpt.llm import LLM +from tests.data.incremental_dev_project.mock import ( + DESIGN_SAMPLE, + REFINED_DESIGN_JSON, + REFINED_PRD_JSON, +) + + +@pytest.fixture() +def llm(): + return LLM() + + +def mock_refined_design_json(): + return REFINED_DESIGN_JSON + + +@pytest.mark.asyncio +async def test_write_design_an(mocker): + root = ActionNode.from_children( + "RefinedDesignAPI", [ActionNode(key="", expected_type=str, instruction="", example="")] + ) + root.instruct_content = BaseModel() + root.instruct_content.model_dump = mock_refined_design_json + mocker.patch("metagpt.actions.design_api_an.REFINED_DESIGN_NODE.fill", return_value=root) + + prompt = NEW_REQ_TEMPLATE.format(old_design=DESIGN_SAMPLE, context=dict_to_markdown(REFINED_PRD_JSON)) + node = await REFINED_DESIGN_NODE.fill(prompt, llm) + + assert "Refined Implementation Approach" in node.instruct_content.model_dump() + assert "Refined File list" in node.instruct_content.model_dump() + assert "Refined Data structures and interfaces" in node.instruct_content.model_dump() + assert "Refined Program call flow" in node.instruct_content.model_dump() diff --git a/tests/metagpt/actions/test_design_api_review.py b/tests/metagpt/actions/test_design_api_review.py index cfc29056f..a648dba3f 100644 --- a/tests/metagpt/actions/test_design_api_review.py +++ b/tests/metagpt/actions/test_design_api_review.py @@ -11,7 +11,7 @@ from metagpt.actions.design_api_review import DesignReview @pytest.mark.asyncio -async def test_design_api_review(): +async def test_design_api_review(context): prd = "我们需要一个音乐播放器，它应该有播放、暂停、上一曲、下一曲等功能。" api_design = """ 数据结构: @@ -26,7 +26,7 @@ API列表: """ _ = "API设计看起来非常合理，满足了PRD中的所有需求。" - design_api_review = DesignReview() + design_api_review = DesignReview(context=context) result = await design_api_review.run(prd, api_design) diff --git a/tests/metagpt/actions/test_fix_bug.py b/tests/metagpt/actions/test_fix_bug.py index b2dc8d0f4..cbd9d0b57 100644 --- a/tests/metagpt/actions/test_fix_bug.py +++ b/tests/metagpt/actions/test_fix_bug.py @@ -12,6 +12,6 @@ from metagpt.actions.fix_bug import FixBug @pytest.mark.asyncio -async def test_fix_bug(): - fix_bug = FixBug() +async def test_fix_bug(context): + fix_bug = FixBug(context=context) assert fix_bug.name == "FixBug" diff --git a/tests/metagpt/actions/test_generate_questions.py b/tests/metagpt/actions/test_generate_questions.py index b7c9d3984..6adb2e617 100644 --- a/tests/metagpt/actions/test_generate_questions.py +++ b/tests/metagpt/actions/test_generate_questions.py @@ -10,7 +10,7 @@ import pytest from metagpt.actions.generate_questions import GenerateQuestions from metagpt.logs import logger -context = """ +msg = """ ## topic 如何做一个生日蛋糕 @@ -20,9 +20,9 @@ context = """ @pytest.mark.asyncio -async def test_generate_questions(): - action = GenerateQuestions() - rsp = await action.run(context) +async def test_generate_questions(context): + action = GenerateQuestions(context=context) + rsp = await action.run(msg) logger.info(f"{rsp.content=}") assert "Questions" in rsp.content diff --git a/tests/metagpt/actions/test_invoice_ocr.py b/tests/metagpt/actions/test_invoice_ocr.py index b4560f61b..4df0cf4f8 100644 --- a/tests/metagpt/actions/test_invoice_ocr.py +++ b/tests/metagpt/actions/test_invoice_ocr.py @@ -23,9 +23,9 @@ from metagpt.const import TEST_DATA_PATH Path("invoices/invoice-4.zip"), ], ) -async def test_invoice_ocr(invoice_path: Path): +async def test_invoice_ocr(invoice_path: Path, context): invoice_path = TEST_DATA_PATH / invoice_path - resp = await InvoiceOCR().run(file_path=Path(invoice_path)) + resp = await InvoiceOCR(context=context).run(file_path=Path(invoice_path)) assert isinstance(resp, list) diff --git a/tests/metagpt/actions/test_make_tools.py b/tests/metagpt/actions/test_make_tools.py deleted file mode 100644 index 8e94c6eee..000000000 --- a/tests/metagpt/actions/test_make_tools.py +++ /dev/null @@ -1,52 +0,0 @@ -import pytest - -from metagpt.actions.execute_code import ExecutePyCode -from metagpt.actions.write_analysis_code import MakeTools -from metagpt.logs import logger - - -@pytest.mark.asyncio -async def test_make_tools(): - code = "import yfinance as yf\n\n# Collect Alibaba stock data\nalibaba = yf.Ticker('BABA')\ndata = alibaba.history(period='1d', start='2022-01-01', end='2022-12-31')\nprint(data.head())" - msgs = [{"role": "assistant", "content": code}] - mt = MakeTools() - tool_code = await mt.run(msgs) - logger.debug(tool_code) - ep = ExecutePyCode() - tool_code = "!pip install yfinance\n" + tool_code - result, res_type = await ep.run(tool_code) - assert res_type is True - logger.debug(result) - - -@pytest.mark.asyncio -async def test_make_tools2(): - code = """import pandas as pd\npath = "./tests/data/test.csv"\ndf = pd.read_csv(path)\ndata = df.copy()\n - data['started_at'] = data['started_at'].apply(lambda r: pd.to_datetime(r))\n - data['ended_at'] = data['ended_at'].apply(lambda r: pd.to_datetime(r))\ndata.head()""" - msgs = [{"role": "assistant", "content": code}] - mt = MakeTools() - tool_code = await mt.run(msgs) - logger.debug(tool_code) - ep = ExecutePyCode() - tool_code = tool_code - result, res_type = await ep.run(tool_code) - assert res_type is True - logger.debug(result) - - -@pytest.mark.asyncio -async def test_make_tools3(): - code = """import pandas as pd\npath = "./tests/data/test.csv"\ndf = pd.read_csv(path)\ndata = df.copy()\n - data['started_at'] = data['started_at'].apply(lambda r: pd.to_datetime(r))\n - data['ended_at'] = data['ended_at'].apply(lambda r: pd.to_datetime(r))\n - data['duration_hour'] = (data['ended_at'] - data['started_at']).dt.seconds/3600\ndata.head()""" - msgs = [{"role": "assistant", "content": code}] - mt = MakeTools() - tool_code = await mt.run(msgs) - logger.debug(tool_code) - ep = ExecutePyCode() - tool_code = tool_code - result, res_type = await ep.run(tool_code) - assert res_type is True - logger.debug(result) diff --git a/tests/metagpt/actions/test_ml_action.py b/tests/metagpt/actions/test_ml_action.py new file mode 100644 index 000000000..2c8d34da8 --- /dev/null +++ b/tests/metagpt/actions/test_ml_action.py @@ -0,0 +1,46 @@ +import pytest + +from metagpt.actions.ml_action import WriteCodeWithToolsML +from metagpt.schema import Plan, Task + + +@pytest.mark.asyncio +async def test_write_code_with_tools(): + write_code_ml = WriteCodeWithToolsML() + + task_map = { + "1": Task( + task_id="1", + instruction="随机生成一个pandas DataFrame数据集", + task_type="other", + dependent_task_ids=[], + code=""" + import pandas as pd + df = pd.DataFrame({ + 'a': [1, 2, 3, 4, 5], + 'b': [1.1, 2.2, 3.3, 4.4, np.nan], + 'c': ['aa', 'bb', 'cc', 'dd', 'ee'], + 'd': [1, 2, 3, 4, 5] + }) + """, + is_finished=True, + ), + "2": Task( + task_id="2", + instruction="对数据集进行数据清洗", + task_type="data_preprocess", + dependent_task_ids=["1"], + ), + } + plan = Plan( + goal="构造数据集并进行数据清洗", + tasks=list(task_map.values()), + task_map=task_map, + current_task_id="2", + ) + column_info = "" + + _, code_with_ml = await write_code_ml.run([], plan, column_info) + code_with_ml = code_with_ml["code"] + assert len(code_with_ml) > 0 + print(code_with_ml) diff --git a/tests/metagpt/actions/test_prepare_documents.py b/tests/metagpt/actions/test_prepare_documents.py index 31c8bcb80..7ad0dee4e 100644 --- a/tests/metagpt/actions/test_prepare_documents.py +++ b/tests/metagpt/actions/test_prepare_documents.py @@ -9,22 +9,19 @@ import pytest from metagpt.actions.prepare_documents import PrepareDocuments -from metagpt.config import CONFIG -from metagpt.const import DOCS_FILE_REPO, REQUIREMENT_FILENAME +from metagpt.const import REQUIREMENT_FILENAME +from metagpt.context import Context from metagpt.schema import Message -from metagpt.utils.file_repository import FileRepository @pytest.mark.asyncio async def test_prepare_documents(): msg = Message(content="New user requirements balabala...") + context = Context() - if CONFIG.git_repo: - CONFIG.git_repo.delete_repository() - CONFIG.git_repo = None - - await PrepareDocuments().run(with_messages=[msg]) - assert CONFIG.git_repo - doc = await FileRepository.get_file(filename=REQUIREMENT_FILENAME, relative_path=DOCS_FILE_REPO) + await PrepareDocuments(context=context).run(with_messages=[msg]) + assert context.git_repo + assert context.repo + doc = await context.repo.docs.get(filename=REQUIREMENT_FILENAME) assert doc assert doc.content == msg.content diff --git a/tests/metagpt/actions/test_prepare_interview.py b/tests/metagpt/actions/test_prepare_interview.py index cd0c850ed..111f24d5f 100644 --- a/tests/metagpt/actions/test_prepare_interview.py +++ b/tests/metagpt/actions/test_prepare_interview.py @@ -12,8 +12,8 @@ from metagpt.logs import logger @pytest.mark.asyncio -async def test_prepare_interview(): - action = PrepareInterview() +async def test_prepare_interview(context): + action = PrepareInterview(context=context) rsp = await action.run("I just graduated and hope to find a job as a Python engineer") logger.info(f"{rsp.content=}") diff --git a/tests/metagpt/actions/test_project_management.py b/tests/metagpt/actions/test_project_management.py index 88263ff29..f3bb405ca 100644 --- a/tests/metagpt/actions/test_project_management.py +++ b/tests/metagpt/actions/test_project_management.py @@ -9,21 +9,18 @@ import pytest from metagpt.actions.project_management import WriteTasks -from metagpt.config import CONFIG -from metagpt.const import PRDS_FILE_REPO, SYSTEM_DESIGN_FILE_REPO from metagpt.logs import logger from metagpt.schema import Message -from metagpt.utils.file_repository import FileRepository from tests.metagpt.actions.mock_json import DESIGN, PRD @pytest.mark.asyncio -async def test_design_api(): - await FileRepository.save_file("1.txt", content=str(PRD), relative_path=PRDS_FILE_REPO) - await FileRepository.save_file("1.txt", content=str(DESIGN), relative_path=SYSTEM_DESIGN_FILE_REPO) - logger.info(CONFIG.git_repo) +async def test_design_api(context): + await context.repo.docs.prd.save("1.txt", content=str(PRD)) + await context.repo.docs.system_design.save("1.txt", content=str(DESIGN)) + logger.info(context.git_repo) - action = WriteTasks() + action = WriteTasks(context=context) result = await action.run(Message(content="", instruct_content=None)) logger.info(result) diff --git a/tests/metagpt/actions/test_project_management_an.py b/tests/metagpt/actions/test_project_management_an.py new file mode 100644 index 000000000..ddbb56569 --- /dev/null +++ b/tests/metagpt/actions/test_project_management_an.py @@ -0,0 +1,45 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/01/03 +@Author : mannaandpoem +@File : test_project_management_an.py +""" +import pytest +from openai._models import BaseModel + +from metagpt.actions.action_node import ActionNode, dict_to_markdown +from metagpt.actions.project_management import NEW_REQ_TEMPLATE +from metagpt.actions.project_management_an import REFINED_PM_NODE +from metagpt.llm import LLM +from tests.data.incremental_dev_project.mock import ( + REFINED_DESIGN_JSON, + REFINED_TASKS_JSON, + TASKS_SAMPLE, +) + + +@pytest.fixture() +def llm(): + return LLM() + + +def mock_refined_tasks_json(): + return REFINED_TASKS_JSON + + +@pytest.mark.asyncio +async def test_project_management_an(mocker): + root = ActionNode.from_children( + "RefinedProjectManagement", [ActionNode(key="", expected_type=str, instruction="", example="")] + ) + root.instruct_content = BaseModel() + root.instruct_content.model_dump = mock_refined_tasks_json + mocker.patch("metagpt.actions.project_management_an.REFINED_PM_NODE.fill", return_value=root) + + prompt = NEW_REQ_TEMPLATE.format(old_task=TASKS_SAMPLE, context=dict_to_markdown(REFINED_DESIGN_JSON)) + node = await REFINED_PM_NODE.fill(prompt, llm) + + assert "Refined Logic Analysis" in node.instruct_content.model_dump() + assert "Refined Task list" in node.instruct_content.model_dump() + assert "Refined Shared Knowledge" in node.instruct_content.model_dump() diff --git a/tests/metagpt/actions/test_rebuild_class_view.py b/tests/metagpt/actions/test_rebuild_class_view.py index 207ba4be1..04b7d91fc 100644 --- a/tests/metagpt/actions/test_rebuild_class_view.py +++ b/tests/metagpt/actions/test_rebuild_class_view.py @@ -11,19 +11,19 @@ from pathlib import Path import pytest from metagpt.actions.rebuild_class_view import RebuildClassView -from metagpt.config import CONFIG -from metagpt.const import GRAPH_REPO_FILE_REPO from metagpt.llm import LLM @pytest.mark.asyncio -async def test_rebuild(): +async def test_rebuild(context): action = RebuildClassView( - name="RedBean", context=str(Path(__file__).parent.parent.parent.parent / "metagpt"), llm=LLM() + name="RedBean", + i_context=str(Path(__file__).parent.parent.parent.parent / "metagpt"), + llm=LLM(), + context=context, ) await action.run() - graph_file_repo = CONFIG.git_repo.new_file_repository(relative_path=GRAPH_REPO_FILE_REPO) - assert graph_file_repo.changed_files + assert context.repo.docs.graph_repo.changed_files @pytest.mark.parametrize( diff --git a/tests/metagpt/actions/test_rebuild_sequence_view.py b/tests/metagpt/actions/test_rebuild_sequence_view.py index 939412fe7..e0e65c3cc 100644 --- a/tests/metagpt/actions/test_rebuild_sequence_view.py +++ b/tests/metagpt/actions/test_rebuild_sequence_view.py @@ -10,33 +10,30 @@ from pathlib import Path import pytest from metagpt.actions.rebuild_sequence_view import RebuildSequenceView -from metagpt.config import CONFIG from metagpt.const import GRAPH_REPO_FILE_REPO from metagpt.llm import LLM from metagpt.utils.common import aread -from metagpt.utils.file_repository import FileRepository from metagpt.utils.git_repository import ChangeType @pytest.mark.asyncio -async def test_rebuild(): +@pytest.mark.skip +async def test_rebuild(context): # Mock data = await aread(filename=Path(__file__).parent / "../../data/graph_db/networkx.json") - graph_db_filename = Path(CONFIG.git_repo.workdir.name).with_suffix(".json") - await FileRepository.save_file( - filename=str(graph_db_filename), - relative_path=GRAPH_REPO_FILE_REPO, - content=data, - ) - CONFIG.git_repo.add_change({f"{GRAPH_REPO_FILE_REPO}/{graph_db_filename}": ChangeType.UNTRACTED}) - CONFIG.git_repo.commit("commit1") + graph_db_filename = Path(context.repo.workdir.name).with_suffix(".json") + await context.repo.docs.graph_repo.save(filename=str(graph_db_filename), content=data) + context.git_repo.add_change({f"{GRAPH_REPO_FILE_REPO}/{graph_db_filename}": ChangeType.UNTRACTED}) + context.git_repo.commit("commit1") action = RebuildSequenceView( - name="RedBean", context=str(Path(__file__).parent.parent.parent.parent / "metagpt"), llm=LLM() + name="RedBean", + i_context=str(Path(__file__).parent.parent.parent.parent / "metagpt"), + llm=LLM(), + context=context, ) await action.run() - graph_file_repo = CONFIG.git_repo.new_file_repository(relative_path=GRAPH_REPO_FILE_REPO) - assert graph_file_repo.changed_files + assert context.repo.docs.graph_repo.changed_files @pytest.mark.parametrize( diff --git a/tests/metagpt/actions/test_research.py b/tests/metagpt/actions/test_research.py index dfbcce4ae..372a1e876 100644 --- a/tests/metagpt/actions/test_research.py +++ b/tests/metagpt/actions/test_research.py @@ -9,10 +9,12 @@ import pytest from metagpt.actions import research +from metagpt.tools import SearchEngineType +from metagpt.tools.search_engine import SearchEngine @pytest.mark.asyncio -async def test_collect_links(mocker): +async def test_collect_links(mocker, search_engine_mocker, context): async def mock_llm_ask(self, prompt: str, system_msgs): if "Please provide up to 2 necessary keywords" in prompt: return '["metagpt", "llm"]' @@ -26,13 +28,15 @@ async def test_collect_links(mocker): return "[1,2]" mocker.patch("metagpt.provider.base_llm.BaseLLM.aask", mock_llm_ask) - resp = await research.CollectLinks().run("The application of MetaGPT") + resp = await research.CollectLinks(search_engine=SearchEngine(SearchEngineType.DUCK_DUCK_GO), context=context).run( + "The application of MetaGPT" + ) for i in ["MetaGPT use cases", "The roadmap of MetaGPT", "The function of MetaGPT", "What llm MetaGPT support"]: assert i in resp @pytest.mark.asyncio -async def test_collect_links_with_rank_func(mocker): +async def test_collect_links_with_rank_func(mocker, search_engine_mocker, context): rank_before = [] rank_after = [] url_per_query = 4 @@ -45,14 +49,16 @@ async def test_collect_links_with_rank_func(mocker): return results mocker.patch("metagpt.provider.base_llm.BaseLLM.aask", mock_collect_links_llm_ask) - resp = await research.CollectLinks(rank_func=rank_func).run("The application of MetaGPT") + resp = await research.CollectLinks( + search_engine=SearchEngine(SearchEngineType.DUCK_DUCK_GO), rank_func=rank_func, context=context + ).run("The application of MetaGPT") for x, y, z in zip(rank_before, rank_after, resp.values()): assert x[::-1] == y assert [i["link"] for i in y] == z @pytest.mark.asyncio -async def test_web_browse_and_summarize(mocker): +async def test_web_browse_and_summarize(mocker, context): async def mock_llm_ask(*args, **kwargs): return "metagpt" @@ -60,20 +66,20 @@ async def test_web_browse_and_summarize(mocker): url = "https://github.com/geekan/MetaGPT" url2 = "https://github.com/trending" query = "What's new in metagpt" - resp = await research.WebBrowseAndSummarize().run(url, query=query) + resp = await research.WebBrowseAndSummarize(context=context).run(url, query=query) assert len(resp) == 1 assert url in resp assert resp[url] == "metagpt" - resp = await research.WebBrowseAndSummarize().run(url, url2, query=query) + resp = await research.WebBrowseAndSummarize(context=context).run(url, url2, query=query) assert len(resp) == 2 async def mock_llm_ask(*args, **kwargs): return "Not relevant." mocker.patch("metagpt.provider.base_llm.BaseLLM.aask", mock_llm_ask) - resp = await research.WebBrowseAndSummarize().run(url, query=query) + resp = await research.WebBrowseAndSummarize(context=context).run(url, query=query) assert len(resp) == 1 assert url in resp @@ -81,7 +87,7 @@ async def test_web_browse_and_summarize(mocker): @pytest.mark.asyncio -async def test_conduct_research(mocker): +async def test_conduct_research(mocker, context): data = None async def mock_llm_ask(*args, **kwargs): @@ -95,7 +101,7 @@ async def test_conduct_research(mocker): "outputs user stories / competitive analysis / requirements / data structures / APIs / documents, etc." ) - resp = await research.ConductResearch().run("The application of MetaGPT", content) + resp = await research.ConductResearch(context=context).run("The application of MetaGPT", content) assert resp == data diff --git a/tests/metagpt/actions/test_run_code.py b/tests/metagpt/actions/test_run_code.py index ad08b5738..afd308da7 100644 --- a/tests/metagpt/actions/test_run_code.py +++ b/tests/metagpt/actions/test_run_code.py @@ -24,19 +24,19 @@ async def test_run_text(): @pytest.mark.asyncio -async def test_run_script(): +async def test_run_script(context): # Successful command - out, err = await RunCode.run_script(".", command=["echo", "Hello World"]) + out, err = await RunCode(context=context).run_script(".", command=["echo", "Hello World"]) assert out.strip() == "Hello World" assert err == "" # Unsuccessful command - out, err = await RunCode.run_script(".", command=["python", "-c", "print(1/0)"]) + out, err = await RunCode(context=context).run_script(".", command=["python", "-c", "print(1/0)"]) assert "ZeroDivisionError" in err @pytest.mark.asyncio -async def test_run(): +async def test_run(context): inputs = [ (RunCodeContext(mode="text", code_filename="a.txt", code="print('Hello, World')"), "PASS"), ( @@ -61,5 +61,5 @@ async def test_run(): ), ] for ctx, result in inputs: - rsp = await RunCode(context=ctx).run() + rsp = await RunCode(i_context=ctx, context=context).run() assert result in rsp.summary diff --git a/tests/metagpt/actions/test_skill_action.py b/tests/metagpt/actions/test_skill_action.py index 69cd8129d..2ebe79b30 100644 --- a/tests/metagpt/actions/test_skill_action.py +++ b/tests/metagpt/actions/test_skill_action.py @@ -47,18 +47,18 @@ class TestSkillAction: assert args.get("size_type") == "512x512" @pytest.mark.asyncio - async def test_parser_action(self, mocker): + async def test_parser_action(self, mocker, context): # mock mocker.patch("metagpt.learn.text_to_image", return_value="https://mock.com/xxx") - parser_action = ArgumentsParingAction(skill=self.skill, ask="Draw an apple") + parser_action = ArgumentsParingAction(skill=self.skill, ask="Draw an apple", context=context) rsp = await parser_action.run() assert rsp assert parser_action.args assert parser_action.args.get("text") == "Draw an apple" assert parser_action.args.get("size_type") == "512x512" - action = SkillAction(skill=self.skill, args=parser_action.args) + action = SkillAction(skill=self.skill, args=parser_action.args, context=context) rsp = await action.run() assert rsp assert "image/png;base64," in rsp.content or "http" in rsp.content @@ -81,8 +81,8 @@ class TestSkillAction: await SkillAction.find_and_call_function("dummy_call", {"a": 1}) @pytest.mark.asyncio - async def test_skill_action_error(self): - action = SkillAction(skill=self.skill, args={}) + async def test_skill_action_error(self, context): + action = SkillAction(skill=self.skill, args={}, context=context) rsp = await action.run() assert "Error" in rsp.content diff --git a/tests/metagpt/actions/test_summarize_code.py b/tests/metagpt/actions/test_summarize_code.py index 7ecb67afd..a404047c1 100644 --- a/tests/metagpt/actions/test_summarize_code.py +++ b/tests/metagpt/actions/test_summarize_code.py @@ -6,14 +6,14 @@ @File : test_summarize_code.py @Modifiled By: mashenquan, 2023-12-6. Unit test for summarize_code.py """ +import uuid +from pathlib import Path + import pytest from metagpt.actions.summarize_code import SummarizeCode -from metagpt.config import CONFIG -from metagpt.const import SYSTEM_DESIGN_FILE_REPO, TASK_FILE_REPO from metagpt.logs import logger from metagpt.schema import CodeSummarizeContext -from metagpt.utils.file_repository import FileRepository DESIGN_CONTENT = """ {"Implementation approach": "To develop this snake game, we will use the Python language and choose the Pygame library. Pygame is an open-source Python module collection specifically designed for writing video games. It provides functionalities such as displaying images and playing sounds, making it suitable for creating intuitive and responsive user interfaces. We will ensure efficient game logic to prevent any delays during gameplay. The scoring system will be simple, with the snake gaining points for each food it eats. We will use Pygame's event handling system to implement pause and resume functionality, as well as high-score tracking. The difficulty will increase by speeding up the snake's movement. In the initial version, we will focus on single-player mode and consider adding multiplayer mode and customizable skins in future updates. Based on the new requirement, we will also add a moving obstacle that appears randomly. If the snake eats this obstacle, the game will end. If the snake does not eat the obstacle, it will disappear after 5 seconds. For this, we need to add mechanisms for obstacle generation, movement, and disappearance in the game logic.", "Project_name": "snake_game", "File list": ["main.py", "game.py", "snake.py", "food.py", "obstacle.py", "scoreboard.py", "constants.py", "assets/styles.css", "assets/index.html"], "Data structures and interfaces": "```mermaid\n classDiagram\n class Game{\n +int score\n +int speed\n +bool game_over\n +bool paused\n +Snake snake\n +Food food\n +Obstacle obstacle\n +Scoreboard scoreboard\n +start_game() void\n +pause_game() void\n +resume_game() void\n +end_game() void\n +increase_difficulty() void\n +update() void\n +render() void\n Game()\n }\n class Snake{\n +list body_parts\n +str direction\n +bool grow\n +move() void\n +grow() void\n +check_collision() bool\n Snake()\n }\n class Food{\n +tuple position\n +spawn() void\n Food()\n }\n class Obstacle{\n +tuple position\n +int lifetime\n +bool active\n +spawn() void\n +move() void\n +check_collision() bool\n +disappear() void\n Obstacle()\n }\n class Scoreboard{\n +int high_score\n +update_score(int) void\n +reset_score() void\n +load_high_score() void\n +save_high_score() void\n Scoreboard()\n }\n class Constants{\n }\n Game \"1\" -- \"1\" Snake: has\n Game \"1\" -- \"1\" Food: has\n Game \"1\" -- \"1\" Obstacle: has\n Game \"1\" -- \"1\" Scoreboard: has\n ```", "Program call flow": "```sequenceDiagram\n participant M as Main\n participant G as Game\n participant S as Snake\n participant F as Food\n participant O as Obstacle\n participant SB as Scoreboard\n M->>G: start_game()\n loop game loop\n G->>S: move()\n G->>S: check_collision()\n G->>F: spawn()\n G->>O: spawn()\n G->>O: move()\n G->>O: check_collision()\n G->>O: disappear()\n G->>SB: update_score(score)\n G->>G: update()\n G->>G: render()\n alt if paused\n M->>G: pause_game()\n M->>G: resume_game()\n end\n alt if game_over\n G->>M: end_game()\n end\n end\n```", "Anything UNCLEAR": "There is no need for further clarification as the requirements are already clear."} @@ -177,19 +177,28 @@ class Snake: @pytest.mark.asyncio -async def test_summarize_code(): - CONFIG.src_workspace = CONFIG.git_repo.workdir / "src" - await FileRepository.save_file(filename="1.json", relative_path=SYSTEM_DESIGN_FILE_REPO, content=DESIGN_CONTENT) - await FileRepository.save_file(filename="1.json", relative_path=TASK_FILE_REPO, content=TASK_CONTENT) - await FileRepository.save_file(filename="food.py", relative_path=CONFIG.src_workspace, content=FOOD_PY) - await FileRepository.save_file(filename="game.py", relative_path=CONFIG.src_workspace, content=GAME_PY) - await FileRepository.save_file(filename="main.py", relative_path=CONFIG.src_workspace, content=MAIN_PY) - await FileRepository.save_file(filename="snake.py", relative_path=CONFIG.src_workspace, content=SNAKE_PY) +async def test_summarize_code(context): + git_dir = Path(__file__).parent / f"unittest/{uuid.uuid4().hex}" + git_dir.mkdir(parents=True, exist_ok=True) - src_file_repo = CONFIG.git_repo.new_file_repository(relative_path=CONFIG.src_workspace) - all_files = src_file_repo.all_files - ctx = CodeSummarizeContext(design_filename="1.json", task_filename="1.json", codes_filenames=all_files) - action = SummarizeCode(context=ctx) + context.src_workspace = context.git_repo.workdir / "src" + await context.repo.docs.system_design.save(filename="1.json", content=DESIGN_CONTENT) + await context.repo.docs.task.save(filename="1.json", content=TASK_CONTENT) + await context.repo.with_src_path(context.src_workspace).srcs.save(filename="food.py", content=FOOD_PY) + assert context.repo.srcs.workdir == context.src_workspace + await context.repo.srcs.save(filename="game.py", content=GAME_PY) + await context.repo.srcs.save(filename="main.py", content=MAIN_PY) + await context.repo.srcs.save(filename="snake.py", content=SNAKE_PY) + + all_files = context.repo.srcs.all_files + summarization_context = CodeSummarizeContext( + design_filename="1.json", task_filename="1.json", codes_filenames=all_files + ) + action = SummarizeCode(context=context, i_context=summarization_context) rsp = await action.run() assert rsp logger.info(rsp) + + +if __name__ == "__main__": + pytest.main([__file__, "-s"]) diff --git a/tests/metagpt/actions/test_talk_action.py b/tests/metagpt/actions/test_talk_action.py index 953fdf44a..206abfbae 100644 --- a/tests/metagpt/actions/test_talk_action.py +++ b/tests/metagpt/actions/test_talk_action.py @@ -9,13 +9,12 @@ import pytest from metagpt.actions.talk_action import TalkAction -from metagpt.config import CONFIG from metagpt.schema import Message @pytest.mark.asyncio @pytest.mark.parametrize( - ("agent_description", "language", "context", "knowledge", "history_summary"), + ("agent_description", "language", "talk_context", "knowledge", "history_summary"), [ ( "mathematician", @@ -33,12 +32,12 @@ from metagpt.schema import Message ), ], ) -async def test_prompt(agent_description, language, context, knowledge, history_summary): +async def test_prompt(agent_description, language, talk_context, knowledge, history_summary, context): # Prerequisites - CONFIG.agent_description = agent_description - CONFIG.language = language + context.kwargs.agent_description = agent_description + context.kwargs.language = language - action = TalkAction(context=context, knowledge=knowledge, history_summary=history_summary) + action = TalkAction(i_context=talk_context, knowledge=knowledge, history_summary=history_summary, context=context) assert "{" not in action.prompt assert "{" not in action.prompt_gpt4 diff --git a/tests/metagpt/actions/test_write_analysis_code.py b/tests/metagpt/actions/test_write_analysis_code.py index e64b4a551..43f23848d 100644 --- a/tests/metagpt/actions/test_write_analysis_code.py +++ b/tests/metagpt/actions/test_write_analysis_code.py @@ -3,16 +3,13 @@ import asyncio import pytest from metagpt.actions.execute_code import ExecutePyCode -from metagpt.actions.write_analysis_code import ( - WriteCodeByGenerate, - WriteCodeWithTools, - WriteCodeWithToolsML, -) +from metagpt.actions.write_analysis_code import WriteCodeByGenerate, WriteCodeWithTools from metagpt.logs import logger from metagpt.plan.planner import STRUCTURAL_CONTEXT from metagpt.schema import Message, Plan, Task +@pytest.mark.skip @pytest.mark.asyncio async def test_write_code_by_list_plan(): write_code = WriteCodeByGenerate() @@ -23,35 +20,31 @@ async def test_write_code_by_list_plan(): print(f"\n任务: {task}\n\n") messages.append(Message(task, role="assistant")) code = await write_code.run(messages) - messages.append(Message(code, role="assistant")) + messages.append(Message(code["code"], role="assistant")) assert len(code) > 0 - output = await execute_code.run(code) + output = await execute_code.run(code["code"]) print(f"\n[Output]: 任务{task}的执行结果是: \n{output}\n") messages.append(output[0]) @pytest.mark.asyncio async def test_tool_recommendation(): - task = "对已经读取的数据集进行数据清洗" - code_steps = """ - step 1: 对数据集进行去重 - step 2: 对数据集进行缺失值处理 - """ + task = "clean and preprocess the data" + code_steps = "" available_tools = { - "fill_missing_value": "Completing missing values with simple strategies", - "split_bins": "Bin continuous data into intervals and return the bin identifier encoded as an integer value", + "FillMissingValue": "Filling missing values", + "SplitBins": "Bin continuous data into intervals and return the bin identifier encoded as an integer value", } write_code = WriteCodeWithTools() - tools = await write_code._tool_recommendation(task, code_steps, available_tools) + tools = await write_code._recommend_tool(task, code_steps, available_tools) assert len(tools) == 1 - assert tools[0] == "fill_missing_value" + assert "FillMissingValue" in tools @pytest.mark.asyncio async def test_write_code_with_tools(): write_code = WriteCodeWithTools() - write_code_ml = WriteCodeWithToolsML() requirement = "构造数据集并进行数据清洗" task_map = { @@ -84,7 +77,6 @@ async def test_write_code_with_tools(): task_map=task_map, current_task_id="2", ) - column_info = "" context = STRUCTURAL_CONTEXT.format( user_requirement=requirement, @@ -95,13 +87,10 @@ async def test_write_code_with_tools(): context_msg = [Message(content=context, role="user")] code = await write_code.run(context_msg, plan) + code = code["code"] assert len(code) > 0 print(code) - code_with_ml = await write_code_ml.run([], plan, column_info) - assert len(code_with_ml) > 0 - print(code_with_ml) - @pytest.mark.asyncio async def test_write_code_to_correct_error(): @@ -150,6 +139,7 @@ async def test_write_code_to_correct_error(): Message(content=error, role="user"), ] new_code = await WriteCodeByGenerate().run(context=context) + new_code = new_code["code"] print(new_code) assert "read_csv" in new_code # should correct read_excel to read_csv @@ -189,10 +179,12 @@ async def test_write_code_reuse_code_simple(): Message(content=structural_context, role="user"), ] code = await WriteCodeByGenerate().run(context=context) + code = code["code"] print(code) assert "pandas" not in code and "read_csv" not in code # should reuse import and read statement from previous one +@pytest.mark.skip @pytest.mark.asyncio async def test_write_code_reuse_code_long(): """test code reuse for long context""" @@ -245,13 +237,14 @@ async def test_write_code_reuse_code_long(): trial_results = await asyncio.gather(*trials) print(*trial_results, sep="\n\n***\n\n") success = [ - "load_iris" not in result and "iris_data" in result for result in trial_results + "load_iris" not in result["code"] and "iris_data" in result["code"] for result in trial_results ] # should reuse iris_data from previous tasks success_rate = sum(success) / trials_num logger.info(f"success rate: {success_rate :.2f}") assert success_rate >= 0.8 +@pytest.mark.skip @pytest.mark.asyncio async def test_write_code_reuse_code_long_for_wine(): """test code reuse for long context""" @@ -318,7 +311,7 @@ async def test_write_code_reuse_code_long_for_wine(): trial_results = await asyncio.gather(*trials) print(*trial_results, sep="\n\n***\n\n") success = [ - "load_wine" not in result and "wine_data" in result for result in trial_results + "load_wine" not in result["code"] and "wine_data" in result["code"] for result in trial_results ] # should reuse iris_data from previous tasks success_rate = sum(success) / trials_num logger.info(f"success rate: {success_rate :.2f}") diff --git a/tests/metagpt/actions/test_write_code.py b/tests/metagpt/actions/test_write_code.py index 249145c92..ee05e0f7d 100644 --- a/tests/metagpt/actions/test_write_code.py +++ b/tests/metagpt/actions/test_write_code.py @@ -12,28 +12,22 @@ from pathlib import Path import pytest from metagpt.actions.write_code import WriteCode -from metagpt.config import CONFIG -from metagpt.const import ( - CODE_SUMMARIES_FILE_REPO, - SYSTEM_DESIGN_FILE_REPO, - TASK_FILE_REPO, - TEST_OUTPUTS_FILE_REPO, -) from metagpt.logs import logger -from metagpt.provider.openai_api import OpenAILLM as LLM from metagpt.schema import CodingContext, Document from metagpt.utils.common import aread -from metagpt.utils.file_repository import FileRepository from tests.metagpt.actions.mock_markdown import TASKS_2, WRITE_CODE_PROMPT_SAMPLE @pytest.mark.asyncio -async def test_write_code(): - context = CodingContext( +async def test_write_code(context): + # Prerequisites + context.src_workspace = context.git_repo.workdir / "writecode" + + coding_ctx = CodingContext( filename="task_filename.py", design_doc=Document(content="设计一个名为'add'的函数，该函数接受两个整数作为输入，并返回它们的和。") ) - doc = Document(content=context.model_dump_json()) - write_code = WriteCode(context=doc) + doc = Document(content=coding_ctx.model_dump_json()) + write_code = WriteCode(i_context=doc, context=context) code = await write_code.run() logger.info(code.model_dump_json()) @@ -44,48 +38,44 @@ async def test_write_code(): @pytest.mark.asyncio -async def test_write_code_directly(): +async def test_write_code_directly(context): prompt = WRITE_CODE_PROMPT_SAMPLE + "\n" + TASKS_2[0] - llm = LLM() + llm = context.llm_with_cost_manager_from_llm_config(context.config.llm) rsp = await llm.aask(prompt) logger.info(rsp) @pytest.mark.asyncio -async def test_write_code_deps(): +async def test_write_code_deps(context): # Prerequisites - CONFIG.src_workspace = CONFIG.git_repo.workdir / "snake1/snake1" + context.src_workspace = context.git_repo.workdir / "snake1/snake1" demo_path = Path(__file__).parent / "../../data/demo_project" - await FileRepository.save_file( - filename="test_game.py.json", - content=await aread(str(demo_path / "test_game.py.json")), - relative_path=TEST_OUTPUTS_FILE_REPO, + await context.repo.test_outputs.save( + filename="test_game.py.json", content=await aread(str(demo_path / "test_game.py.json")) ) - await FileRepository.save_file( + await context.repo.docs.code_summary.save( filename="20231221155954.json", content=await aread(str(demo_path / "code_summaries.json")), - relative_path=CODE_SUMMARIES_FILE_REPO, ) - await FileRepository.save_file( + await context.repo.docs.system_design.save( filename="20231221155954.json", content=await aread(str(demo_path / "system_design.json")), - relative_path=SYSTEM_DESIGN_FILE_REPO, ) - await FileRepository.save_file( - filename="20231221155954.json", content=await aread(str(demo_path / "tasks.json")), relative_path=TASK_FILE_REPO + await context.repo.docs.task.save( + filename="20231221155954.json", content=await aread(str(demo_path / "tasks.json")) ) - await FileRepository.save_file( - filename="main.py", content='if __name__ == "__main__":\nmain()', relative_path=CONFIG.src_workspace + await context.repo.with_src_path(context.src_workspace).srcs.save( + filename="main.py", content='if __name__ == "__main__":\nmain()' ) - context = CodingContext( + ccontext = CodingContext( filename="game.py", - design_doc=await FileRepository.get_file(filename="20231221155954.json", relative_path=SYSTEM_DESIGN_FILE_REPO), - task_doc=await FileRepository.get_file(filename="20231221155954.json", relative_path=TASK_FILE_REPO), + design_doc=await context.repo.docs.system_design.get(filename="20231221155954.json"), + task_doc=await context.repo.docs.task.get(filename="20231221155954.json"), code_doc=Document(filename="game.py", content="", root_path="snake1"), ) - coding_doc = Document(root_path="snake1", filename="game.py", content=context.json()) + coding_doc = Document(root_path="snake1", filename="game.py", content=ccontext.json()) - action = WriteCode(context=coding_doc) + action = WriteCode(i_context=coding_doc, context=context) rsp = await action.run() assert rsp assert rsp.code_doc.content diff --git a/tests/metagpt/actions/test_write_code_plan_and_change_an.py b/tests/metagpt/actions/test_write_code_plan_and_change_an.py new file mode 100644 index 000000000..9cd51398f --- /dev/null +++ b/tests/metagpt/actions/test_write_code_plan_and_change_an.py @@ -0,0 +1,71 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/01/03 +@Author : mannaandpoem +@File : test_write_code_plan_and_change_an.py +""" +import pytest +from openai._models import BaseModel + +from metagpt.actions.action_node import ActionNode +from metagpt.actions.write_code import WriteCode +from metagpt.actions.write_code_plan_and_change_an import ( + REFINED_TEMPLATE, + WriteCodePlanAndChange, +) +from metagpt.schema import CodePlanAndChangeContext +from tests.data.incremental_dev_project.mock import ( + CODE_PLAN_AND_CHANGE_SAMPLE, + DESIGN_SAMPLE, + NEW_REQUIREMENT_SAMPLE, + REFINED_CODE_INPUT_SAMPLE, + REFINED_CODE_SAMPLE, + TASKS_SAMPLE, +) + + +def mock_code_plan_and_change(): + return CODE_PLAN_AND_CHANGE_SAMPLE + + +@pytest.mark.asyncio +async def test_write_code_plan_and_change_an(mocker): + root = ActionNode.from_children( + "WriteCodePlanAndChange", [ActionNode(key="", expected_type=str, instruction="", example="")] + ) + root.instruct_content = BaseModel() + root.instruct_content.model_dump = mock_code_plan_and_change + mocker.patch("metagpt.actions.write_code_plan_and_change_an.WriteCodePlanAndChange.run", return_value=root) + + requirement = "New requirement" + prd_filename = "prd.md" + design_filename = "design.md" + task_filename = "task.md" + code_plan_and_change_context = CodePlanAndChangeContext( + requirement=requirement, + prd_filename=prd_filename, + design_filename=design_filename, + task_filename=task_filename, + ) + node = await WriteCodePlanAndChange(i_context=code_plan_and_change_context).run() + + assert "Code Plan And Change" in node.instruct_content.model_dump() + + +@pytest.mark.asyncio +async def test_refine_code(mocker): + mocker.patch.object(WriteCode, "_aask", return_value=REFINED_CODE_SAMPLE) + prompt = REFINED_TEMPLATE.format( + user_requirement=NEW_REQUIREMENT_SAMPLE, + code_plan_and_change=CODE_PLAN_AND_CHANGE_SAMPLE, + design=DESIGN_SAMPLE, + task=TASKS_SAMPLE, + code=REFINED_CODE_INPUT_SAMPLE, + logs="", + feedback="", + filename="game.py", + summary_log="", + ) + code = await WriteCode().write_code(prompt=prompt) + assert "def" in code diff --git a/tests/metagpt/actions/test_write_code_review.py b/tests/metagpt/actions/test_write_code_review.py index 3343b42b4..a08dd07bc 100644 --- a/tests/metagpt/actions/test_write_code_review.py +++ b/tests/metagpt/actions/test_write_code_review.py @@ -12,28 +12,25 @@ from metagpt.schema import CodingContext, Document @pytest.mark.asyncio -async def test_write_code_review(capfd): +async def test_write_code_review(capfd, context): + context.src_workspace = context.repo.workdir / "srcs" code = """ def add(a, b): return a + """ - context = CodingContext( + coding_context = CodingContext( filename="math.py", design_doc=Document(content="编写一个从a加b的函数，返回a+b"), code_doc=Document(content=code) ) - context = await WriteCodeReview(context=context).run() + await WriteCodeReview(i_context=coding_context, context=context).run() # 我们不能精确地预测生成的代码评审，但我们可以检查返回的是否为字符串 - assert isinstance(context.code_doc.content, str) - assert len(context.code_doc.content) > 0 + assert isinstance(coding_context.code_doc.content, str) + assert len(coding_context.code_doc.content) > 0 captured = capfd.readouterr() print(f"输出内容: {captured.out}") -# @pytest.mark.asyncio -# async def test_write_code_review_directly(): -# code = SEARCH_CODE_SAMPLE -# write_code_review = WriteCodeReview("write_code_review") -# review = await write_code_review.run(code) -# logger.info(review) +if __name__ == "__main__": + pytest.main([__file__, "-s"]) diff --git a/tests/metagpt/actions/test_write_docstring.py b/tests/metagpt/actions/test_write_docstring.py index a0fc46ebd..ebb7e8cb1 100644 --- a/tests/metagpt/actions/test_write_docstring.py +++ b/tests/metagpt/actions/test_write_docstring.py @@ -27,8 +27,8 @@ class Person: ], ids=["google", "numpy", "sphinx"], ) -async def test_write_docstring(style: str, part: str): - ret = await WriteDocstring().run(code, style=style) +async def test_write_docstring(style: str, part: str, context): + ret = await WriteDocstring(context=context).run(code, style=style) assert part in ret diff --git a/tests/metagpt/actions/test_write_prd.py b/tests/metagpt/actions/test_write_prd.py index 7317bba76..31d20018e 100644 --- a/tests/metagpt/actions/test_write_prd.py +++ b/tests/metagpt/actions/test_write_prd.py @@ -9,21 +9,19 @@ import pytest from metagpt.actions import UserRequirement, WritePRD -from metagpt.config import CONFIG -from metagpt.const import DOCS_FILE_REPO, PRDS_FILE_REPO, REQUIREMENT_FILENAME +from metagpt.const import REQUIREMENT_FILENAME from metagpt.logs import logger from metagpt.roles.product_manager import ProductManager from metagpt.roles.role import RoleReactMode from metagpt.schema import Message from metagpt.utils.common import any_to_str -from metagpt.utils.file_repository import FileRepository @pytest.mark.asyncio -async def test_write_prd(new_filename): - product_manager = ProductManager() +async def test_write_prd(new_filename, context): + product_manager = ProductManager(context=context) requirements = "开发一个基于大语言模型与私有知识库的搜索引擎，希望可以基于大语言模型进行搜索总结" - await FileRepository.save_file(filename=REQUIREMENT_FILENAME, content=requirements, relative_path=DOCS_FILE_REPO) + await context.repo.docs.save(filename=REQUIREMENT_FILENAME, content=requirements) product_manager.rc.react_mode = RoleReactMode.BY_ORDER prd = await product_manager.run(Message(content=requirements, cause_by=UserRequirement)) assert prd.cause_by == any_to_str(WritePRD) @@ -33,7 +31,7 @@ async def test_write_prd(new_filename): # Assert the prd is not None or empty assert prd is not None assert prd.content != "" - assert CONFIG.git_repo.new_file_repository(relative_path=PRDS_FILE_REPO).changed_files + assert product_manager.context.repo.docs.prd.changed_files if __name__ == "__main__": diff --git a/tests/metagpt/actions/test_write_prd_an.py b/tests/metagpt/actions/test_write_prd_an.py new file mode 100644 index 000000000..378ce42c3 --- /dev/null +++ b/tests/metagpt/actions/test_write_prd_an.py @@ -0,0 +1,48 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/01/03 +@Author : mannaandpoem +@File : test_write_prd_an.py +""" +import pytest +from openai._models import BaseModel + +from metagpt.actions.action_node import ActionNode +from metagpt.actions.write_prd import NEW_REQ_TEMPLATE +from metagpt.actions.write_prd_an import REFINED_PRD_NODE +from metagpt.llm import LLM +from tests.data.incremental_dev_project.mock import ( + NEW_REQUIREMENT_SAMPLE, + PRD_SAMPLE, + REFINED_PRD_JSON, +) + + +@pytest.fixture() +def llm(): + return LLM() + + +def mock_refined_prd_json(): + return REFINED_PRD_JSON + + +@pytest.mark.asyncio +async def test_write_prd_an(mocker): + root = ActionNode.from_children("RefinedPRD", [ActionNode(key="", expected_type=str, instruction="", example="")]) + root.instruct_content = BaseModel() + root.instruct_content.model_dump = mock_refined_prd_json + mocker.patch("metagpt.actions.write_prd_an.REFINED_PRD_NODE.fill", return_value=root) + + prompt = NEW_REQ_TEMPLATE.format( + requirements=NEW_REQUIREMENT_SAMPLE, + old_prd=PRD_SAMPLE, + ) + node = await REFINED_PRD_NODE.fill(prompt, llm) + + assert "Refined Requirements" in node.instruct_content.model_dump() + assert "Refined Product Goals" in node.instruct_content.model_dump() + assert "Refined User Stories" in node.instruct_content.model_dump() + assert "Refined Requirement Analysis" in node.instruct_content.model_dump() + assert "Refined Requirement Pool" in node.instruct_content.model_dump() diff --git a/tests/metagpt/actions/test_write_prd_review.py b/tests/metagpt/actions/test_write_prd_review.py index 9b3f0a285..8e1601b2e 100644 --- a/tests/metagpt/actions/test_write_prd_review.py +++ b/tests/metagpt/actions/test_write_prd_review.py @@ -11,7 +11,7 @@ from metagpt.actions.write_prd_review import WritePRDReview @pytest.mark.asyncio -async def test_write_prd_review(): +async def test_write_prd_review(context): prd = """ Introduction: This is a new feature for our product. Goals: The goal is to improve user engagement. @@ -23,7 +23,7 @@ async def test_write_prd_review(): Timeline: The feature should be ready for testing in 1.5 months. """ - write_prd_review = WritePRDReview(name="write_prd_review") + write_prd_review = WritePRDReview(name="write_prd_review", context=context) prd_review = await write_prd_review.run(prd) diff --git a/tests/metagpt/actions/test_write_review.py b/tests/metagpt/actions/test_write_review.py index 2d188b720..0274a3532 100644 --- a/tests/metagpt/actions/test_write_review.py +++ b/tests/metagpt/actions/test_write_review.py @@ -9,7 +9,7 @@ import pytest from metagpt.actions.write_review import WriteReview -CONTEXT = """ +TEMPLATE_CONTEXT = """ { "Language": "zh_cn", "Programming Language": "Python", @@ -46,8 +46,8 @@ CONTEXT = """ @pytest.mark.asyncio -async def test_write_review(): - write_review = WriteReview() - review = await write_review.run(CONTEXT) +async def test_write_review(context): + write_review = WriteReview(context=context) + review = await write_review.run(TEMPLATE_CONTEXT) assert review.instruct_content assert review.get("LGTM") in ["LGTM", "LBTM"] diff --git a/tests/metagpt/actions/test_write_teaching_plan.py b/tests/metagpt/actions/test_write_teaching_plan.py index 57a4f5eb0..bb68d4286 100644 --- a/tests/metagpt/actions/test_write_teaching_plan.py +++ b/tests/metagpt/actions/test_write_teaching_plan.py @@ -13,11 +13,11 @@ from metagpt.actions.write_teaching_plan import WriteTeachingPlanPart @pytest.mark.asyncio @pytest.mark.parametrize( - ("topic", "context"), + ("topic", "content"), [("Title", "Lesson 1: Learn to draw an apple."), ("Teaching Content", "Lesson 1: Learn to draw an apple.")], ) -async def test_write_teaching_plan_part(topic, context): - action = WriteTeachingPlanPart(topic=topic, context=context) +async def test_write_teaching_plan_part(topic, content, context): + action = WriteTeachingPlanPart(topic=topic, i_context=content, context=context) rsp = await action.run() assert rsp diff --git a/tests/metagpt/actions/test_write_test.py b/tests/metagpt/actions/test_write_test.py index 9649b9abb..9469dd312 100644 --- a/tests/metagpt/actions/test_write_test.py +++ b/tests/metagpt/actions/test_write_test.py @@ -13,7 +13,7 @@ from metagpt.schema import Document, TestingContext @pytest.mark.asyncio -async def test_write_test(): +async def test_write_test(context): code = """ import random from typing import Tuple @@ -25,8 +25,8 @@ async def test_write_test(): def generate(self, max_y: int, max_x: int): self.position = (random.randint(1, max_y - 1), random.randint(1, max_x - 1)) """ - context = TestingContext(filename="food.py", code_doc=Document(filename="food.py", content=code)) - write_test = WriteTest(context=context) + testing_context = TestingContext(filename="food.py", code_doc=Document(filename="food.py", content=code)) + write_test = WriteTest(i_context=testing_context, context=context) context = await write_test.run() logger.info(context.model_dump_json()) @@ -39,12 +39,12 @@ async def test_write_test(): @pytest.mark.asyncio -async def test_write_code_invalid_code(mocker): +async def test_write_code_invalid_code(mocker, context): # Mock the _aask method to return an invalid code string mocker.patch.object(WriteTest, "_aask", return_value="Invalid Code String") # Create an instance of WriteTest - write_test = WriteTest() + write_test = WriteTest(context=context) # Call the write_code method code = await write_test.write_code("Some prompt:") diff --git a/tests/metagpt/actions/test_write_tutorial.py b/tests/metagpt/actions/test_write_tutorial.py index 27a323b44..a83da1a1c 100644 --- a/tests/metagpt/actions/test_write_tutorial.py +++ b/tests/metagpt/actions/test_write_tutorial.py @@ -14,8 +14,8 @@ from metagpt.actions.write_tutorial import WriteContent, WriteDirectory @pytest.mark.asyncio @pytest.mark.parametrize(("language", "topic"), [("English", "Write a tutorial about Python")]) -async def test_write_directory(language: str, topic: str): - ret = await WriteDirectory(language=language).run(topic=topic) +async def test_write_directory(language: str, topic: str, context): + ret = await WriteDirectory(language=language, context=context).run(topic=topic) assert isinstance(ret, dict) assert "title" in ret assert "directory" in ret @@ -29,8 +29,8 @@ async def test_write_directory(language: str, topic: str): ("language", "topic", "directory"), [("English", "Write a tutorial about Python", {"Introduction": ["What is Python?", "Why learn Python?"]})], ) -async def test_write_content(language: str, topic: str, directory: Dict): - ret = await WriteContent(language=language, directory=directory).run(topic=topic) +async def test_write_content(language: str, topic: str, directory: Dict, context): + ret = await WriteContent(language=language, directory=directory, context=context).run(topic=topic) assert isinstance(ret, str) assert list(directory.keys())[0] in ret for value in list(directory.values())[0]: diff --git a/tests/metagpt/learn/test_skill_loader.py b/tests/metagpt/learn/test_skill_loader.py index 529a490c8..f1952c275 100644 --- a/tests/metagpt/learn/test_skill_loader.py +++ b/tests/metagpt/learn/test_skill_loader.py @@ -10,13 +10,12 @@ from pathlib import Path import pytest -from metagpt.config import CONFIG from metagpt.learn.skill_loader import SkillsDeclaration @pytest.mark.asyncio -async def test_suite(): - CONFIG.agent_skills = [ +async def test_suite(context): + context.kwargs.agent_skills = [ {"id": 1, "name": "text_to_speech", "type": "builtin", "config": {}, "enabled": True}, {"id": 2, "name": "text_to_image", "type": "builtin", "config": {}, "enabled": True}, {"id": 3, "name": "ai_call", "type": "builtin", "config": {}, "enabled": True}, @@ -27,7 +26,7 @@ async def test_suite(): ] pathname = Path(__file__).parent / "../../../docs/.well-known/skills.yaml" loader = await SkillsDeclaration.load(skill_yaml_file_name=pathname) - skills = loader.get_skill_list() + skills = loader.get_skill_list(context=context) assert skills assert len(skills) >= 3 for desc, name in skills.items(): diff --git a/tests/metagpt/learn/test_text_to_embedding.py b/tests/metagpt/learn/test_text_to_embedding.py index cbd1bbbbc..f50f6a7aa 100644 --- a/tests/metagpt/learn/test_text_to_embedding.py +++ b/tests/metagpt/learn/test_text_to_embedding.py @@ -6,19 +6,33 @@ @File : test_text_to_embedding.py @Desc : Unit tests. """ +import json +from pathlib import Path import pytest -from metagpt.config import CONFIG +from metagpt.config2 import Config from metagpt.learn.text_to_embedding import text_to_embedding +from metagpt.utils.common import aread @pytest.mark.asyncio -async def test_text_to_embedding(): - # Prerequisites - assert CONFIG.OPENAI_API_KEY +async def test_text_to_embedding(mocker): + # mock + config = Config.default() + mock_post = mocker.patch("aiohttp.ClientSession.post") + mock_response = mocker.AsyncMock() + mock_response.status = 200 + data = await aread(Path(__file__).parent / "../../data/openai/embedding.json") + mock_response.json.return_value = json.loads(data) + mock_post.return_value.__aenter__.return_value = mock_response + config.get_openai_llm().proxy = mocker.PropertyMock(return_value="http://mock.proxy") - v = await text_to_embedding(text="Panda emoji") + # Prerequisites + assert config.get_openai_llm().api_key + assert config.get_openai_llm().proxy + + v = await text_to_embedding(text="Panda emoji", config=config) assert len(v.data) > 0 diff --git a/tests/metagpt/learn/test_text_to_image.py b/tests/metagpt/learn/test_text_to_image.py index 1485df5c6..167a35891 100644 --- a/tests/metagpt/learn/test_text_to_image.py +++ b/tests/metagpt/learn/test_text_to_image.py @@ -6,11 +6,13 @@ @File : test_text_to_image.py @Desc : Unit tests. """ +import base64 - +import openai import pytest +from pydantic import BaseModel -from metagpt.config import CONFIG +from metagpt.config2 import Config from metagpt.learn.text_to_image import text_to_image from metagpt.tools.metagpt_text_to_image import MetaGPTText2Image from metagpt.tools.openai_text_to_image import OpenAIText2Image @@ -24,23 +26,37 @@ async def test_text_to_image(mocker): mocker.patch.object(OpenAIText2Image, "text_2_image", return_value=b"mock OpenAIText2Image") mocker.patch.object(S3, "cache", return_value="http://mock/s3") - # Prerequisites - assert CONFIG.METAGPT_TEXT_TO_IMAGE_MODEL_URL - assert CONFIG.OPENAI_API_KEY + config = Config.default() + assert config.METAGPT_TEXT_TO_IMAGE_MODEL_URL - data = await text_to_image("Panda emoji", size_type="512x512") + data = await text_to_image("Panda emoji", size_type="512x512", config=config) assert "base64" in data or "http" in data - # Mock session env - old_options = CONFIG.options.copy() - new_options = old_options.copy() - new_options["METAGPT_TEXT_TO_IMAGE_MODEL_URL"] = None - CONFIG.set_context(new_options) - try: - data = await text_to_image("Panda emoji", size_type="512x512") - assert "base64" in data or "http" in data - finally: - CONFIG.set_context(old_options) + +@pytest.mark.asyncio +async def test_openai_text_to_image(mocker): + # mocker + mock_url = mocker.Mock() + mock_url.url.return_value = "http://mock.com/0.png" + + class _MockData(BaseModel): + data: list + + mock_data = _MockData(data=[mock_url]) + mocker.patch.object(openai.resources.images.AsyncImages, "generate", return_value=mock_data) + mock_post = mocker.patch("aiohttp.ClientSession.get") + mock_response = mocker.AsyncMock() + mock_response.status = 200 + mock_response.read.return_value = base64.b64encode(b"success") + mock_post.return_value.__aenter__.return_value = mock_response + mocker.patch.object(S3, "cache", return_value="http://mock.s3.com/0.png") + + config = Config.default() + config.METAGPT_TEXT_TO_IMAGE_MODEL_URL = None + assert config.get_openai_llm() + + data = await text_to_image("Panda emoji", size_type="512x512", config=config) + assert "base64" in data or "http" in data if __name__ == "__main__": diff --git a/tests/metagpt/learn/test_text_to_speech.py b/tests/metagpt/learn/test_text_to_speech.py index aca08b9a2..38e051cc6 100644 --- a/tests/metagpt/learn/test_text_to_speech.py +++ b/tests/metagpt/learn/test_text_to_speech.py @@ -8,35 +8,65 @@ """ import pytest +from azure.cognitiveservices.speech import ResultReason, SpeechSynthesizer -from metagpt.config import CONFIG +from metagpt.config2 import Config from metagpt.learn.text_to_speech import text_to_speech +from metagpt.tools.iflytek_tts import IFlyTekTTS +from metagpt.utils.s3 import S3 @pytest.mark.asyncio -async def test_text_to_speech(): - # Prerequisites - assert CONFIG.IFLYTEK_APP_ID - assert CONFIG.IFLYTEK_API_KEY - assert CONFIG.IFLYTEK_API_SECRET - assert CONFIG.AZURE_TTS_SUBSCRIPTION_KEY and CONFIG.AZURE_TTS_SUBSCRIPTION_KEY != "YOUR_API_KEY" - assert CONFIG.AZURE_TTS_REGION +async def test_azure_text_to_speech(mocker): + # mock + config = Config.default() + config.IFLYTEK_API_KEY = None + config.IFLYTEK_API_SECRET = None + config.IFLYTEK_APP_ID = None + mock_result = mocker.Mock() + mock_result.audio_data = b"mock audio data" + mock_result.reason = ResultReason.SynthesizingAudioCompleted + mock_data = mocker.Mock() + mock_data.get.return_value = mock_result + mocker.patch.object(SpeechSynthesizer, "speak_ssml_async", return_value=mock_data) + mocker.patch.object(S3, "cache", return_value="http://mock.s3.com/1.wav") + # Prerequisites + assert not config.IFLYTEK_APP_ID + assert not config.IFLYTEK_API_KEY + assert not config.IFLYTEK_API_SECRET + assert config.AZURE_TTS_SUBSCRIPTION_KEY and config.AZURE_TTS_SUBSCRIPTION_KEY != "YOUR_API_KEY" + assert config.AZURE_TTS_REGION + + config.copy() # test azure - data = await text_to_speech("panda emoji") + data = await text_to_speech("panda emoji", config=config) assert "base64" in data or "http" in data - # test iflytek - ## Mock session env - old_options = CONFIG.options.copy() - new_options = old_options.copy() - new_options["AZURE_TTS_SUBSCRIPTION_KEY"] = "" - CONFIG.set_context(new_options) - try: - data = await text_to_speech("panda emoji") - assert "base64" in data or "http" in data - finally: - CONFIG.set_context(old_options) + +@pytest.mark.asyncio +async def test_iflytek_text_to_speech(mocker): + # mock + config = Config.default() + config.AZURE_TTS_SUBSCRIPTION_KEY = None + config.AZURE_TTS_REGION = None + mocker.patch.object(IFlyTekTTS, "synthesize_speech", return_value=None) + mock_data = mocker.AsyncMock() + mock_data.read.return_value = b"mock iflytek" + mock_reader = mocker.patch("aiofiles.open") + mock_reader.return_value.__aenter__.return_value = mock_data + mocker.patch.object(S3, "cache", return_value="http://mock.s3.com/1.mp3") + + # Prerequisites + assert config.IFLYTEK_APP_ID + assert config.IFLYTEK_API_KEY + assert config.IFLYTEK_API_SECRET + assert not config.AZURE_TTS_SUBSCRIPTION_KEY or config.AZURE_TTS_SUBSCRIPTION_KEY == "YOUR_API_KEY" + assert not config.AZURE_TTS_REGION + + # test azure + data = await text_to_speech("panda emoji", config=config) + assert "base64" in data or "http" in data if __name__ == "__main__": diff --git a/tests/metagpt/memory/test_brain_memory.py b/tests/metagpt/memory/test_brain_memory.py index 32dcd672a..72ffcc538 100644 --- a/tests/metagpt/memory/test_brain_memory.py +++ b/tests/metagpt/memory/test_brain_memory.py @@ -8,7 +8,6 @@ import pytest -from metagpt.config import LLMProviderEnum from metagpt.llm import LLM from metagpt.memory.brain_memory import BrainMemory from metagpt.schema import Message @@ -46,7 +45,7 @@ def test_extract_info(input, tag, val): @pytest.mark.asyncio -@pytest.mark.parametrize("llm", [LLM(provider=LLMProviderEnum.OPENAI), LLM(provider=LLMProviderEnum.METAGPT)]) +@pytest.mark.parametrize("llm", [LLM()]) async def test_memory_llm(llm): memory = BrainMemory() for i in range(500): diff --git a/tests/metagpt/memory/test_longterm_memory.py b/tests/metagpt/memory/test_longterm_memory.py index 0f7a4fac4..5c71ddd13 100644 --- a/tests/metagpt/memory/test_longterm_memory.py +++ b/tests/metagpt/memory/test_longterm_memory.py @@ -2,7 +2,6 @@ # -*- coding: utf-8 -*- """ @Desc : unittest of `metagpt/memory/longterm_memory.py` -@Modified By: mashenquan, 2023/8/20. Remove global configuration `CONFIG`, enable configuration support for business isolation. """ import os @@ -10,17 +9,15 @@ import os import pytest from metagpt.actions import UserRequirement -from metagpt.config import CONFIG +from metagpt.config2 import config from metagpt.memory.longterm_memory import LongTermMemory from metagpt.roles.role import RoleContext from metagpt.schema import Message +os.environ.setdefault("OPENAI_API_KEY", config.get_openai_llm().api_key) + def test_ltm_search(): - assert hasattr(CONFIG, "long_term_memory") is True - os.environ.setdefault("OPENAI_API_KEY", CONFIG.openai_api_key) - assert len(CONFIG.openai_api_key) > 20 - role_id = "UTUserLtm(Product Manager)" from metagpt.environment import Environment diff --git a/tests/metagpt/memory/test_memory.py b/tests/metagpt/memory/test_memory.py index 36d7ad488..a072b61de 100644 --- a/tests/metagpt/memory/test_memory.py +++ b/tests/metagpt/memory/test_memory.py @@ -32,7 +32,7 @@ def test_memory(): messages = memory.get_by_action(UserRequirement) assert len(messages) == 2 - messages = memory.get_by_actions([UserRequirement]) + messages = memory.get_by_actions({UserRequirement}) assert len(messages) == 2 messages = memory.try_remember("test message") diff --git a/tests/metagpt/memory/test_memory_storage.py b/tests/metagpt/memory/test_memory_storage.py index 0eb1069d5..e82a82fc8 100644 --- a/tests/metagpt/memory/test_memory_storage.py +++ b/tests/metagpt/memory/test_memory_storage.py @@ -11,12 +11,12 @@ from typing import List from metagpt.actions import UserRequirement, WritePRD from metagpt.actions.action_node import ActionNode -from metagpt.config import CONFIG +from metagpt.config2 import config from metagpt.const import DATA_PATH from metagpt.memory.memory_storage import MemoryStorage from metagpt.schema import Message -os.environ.setdefault("OPENAI_API_KEY", CONFIG.openai_api_key) +os.environ.setdefault("OPENAI_API_KEY", config.get_openai_llm().api_key) def test_idea_message(): diff --git a/tests/metagpt/provider/mock_llm_config.py b/tests/metagpt/provider/mock_llm_config.py new file mode 100644 index 000000000..e2f626a6a --- /dev/null +++ b/tests/metagpt/provider/mock_llm_config.py @@ -0,0 +1,44 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/1/8 17:03 +@Author : alexanderwu +@File : mock_llm_config.py +""" + +from metagpt.configs.llm_config import LLMConfig + +mock_llm_config = LLMConfig( + llm_type="mock", + api_key="mock_api_key", + base_url="mock_base_url", + app_id="mock_app_id", + api_secret="mock_api_secret", + domain="mock_domain", +) + + +mock_llm_config_proxy = LLMConfig( + llm_type="mock", + api_key="mock_api_key", + base_url="mock_base_url", + proxy="http://localhost:8080", +) + + +mock_llm_config_azure = LLMConfig( + llm_type="azure", + api_version="2023-09-01-preview", + api_key="mock_api_key", + base_url="mock_base_url", + proxy="http://localhost:8080", +) + + +mock_llm_config_zhipu = LLMConfig( + llm_type="zhipu", + api_key="mock_api_key.zhipu", + base_url="mock_base_url", + model="mock_zhipu_model", + proxy="http://localhost:8080", +) diff --git a/tests/metagpt/provider/test_anthropic_api.py b/tests/metagpt/provider/test_anthropic_api.py index 4410717a9..6962ab064 100644 --- a/tests/metagpt/provider/test_anthropic_api.py +++ b/tests/metagpt/provider/test_anthropic_api.py @@ -6,10 +6,8 @@ import pytest from anthropic.resources.completions import Completion -from metagpt.config import CONFIG from metagpt.provider.anthropic_api import Claude2 - -CONFIG.anthropic_api_key = "xxx" +from tests.metagpt.provider.mock_llm_config import mock_llm_config prompt = "who are you" resp = "I'am Claude2" @@ -25,10 +23,10 @@ async def mock_anthropic_acompletions_create(self, model: str, prompt: str, max_ def test_claude2_ask(mocker): mocker.patch("anthropic.resources.completions.Completions.create", mock_anthropic_completions_create) - assert resp == Claude2().ask(prompt) + assert resp == Claude2(mock_llm_config).ask(prompt) @pytest.mark.asyncio async def test_claude2_aask(mocker): mocker.patch("anthropic.resources.completions.AsyncCompletions.create", mock_anthropic_acompletions_create) - assert resp == await Claude2().aask(prompt) + assert resp == await Claude2(mock_llm_config).aask(prompt) diff --git a/tests/metagpt/provider/test_azure_llm.py b/tests/metagpt/provider/test_azure_llm.py new file mode 100644 index 000000000..51e051145 --- /dev/null +++ b/tests/metagpt/provider/test_azure_llm.py @@ -0,0 +1,12 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +# @Desc : + +from metagpt.provider import AzureOpenAILLM +from tests.metagpt.provider.mock_llm_config import mock_llm_config_azure + + +def test_azure_llm(): + llm = AzureOpenAILLM(mock_llm_config_azure) + kwargs = llm._make_client_kwargs() + assert kwargs["azure_endpoint"] == mock_llm_config_azure.base_url diff --git a/tests/metagpt/provider/test_azure_openai_api.py b/tests/metagpt/provider/test_azure_openai_api.py deleted file mode 100644 index f36740e65..000000000 --- a/tests/metagpt/provider/test_azure_openai_api.py +++ /dev/null @@ -1,14 +0,0 @@ -#!/usr/bin/env python -# -*- coding: utf-8 -*- -# @Desc : - - -from metagpt.config import CONFIG -from metagpt.provider.azure_openai_api import AzureOpenAILLM - -CONFIG.OPENAI_API_VERSION = "xx" -CONFIG.openai_proxy = "http://127.0.0.1:80" # fake value - - -def test_azure_openai_api(): - _ = AzureOpenAILLM() diff --git a/tests/metagpt/provider/test_base_gpt_api.py b/tests/metagpt/provider/test_base_llm.py similarity index 93% rename from tests/metagpt/provider/test_base_gpt_api.py rename to tests/metagpt/provider/test_base_llm.py index 3443b5078..cc781f78a 100644 --- a/tests/metagpt/provider/test_base_gpt_api.py +++ b/tests/metagpt/provider/test_base_llm.py @@ -8,6 +8,7 @@ import pytest +from metagpt.configs.llm_config import LLMConfig from metagpt.provider.base_llm import BaseLLM from metagpt.schema import Message @@ -28,6 +29,9 @@ resp_content = default_chat_resp["choices"][0]["message"]["content"] class MockBaseLLM(BaseLLM): + def __init__(self, config: LLMConfig = None): + pass + def completion(self, messages: list[dict], timeout=3): return default_chat_resp @@ -102,5 +106,5 @@ async def test_async_base_llm(): resp = await base_llm.aask_batch([prompt_msg]) assert resp == resp_content - resp = await base_llm.aask_code([prompt_msg]) - assert resp == resp_content + # resp = await base_llm.aask_code([prompt_msg]) + # assert resp == resp_content diff --git a/tests/metagpt/provider/test_fireworks_api.py b/tests/metagpt/provider/test_fireworks_llm.py similarity index 96% rename from tests/metagpt/provider/test_fireworks_api.py rename to tests/metagpt/provider/test_fireworks_llm.py index d48686eaa..66b55e5b2 100644 --- a/tests/metagpt/provider/test_fireworks_api.py +++ b/tests/metagpt/provider/test_fireworks_llm.py @@ -13,17 +13,13 @@ from openai.types.chat.chat_completion_chunk import Choice as AChoice from openai.types.chat.chat_completion_chunk import ChoiceDelta from openai.types.completion_usage import CompletionUsage -from metagpt.config import CONFIG from metagpt.provider.fireworks_api import ( MODEL_GRADE_TOKEN_COSTS, FireworksCostManager, FireworksLLM, ) from metagpt.utils.cost_manager import Costs - -CONFIG.fireworks_api_key = "xxx" -CONFIG.max_budget = 10 -CONFIG.calc_usage = True +from tests.metagpt.provider.mock_llm_config import mock_llm_config resp_content = "I'm fireworks" default_resp = ChatCompletion( @@ -92,7 +88,7 @@ async def mock_openai_acompletions_create(self, stream: bool = False, **kwargs) async def test_fireworks_acompletion(mocker): mocker.patch("openai.resources.chat.completions.AsyncCompletions.create", mock_openai_acompletions_create) - fireworks_gpt = FireworksLLM() + fireworks_gpt = FireworksLLM(mock_llm_config) fireworks_gpt.model = "llama-v2-13b-chat" fireworks_gpt._update_costs( diff --git a/tests/metagpt/provider/test_google_gemini_api.py b/tests/metagpt/provider/test_google_gemini_api.py index ffd10df7f..404ae1e90 100644 --- a/tests/metagpt/provider/test_google_gemini_api.py +++ b/tests/metagpt/provider/test_google_gemini_api.py @@ -9,10 +9,8 @@ import pytest from google.ai import generativelanguage as glm from google.generativeai.types import content_types -from metagpt.config import CONFIG from metagpt.provider.google_gemini_api import GeminiLLM - -CONFIG.gemini_api_key = "xx" +from tests.metagpt.provider.mock_llm_config import mock_llm_config @dataclass @@ -62,7 +60,7 @@ async def test_gemini_acompletion(mocker): mock_gemini_generate_content_async, ) - gemini_gpt = GeminiLLM() + gemini_gpt = GeminiLLM(mock_llm_config) assert gemini_gpt._user_msg(prompt_msg) == {"role": "user", "parts": [prompt_msg]} assert gemini_gpt._assistant_msg(prompt_msg) == {"role": "model", "parts": [prompt_msg]} diff --git a/tests/metagpt/provider/test_human_provider.py b/tests/metagpt/provider/test_human_provider.py index 3f63410c0..97ed8bae6 100644 --- a/tests/metagpt/provider/test_human_provider.py +++ b/tests/metagpt/provider/test_human_provider.py @@ -5,6 +5,7 @@ import pytest from metagpt.provider.human_provider import HumanProvider +from tests.metagpt.provider.mock_llm_config import mock_llm_config resp_content = "test" resp_exit = "exit" @@ -13,7 +14,7 @@ resp_exit = "exit" @pytest.mark.asyncio async def test_async_human_provider(mocker): mocker.patch("builtins.input", lambda _: resp_content) - human_provider = HumanProvider() + human_provider = HumanProvider(mock_llm_config) resp = human_provider.ask(resp_content) assert resp == resp_content diff --git a/tests/metagpt/provider/test_metagpt_api.py b/tests/metagpt/provider/test_metagpt_api.py deleted file mode 100644 index 1f00cb653..000000000 --- a/tests/metagpt/provider/test_metagpt_api.py +++ /dev/null @@ -1,14 +0,0 @@ -#!/usr/bin/env python -# -*- coding: utf-8 -*- -""" -@Time : 2023/12/28 -@Author : mashenquan -@File : test_metagpt_api.py -""" -from metagpt.config import LLMProviderEnum -from metagpt.llm import LLM - - -def test_llm(): - llm = LLM(provider=LLMProviderEnum.METAGPT) - assert llm diff --git a/tests/metagpt/provider/test_metagpt_llm_api.py b/tests/metagpt/provider/test_metagpt_llm.py similarity index 63% rename from tests/metagpt/provider/test_metagpt_llm_api.py rename to tests/metagpt/provider/test_metagpt_llm.py index 8fce6b6b0..0263fe508 100644 --- a/tests/metagpt/provider/test_metagpt_llm_api.py +++ b/tests/metagpt/provider/test_metagpt_llm.py @@ -3,13 +3,14 @@ """ @Time : 2023/8/30 @Author : mashenquan -@File : test_metagpt_llm_api.py +@File : test_metagpt_llm.py """ from metagpt.provider.metagpt_api import MetaGPTLLM +from tests.metagpt.provider.mock_llm_config import mock_llm_config def test_metagpt(): - llm = MetaGPTLLM() + llm = MetaGPTLLM(mock_llm_config) assert llm diff --git a/tests/metagpt/provider/test_ollama_api.py b/tests/metagpt/provider/test_ollama_api.py index 1c604768e..5d942598b 100644 --- a/tests/metagpt/provider/test_ollama_api.py +++ b/tests/metagpt/provider/test_ollama_api.py @@ -7,8 +7,8 @@ from typing import Any, Tuple import pytest -from metagpt.config import CONFIG from metagpt.provider.ollama_api import OllamaLLM +from tests.metagpt.provider.mock_llm_config import mock_llm_config prompt_msg = "who are you" messages = [{"role": "user", "content": prompt_msg}] @@ -16,9 +16,6 @@ messages = [{"role": "user", "content": prompt_msg}] resp_content = "I'm ollama" default_resp = {"message": {"role": "assistant", "content": resp_content}} -CONFIG.ollama_api_base = "http://xxx" -CONFIG.max_budget = 10 - async def mock_ollama_arequest(self, stream: bool = False, **kwargs) -> Tuple[Any, Any, bool]: if stream: @@ -44,7 +41,7 @@ async def mock_ollama_arequest(self, stream: bool = False, **kwargs) -> Tuple[An async def test_gemini_acompletion(mocker): mocker.patch("metagpt.provider.general_api_requestor.GeneralAPIRequestor.arequest", mock_ollama_arequest) - ollama_gpt = OllamaLLM() + ollama_gpt = OllamaLLM(mock_llm_config) resp = await ollama_gpt.acompletion(messages) assert resp["message"]["content"] == default_resp["message"]["content"] diff --git a/tests/metagpt/provider/test_open_llm_api.py b/tests/metagpt/provider/test_open_llm_api.py index 85069c5e1..fc7b510cc 100644 --- a/tests/metagpt/provider/test_open_llm_api.py +++ b/tests/metagpt/provider/test_open_llm_api.py @@ -13,12 +13,9 @@ from openai.types.chat.chat_completion_chunk import Choice as AChoice from openai.types.chat.chat_completion_chunk import ChoiceDelta from openai.types.completion_usage import CompletionUsage -from metagpt.config import CONFIG from metagpt.provider.open_llm_api import OpenLLM from metagpt.utils.cost_manager import Costs - -CONFIG.max_budget = 10 -CONFIG.calc_usage = True +from tests.metagpt.provider.mock_llm_config import mock_llm_config resp_content = "I'm llama2" default_resp = ChatCompletion( @@ -71,7 +68,7 @@ async def mock_openai_acompletions_create(self, stream: bool = False, **kwargs) async def test_openllm_acompletion(mocker): mocker.patch("openai.resources.chat.completions.AsyncCompletions.create", mock_openai_acompletions_create) - openllm_gpt = OpenLLM() + openllm_gpt = OpenLLM(mock_llm_config) openllm_gpt.model = "llama-v2-13b-chat" openllm_gpt._update_costs(usage=CompletionUsage(prompt_tokens=100, completion_tokens=100, total_tokens=200)) diff --git a/tests/metagpt/provider/test_openai.py b/tests/metagpt/provider/test_openai.py index 77820a5f8..a49d7e85b 100644 --- a/tests/metagpt/provider/test_openai.py +++ b/tests/metagpt/provider/test_openai.py @@ -1,5 +1,3 @@ -from unittest.mock import Mock - import pytest from openai.types.chat import ( ChatCompletion, @@ -9,134 +7,92 @@ from openai.types.chat import ( from openai.types.chat.chat_completion import Choice from openai.types.chat.chat_completion_message_tool_call import Function -from metagpt.config import CONFIG +from metagpt.const import TEST_DATA_PATH +from metagpt.llm import LLM from metagpt.logs import logger -from metagpt.provider.openai_api import OpenAILLM +from metagpt.provider import OpenAILLM +from tests.metagpt.provider.mock_llm_config import ( + mock_llm_config, + mock_llm_config_proxy, +) -CONFIG.openai_proxy = None + +@pytest.mark.asyncio +async def test_text_to_speech(): + llm = LLM() + resp = await llm.atext_to_speech( + model="tts-1", + voice="alloy", + input="人生说起来长，但直到一个岁月回头看，许多事件仅是仓促的。一段一段拼凑一起，合成了人生。苦难当头时，当下不免觉得是折磨；回头看，也不够是一段短短的人生旅程。", + ) + assert 200 == resp.response.status_code + + +@pytest.mark.asyncio +async def test_speech_to_text(): + llm = LLM() + audio_file = open(f"{TEST_DATA_PATH}/audio/hello.mp3", "rb") + resp = await llm.aspeech_to_text(file=audio_file, model="whisper-1") + assert "你好" == resp.text + + +@pytest.fixture +def tool_calls_rsp(): + function_rsps = [ + Function(arguments='{\n"language": "python",\n"code": "print(\'hello world\')"}', name="execute"), + Function(arguments='{\n"language": "python",\n"code": \'print("hello world")\'}', name="execute"), + Function(arguments='{\n"language": \'python\',\n"code": "print(\'hello world\')"}', name="execute"), + Function(arguments='{\n"language": "python",\n"code": "print(\'hello world\')"}', name="execute"), + Function(arguments='{\n"language": "python",\n"code": ```print("hello world")```}', name="execute"), + Function(arguments='{\n"language": "python",\n"code": """print("hello world")"""}', name="execute"), + Function(arguments='\nprint("hello world")\\n', name="execute"), + # only `{` in arguments + Function(arguments='{\n"language": "python",\n"code": "print(\'hello world\')"', name="execute"), + # no `{`, `}` in arguments + Function(arguments='\n"language": "python",\n"code": "print(\'hello world\')"', name="execute"), + ] + tool_calls = [ + ChatCompletionMessageToolCall(type="function", id=f"call_{i}", function=f) for i, f in enumerate(function_rsps) + ] + messages = [ChatCompletionMessage(content=None, role="assistant", tool_calls=[t]) for t in tool_calls] + # 添加一个纯文本响应 + messages.append( + ChatCompletionMessage(content="Completed a python code for hello world!", role="assistant", tool_calls=None) + ) + # 添加 openai tool calls respond bug, code 出现在ChatCompletionMessage.content中 + messages.extend( + [ + ChatCompletionMessage(content="```python\nprint('hello world')```", role="assistant", tool_calls=None), + ChatCompletionMessage(content="'''python\nprint('hello world')'''", role="assistant", tool_calls=None), + ChatCompletionMessage(content='"""python\nprint(\'hello world\')"""', role="assistant", tool_calls=None), + ChatCompletionMessage(content="'''python\nprint(\"hello world\")'''", role="assistant", tool_calls=None), + ChatCompletionMessage(content="```python\nprint('hello world')```", role="assistant", tool_calls=None), + ] + ) + choices = [ + Choice(finish_reason="tool_calls", logprobs=None, index=i, message=msg) for i, msg in enumerate(messages) + ] + return [ + ChatCompletion(id=str(i), choices=[c], created=i, model="gpt-4", object="chat.completion") + for i, c in enumerate(choices) + ] class TestOpenAI: - @pytest.fixture - def config(self): - return Mock( - openai_api_key="test_key", - OPENAI_API_KEY="test_key", - openai_base_url="test_url", - OPENAI_BASE_URL="test_url", - openai_proxy=None, - openai_api_type="other", - ) - - @pytest.fixture - def config_azure(self): - return Mock( - openai_api_key="test_key", - OPENAI_API_KEY="test_key", - openai_api_version="test_version", - openai_base_url="test_url", - OPENAI_BASE_URL="test_url", - openai_proxy=None, - openai_api_type="azure", - ) - - @pytest.fixture - def config_proxy(self): - return Mock( - openai_api_key="test_key", - OPENAI_API_KEY="test_key", - openai_base_url="test_url", - OPENAI_BASE_URL="test_url", - openai_proxy="http://proxy.com", - openai_api_type="other", - ) - - @pytest.fixture - def config_azure_proxy(self): - return Mock( - openai_api_key="test_key", - OPENAI_API_KEY="test_key", - openai_api_version="test_version", - openai_base_url="test_url", - OPENAI_BASE_URL="test_url", - openai_proxy="http://proxy.com", - openai_api_type="azure", - ) - - @pytest.fixture - def tool_calls_rsp(self): - function_rsps = [ - Function(arguments='{\n"language": "python",\n"code": "print(\'hello world\')"}', name="execute"), - Function(arguments='{\n"language": "python",\n"code": \'print("hello world")\'}', name="execute"), - Function(arguments='{\n"language": \'python\',\n"code": "print(\'hello world\')"}', name="execute"), - Function(arguments='{\n"language": "python",\n"code": "print(\'hello world\')"}', name="execute"), - Function(arguments='{\n"language": "python",\n"code": ```print("hello world")```}', name="execute"), - Function(arguments='{\n"language": "python",\n"code": """print("hello world")"""}', name="execute"), - Function(arguments='\nprint("hello world")\\n', name="execute"), - # only `{` in arguments - Function(arguments='{\n"language": "python",\n"code": "print(\'hello world\')"', name="execute"), - # no `{`, `}` in arguments - Function(arguments='\n"language": "python",\n"code": "print(\'hello world\')"', name="execute"), - ] - tool_calls = [ - ChatCompletionMessageToolCall(type="function", id=f"call_{i}", function=f) - for i, f in enumerate(function_rsps) - ] - messages = [ChatCompletionMessage(content=None, role="assistant", tool_calls=[t]) for t in tool_calls] - # 添加一个纯文本响应 - messages.append( - ChatCompletionMessage(content="Completed a python code for hello world!", role="assistant", tool_calls=None) - ) - # 添加 openai tool calls respond bug, code 出现在ChatCompletionMessage.content中 - messages.extend( - [ - ChatCompletionMessage(content="```python\nprint('hello world')```", role="assistant", tool_calls=None), - ChatCompletionMessage(content="'''python\nprint('hello world')'''", role="assistant", tool_calls=None), - ChatCompletionMessage( - content='"""python\nprint(\'hello world\')"""', role="assistant", tool_calls=None - ), - ChatCompletionMessage( - content="'''python\nprint(\"hello world\")'''", role="assistant", tool_calls=None - ), - ChatCompletionMessage(content="```python\nprint('hello world')```", role="assistant", tool_calls=None), - ] - ) - choices = [ - Choice(finish_reason="tool_calls", logprobs=None, index=i, message=msg) for i, msg in enumerate(messages) - ] - return [ - ChatCompletion(id=str(i), choices=[c], created=i, model="gpt-4", object="chat.completion") - for i, c in enumerate(choices) - ] - - def test_make_client_kwargs_without_proxy(self, config): - instance = OpenAILLM() - instance.config = config + def test_make_client_kwargs_without_proxy(self): + instance = OpenAILLM(mock_llm_config) kwargs = instance._make_client_kwargs() - assert kwargs == {"api_key": "test_key", "base_url": "test_url"} + assert kwargs["api_key"] == "mock_api_key" + assert kwargs["base_url"] == "mock_base_url" assert "http_client" not in kwargs - def test_make_client_kwargs_without_proxy_azure(self, config_azure): - instance = OpenAILLM() - instance.config = config_azure - kwargs = instance._make_client_kwargs() - assert kwargs == {"api_key": "test_key", "base_url": "test_url"} - assert "http_client" not in kwargs - - def test_make_client_kwargs_with_proxy(self, config_proxy): - instance = OpenAILLM() - instance.config = config_proxy - kwargs = instance._make_client_kwargs() - assert "http_client" in kwargs - - def test_make_client_kwargs_with_proxy_azure(self, config_azure_proxy): - instance = OpenAILLM() - instance.config = config_azure_proxy + def test_make_client_kwargs_with_proxy(self): + instance = OpenAILLM(mock_llm_config_proxy) kwargs = instance._make_client_kwargs() assert "http_client" in kwargs def test_get_choice_function_arguments_for_aask_code(self, tool_calls_rsp): - instance = OpenAILLM() + instance = OpenAILLM(mock_llm_config_proxy) for i, rsp in enumerate(tool_calls_rsp): code = instance.get_choice_function_arguments(rsp) logger.info(f"\ntest get function call arguments {i}: {code}") diff --git a/tests/metagpt/provider/test_spark_api.py b/tests/metagpt/provider/test_spark_api.py index ee2d02c97..f5a6f66fd 100644 --- a/tests/metagpt/provider/test_spark_api.py +++ b/tests/metagpt/provider/test_spark_api.py @@ -4,14 +4,9 @@ import pytest -from metagpt.config import CONFIG +from metagpt.config2 import Config from metagpt.provider.spark_api import GetMessageFromWeb, SparkLLM - -CONFIG.spark_appid = "xxx" -CONFIG.spark_api_secret = "xxx" -CONFIG.spark_api_key = "xxx" -CONFIG.domain = "xxxxxx" -CONFIG.spark_url = "xxxx" +from tests.metagpt.provider.mock_llm_config import mock_llm_config prompt_msg = "who are you" resp_content = "I'm Spark" @@ -28,8 +23,8 @@ class MockWebSocketApp(object): def test_get_msg_from_web(mocker): mocker.patch("websocket.WebSocketApp", MockWebSocketApp) - get_msg_from_web = GetMessageFromWeb(text=prompt_msg) - assert get_msg_from_web.gen_params()["parameter"]["chat"]["domain"] == "xxxxxx" + get_msg_from_web = GetMessageFromWeb(prompt_msg, mock_llm_config) + assert get_msg_from_web.gen_params()["parameter"]["chat"]["domain"] == "mock_domain" ret = get_msg_from_web.run() assert ret == "" @@ -39,11 +34,19 @@ def mock_spark_get_msg_from_web_run(self) -> str: return resp_content +@pytest.mark.asyncio +async def test_spark_aask(): + llm = SparkLLM(Config.from_home("spark.yaml").llm) + + resp = await llm.aask("Hello!") + print(resp) + + @pytest.mark.asyncio async def test_spark_acompletion(mocker): mocker.patch("metagpt.provider.spark_api.GetMessageFromWeb.run", mock_spark_get_msg_from_web_run) - spark_gpt = SparkLLM() + spark_gpt = SparkLLM(mock_llm_config) resp = await spark_gpt.acompletion([]) assert resp == resp_content diff --git a/tests/metagpt/provider/test_zhipuai_api.py b/tests/metagpt/provider/test_zhipuai_api.py index ab240260c..798209710 100644 --- a/tests/metagpt/provider/test_zhipuai_api.py +++ b/tests/metagpt/provider/test_zhipuai_api.py @@ -3,47 +3,25 @@ # @Desc : the unittest of ZhiPuAILLM import pytest -from zhipuai.utils.sse_client import Event -from metagpt.config import CONFIG from metagpt.provider.zhipuai_api import ZhiPuAILLM - -CONFIG.zhipuai_api_key = "xxx.xxx" +from tests.metagpt.provider.mock_llm_config import mock_llm_config_zhipu prompt_msg = "who are you" messages = [{"role": "user", "content": prompt_msg}] resp_content = "I'm chatglm-turbo" default_resp = { - "code": 200, - "data": { - "choices": [{"role": "assistant", "content": resp_content}], - "usage": {"prompt_tokens": 20, "completion_tokens": 20}, - }, + "choices": [{"finish_reason": "stop", "index": 0, "message": {"content": resp_content, "role": "assistant"}}], + "usage": {"completion_tokens": 22, "prompt_tokens": 19, "total_tokens": 41}, } -def mock_zhipuai_invoke(**kwargs) -> dict: - return default_resp - - -async def mock_zhipuai_ainvoke(**kwargs) -> dict: - return default_resp - - -async def mock_zhipuai_asse_invoke(**kwargs): +async def mock_zhipuai_acreate_stream(**kwargs): class MockResponse(object): async def _aread(self): class Iterator(object): - events = [ - Event(id="xxx", event="add", data=resp_content, retry=0), - Event( - id="xxx", - event="finish", - data="", - meta='{"usage": {"completion_tokens": 20,"prompt_tokens": 20}}', - ), - ] + events = [{"choices": [{"index": 0, "delta": {"content": resp_content, "role": "assistant"}}]}] async def __aiter__(self): for event in self.events: @@ -52,23 +30,26 @@ async def mock_zhipuai_asse_invoke(**kwargs): async for chunk in Iterator(): yield chunk - async def async_events(self): + async def stream(self): async for chunk in self._aread(): yield chunk return MockResponse() +async def mock_zhipuai_acreate(**kwargs) -> dict: + return default_resp + + @pytest.mark.asyncio async def test_zhipuai_acompletion(mocker): - mocker.patch("metagpt.provider.zhipuai.zhipu_model_api.ZhiPuModelAPI.invoke", mock_zhipuai_invoke) - mocker.patch("metagpt.provider.zhipuai.zhipu_model_api.ZhiPuModelAPI.ainvoke", mock_zhipuai_ainvoke) - mocker.patch("metagpt.provider.zhipuai.zhipu_model_api.ZhiPuModelAPI.asse_invoke", mock_zhipuai_asse_invoke) + mocker.patch("metagpt.provider.zhipuai.zhipu_model_api.ZhiPuModelAPI.acreate", mock_zhipuai_acreate) + mocker.patch("metagpt.provider.zhipuai.zhipu_model_api.ZhiPuModelAPI.acreate_stream", mock_zhipuai_acreate_stream) - zhipu_gpt = ZhiPuAILLM() + zhipu_gpt = ZhiPuAILLM(mock_llm_config_zhipu) resp = await zhipu_gpt.acompletion(messages) - assert resp["data"]["choices"][0]["content"] == resp_content + assert resp["choices"][0]["message"]["content"] == resp_content resp = await zhipu_gpt.aask(prompt_msg, stream=False) assert resp == resp_content @@ -84,6 +65,7 @@ async def test_zhipuai_acompletion(mocker): def test_zhipuai_proxy(): - # CONFIG.openai_proxy = "http://127.0.0.1:8080" - _ = ZhiPuAILLM() - # assert openai.proxy == CONFIG.openai_proxy + # it seems like zhipuai would be inflected by the proxy of openai, maybe it's a bug + # but someone may want to use openai.proxy, so we keep this test case + # assert openai.proxy == config.llm.proxy + _ = ZhiPuAILLM(mock_llm_config_zhipu) diff --git a/tests/metagpt/provider/zhipuai/test_async_sse_client.py b/tests/metagpt/provider/zhipuai/test_async_sse_client.py index 2649f595b..31b2d3d64 100644 --- a/tests/metagpt/provider/zhipuai/test_async_sse_client.py +++ b/tests/metagpt/provider/zhipuai/test_async_sse_client.py @@ -11,16 +11,16 @@ from metagpt.provider.zhipuai.async_sse_client import AsyncSSEClient async def test_async_sse_client(): class Iterator(object): async def __aiter__(self): - yield b"data: test_value" + yield b'data: {"test_key": "test_value"}' async_sse_client = AsyncSSEClient(event_source=Iterator()) - async for event in async_sse_client.async_events(): - assert event.data, "test_value" + async for chunk in async_sse_client.stream(): + assert "test_value" in chunk.values() class InvalidIterator(object): async def __aiter__(self): yield b"invalid: test_value" async_sse_client = AsyncSSEClient(event_source=InvalidIterator()) - async for event in async_sse_client.async_events(): - assert not event + async for chunk in async_sse_client.stream(): + assert not chunk diff --git a/tests/metagpt/provider/zhipuai/test_zhipu_model_api.py b/tests/metagpt/provider/zhipuai/test_zhipu_model_api.py index 1f0a42fa6..15673c51c 100644 --- a/tests/metagpt/provider/zhipuai/test_zhipu_model_api.py +++ b/tests/metagpt/provider/zhipuai/test_zhipu_model_api.py @@ -6,15 +6,13 @@ from typing import Any, Tuple import pytest import zhipuai -from zhipuai.model_api.api import InvokeType -from zhipuai.utils.http_client import headers as zhipuai_default_headers from metagpt.provider.zhipuai.zhipu_model_api import ZhiPuModelAPI api_key = "xxx.xxx" zhipuai.api_key = api_key -default_resp = b'{"result": "test response"}' +default_resp = b'{"choices": [{"finish_reason": "stop", "index": 0, "message": {"content": "test response", "role": "assistant"}}]}' async def mock_requestor_arequest(self, **kwargs) -> Tuple[Any, Any, str]: @@ -23,22 +21,15 @@ async def mock_requestor_arequest(self, **kwargs) -> Tuple[Any, Any, str]: @pytest.mark.asyncio async def test_zhipu_model_api(mocker): - header = ZhiPuModelAPI.get_header() - zhipuai_default_headers.update({"Authorization": api_key}) - assert header == zhipuai_default_headers - - sse_header = ZhiPuModelAPI.get_sse_header() - assert len(sse_header["Authorization"]) == 191 - - url_prefix, url_suffix = ZhiPuModelAPI.split_zhipu_api_url(InvokeType.SYNC, kwargs={"model": "chatglm_turbo"}) + url_prefix, url_suffix = ZhiPuModelAPI(api_key=api_key).split_zhipu_api_url() assert url_prefix == "https://open.bigmodel.cn/api" - assert url_suffix == "/paas/v3/model-api/chatglm_turbo/invoke" + assert url_suffix == "/paas/v4/chat/completions" mocker.patch("metagpt.provider.general_api_requestor.GeneralAPIRequestor.arequest", mock_requestor_arequest) - result = await ZhiPuModelAPI.arequest( - InvokeType.SYNC, stream=False, method="get", headers={}, kwargs={"model": "chatglm_turbo"} + result = await ZhiPuModelAPI(api_key=api_key).arequest( + stream=False, method="get", headers={}, kwargs={"model": "glm-3-turbo"} ) assert result == default_resp - result = await ZhiPuModelAPI.ainvoke() - assert result["result"] == "test response" + result = await ZhiPuModelAPI(api_key=api_key).acreate() + assert result["choices"][0]["message"]["content"] == "test response" diff --git a/tests/metagpt/roles/run_code_interpreter.py b/tests/metagpt/roles/run_code_interpreter.py index 766a25998..1c5b2873f 100644 --- a/tests/metagpt/roles/run_code_interpreter.py +++ b/tests/metagpt/roles/run_code_interpreter.py @@ -9,9 +9,7 @@ from metagpt.schema import Plan from metagpt.utils.recovery_util import load_history, save_history -async def run_code_interpreter( - role_class, requirement, auto_run, use_tools, use_code_steps, make_udfs, use_udfs, save_dir, tools -): +async def run_code_interpreter(role_class, requirement, auto_run, use_tools, save_dir, tools): """ The main function to run the MLEngineer with optional history loading. @@ -25,16 +23,11 @@ async def run_code_interpreter( """ if role_class == "ci": - role = CodeInterpreter( - goal=requirement, auto_run=auto_run, use_tools=use_tools, make_udfs=make_udfs, tools=tools - ) + role = CodeInterpreter(auto_run=auto_run, use_tools=use_tools, tools=tools) else: role = MLEngineer( - goal=requirement, auto_run=auto_run, use_tools=use_tools, - use_code_steps=use_code_steps, - make_udfs=make_udfs, tools=tools, ) @@ -50,10 +43,10 @@ async def run_code_interpreter( try: await role.run(requirement) except Exception as e: - save_path = save_history(role, save_dir) - logger.exception(f"An error occurred: {e}, save trajectory here: {save_path}") + save_history(role, save_dir) + if __name__ == "__main__": # requirement = "Run data analysis on sklearn Iris dataset, include a plot" @@ -73,8 +66,6 @@ if __name__ == "__main__": role_class = "mle" auto_run = True use_tools = True - make_udfs = False - use_udfs = False tools = [] # tools = ["FillMissingValue", "CatCross", "non_existing_test"] @@ -83,14 +74,9 @@ if __name__ == "__main__": requirement: str = requirement, auto_run: bool = auto_run, use_tools: bool = use_tools, - use_code_steps: bool = False, - make_udfs: bool = make_udfs, - use_udfs: bool = use_udfs, save_dir: str = save_dir, tools=tools, ): - await run_code_interpreter( - role_class, requirement, auto_run, use_tools, use_code_steps, make_udfs, use_udfs, save_dir, tools - ) + await run_code_interpreter(role_class, requirement, auto_run, use_tools, save_dir, tools) fire.Fire(main) diff --git a/tests/metagpt/roles/test_architect.py b/tests/metagpt/roles/test_architect.py index 06e4b2d11..b02242ed2 100644 --- a/tests/metagpt/roles/test_architect.py +++ b/tests/metagpt/roles/test_architect.py @@ -12,7 +12,6 @@ import uuid import pytest from metagpt.actions import WriteDesign, WritePRD -from metagpt.config import CONFIG from metagpt.const import PRDS_FILE_REPO from metagpt.logs import logger from metagpt.roles import Architect @@ -22,12 +21,12 @@ from tests.metagpt.roles.mock import MockMessages @pytest.mark.asyncio -async def test_architect(): +async def test_architect(context): # Prerequisites filename = uuid.uuid4().hex + ".json" - await awrite(CONFIG.git_repo.workdir / PRDS_FILE_REPO / filename, data=MockMessages.prd.content) + await awrite(context.repo.workdir / PRDS_FILE_REPO / filename, data=MockMessages.prd.content) - role = Architect() + role = Architect(context=context) rsp = await role.run(with_message=Message(content="", cause_by=WritePRD)) logger.info(rsp) assert len(rsp.content) > 0 diff --git a/tests/metagpt/roles/test_assistant.py b/tests/metagpt/roles/test_assistant.py index 24096b357..bd0efea35 100644 --- a/tests/metagpt/roles/test_assistant.py +++ b/tests/metagpt/roles/test_assistant.py @@ -12,7 +12,6 @@ from pydantic import BaseModel from metagpt.actions.skill_action import SkillAction from metagpt.actions.talk_action import TalkAction -from metagpt.config import CONFIG from metagpt.memory.brain_memory import BrainMemory from metagpt.roles.assistant import Assistant from metagpt.schema import Message @@ -20,8 +19,11 @@ from metagpt.utils.common import any_to_str @pytest.mark.asyncio -async def test_run(): - CONFIG.language = "Chinese" +async def test_run(mocker, context): + # mock + mocker.patch("metagpt.learn.text_to_image", return_value="http://mock.com/1.png") + + context.kwargs.language = "Chinese" class Input(BaseModel): memory: BrainMemory @@ -65,7 +67,7 @@ async def test_run(): "cause_by": any_to_str(SkillAction), }, ] - CONFIG.agent_skills = [ + agent_skills = [ {"id": 1, "name": "text_to_speech", "type": "builtin", "config": {}, "enabled": True}, {"id": 2, "name": "text_to_image", "type": "builtin", "config": {}, "enabled": True}, {"id": 3, "name": "ai_call", "type": "builtin", "config": {}, "enabled": True}, @@ -77,9 +79,11 @@ async def test_run(): for i in inputs: seed = Input(**i) - CONFIG.language = seed.language - CONFIG.agent_description = seed.agent_description - role = Assistant(language="Chinese") + role = Assistant(language="Chinese", context=context) + role.context.kwargs.language = seed.language + role.context.kwargs.agent_description = seed.agent_description + role.context.kwargs.agent_skills = agent_skills + role.memory = seed.memory # Restore historical conversation content. while True: has_action = await role.think() @@ -110,21 +114,16 @@ async def test_run(): ], ) @pytest.mark.asyncio -async def test_memory(memory): - role = Assistant() +async def test_memory(memory, context): + role = Assistant(context=context) + role.context.kwargs.agent_skills = [] role.load_memory(memory) val = role.get_memory() assert val await role.talk("draw apple") - - agent_skills = CONFIG.agent_skills - CONFIG.agent_skills = [] - try: - await role.think() - finally: - CONFIG.agent_skills = agent_skills + await role.think() assert isinstance(role.rc.todo, TalkAction) diff --git a/tests/metagpt/roles/test_code_interpreter.py b/tests/metagpt/roles/test_code_interpreter.py index 8595b9b15..b78f7a9ef 100644 --- a/tests/metagpt/roles/test_code_interpreter.py +++ b/tests/metagpt/roles/test_code_interpreter.py @@ -7,7 +7,9 @@ from metagpt.roles.code_interpreter import CodeInterpreter @pytest.mark.asyncio async def test_code_interpreter(): requirement = "Run data analysis on sklearn Iris dataset, include a plot" - ci = CodeInterpreter(goal=requirement, auto_run=True, use_tools=False) + tools = [] + + ci = CodeInterpreter(auto_run=True, use_tools=True, tools=tools) rsp = await ci.run(requirement) logger.info(rsp) assert len(rsp.content) > 0 diff --git a/tests/metagpt/roles/test_daml.py b/tests/metagpt/roles/test_daml.py deleted file mode 100644 index 2e2c003d9..000000000 --- a/tests/metagpt/roles/test_daml.py +++ /dev/null @@ -1,50 +0,0 @@ -import pytest -from tqdm import tqdm - -from metagpt.logs import logger -from metagpt.roles.ml_engineer import ExecutePyCode, MLEngineer -from metagpt.schema import Plan - - -def reset(role): - """Restart role with the same goal.""" - role.working_memory.clear() - role.planner.plan = Plan(goal=role.planner.plan.goal) - role.execute_code = ExecutePyCode() - - -async def make_use_tools(requirement: str, auto_run: bool = True): - """make and use tools for requirement.""" - role = MLEngineer(goal=requirement, auto_run=auto_run) - # make udfs - role.use_tools = False - role.use_code_steps = False - role.make_udfs = True - role.use_udfs = False - await role.run(requirement) - # use udfs - reset(role) - role.make_udfs = False - role.use_udfs = True - role.use_code_steps = False - role.use_tools = False - await role.run(requirement) - - -@pytest.mark.asyncio -async def test_make_use_tools(): - requirements = [ - "Run data analysis on sklearn Iris dataset, include a plot", - "Run data analysis on sklearn Diabetes dataset, include a plot", - "Run data analysis on sklearn Wine recognition dataset, include a plot, and train a model to predict wine class (20% as validation), and show validation accuracy", - "Run data analysis on sklearn Wisconsin Breast Cancer dataset, include a plot, train a model to predict targets (20% as validation), and show validation accuracy", - "Run EDA and visualization on this dataset, train a model to predict survival, report metrics on validation set (20%), dataset: tests/data/titanic.csv", - ] - success = 0 - for requirement in tqdm(requirements, total=len(requirements)): - try: - await make_use_tools(requirement) - success += 1 - except Exception as e: - logger.error(f"Found Error in {requirement}, {e}") - logger.info(f"success: {round(success/len(requirements), 1)*100}%") diff --git a/tests/metagpt/roles/test_engineer.py b/tests/metagpt/roles/test_engineer.py index d03aea0a6..d263a8a2f 100644 --- a/tests/metagpt/roles/test_engineer.py +++ b/tests/metagpt/roles/test_engineer.py @@ -13,40 +13,30 @@ from pathlib import Path import pytest from metagpt.actions import WriteCode, WriteTasks -from metagpt.config import CONFIG -from metagpt.const import ( - PRDS_FILE_REPO, - REQUIREMENT_FILENAME, - SYSTEM_DESIGN_FILE_REPO, - TASK_FILE_REPO, -) +from metagpt.const import REQUIREMENT_FILENAME, SYSTEM_DESIGN_FILE_REPO, TASK_FILE_REPO from metagpt.logs import logger from metagpt.roles.engineer import Engineer from metagpt.schema import CodingContext, Message from metagpt.utils.common import CodeParser, any_to_name, any_to_str, aread, awrite -from metagpt.utils.file_repository import FileRepository from metagpt.utils.git_repository import ChangeType from tests.metagpt.roles.mock import STRS_FOR_PARSING, TASKS, MockMessages @pytest.mark.asyncio -async def test_engineer(): +async def test_engineer(context): # Prerequisites rqno = "20231221155954.json" - await FileRepository.save_file(REQUIREMENT_FILENAME, content=MockMessages.req.content) - await FileRepository.save_file(rqno, relative_path=PRDS_FILE_REPO, content=MockMessages.prd.content) - await FileRepository.save_file( - rqno, relative_path=SYSTEM_DESIGN_FILE_REPO, content=MockMessages.system_design.content - ) - await FileRepository.save_file(rqno, relative_path=TASK_FILE_REPO, content=MockMessages.json_tasks.content) + await context.repo.save(REQUIREMENT_FILENAME, content=MockMessages.req.content) + await context.repo.docs.prd.save(rqno, content=MockMessages.prd.content) + await context.repo.docs.system_design.save(rqno, content=MockMessages.system_design.content) + await context.repo.docs.task.save(rqno, content=MockMessages.json_tasks.content) - engineer = Engineer() + engineer = Engineer(context=context) rsp = await engineer.run(Message(content="", cause_by=WriteTasks)) logger.info(rsp) assert rsp.cause_by == any_to_str(WriteCode) - src_file_repo = CONFIG.git_repo.new_file_repository(CONFIG.src_workspace) - assert src_file_repo.changed_files + assert context.repo.with_src_path(context.src_workspace).srcs.changed_files def test_parse_str(): @@ -109,54 +99,52 @@ def test_parse_code(): def test_todo(): role = Engineer() - assert role.todo == any_to_name(WriteCode) + assert role.action_description == any_to_name(WriteCode) @pytest.mark.asyncio -async def test_new_coding_context(): +async def test_new_coding_context(context): # Prerequisites demo_path = Path(__file__).parent / "../../data/demo_project" deps = json.loads(await aread(demo_path / "dependencies.json")) - dependency = await CONFIG.git_repo.get_dependency() + dependency = await context.git_repo.get_dependency() for k, v in deps.items(): await dependency.update(k, set(v)) data = await aread(demo_path / "system_design.json") rqno = "20231221155954.json" - await awrite(CONFIG.git_repo.workdir / SYSTEM_DESIGN_FILE_REPO / rqno, data) + await awrite(context.repo.workdir / SYSTEM_DESIGN_FILE_REPO / rqno, data) data = await aread(demo_path / "tasks.json") - await awrite(CONFIG.git_repo.workdir / TASK_FILE_REPO / rqno, data) + await awrite(context.repo.workdir / TASK_FILE_REPO / rqno, data) - CONFIG.src_workspace = Path(CONFIG.git_repo.workdir) / "game_2048" - src_file_repo = CONFIG.git_repo.new_file_repository(relative_path=CONFIG.src_workspace) - task_file_repo = CONFIG.git_repo.new_file_repository(relative_path=TASK_FILE_REPO) - design_file_repo = CONFIG.git_repo.new_file_repository(relative_path=SYSTEM_DESIGN_FILE_REPO) + context.src_workspace = Path(context.repo.workdir) / "game_2048" - filename = "game.py" - ctx_doc = await Engineer._new_coding_doc( - filename=filename, - src_file_repo=src_file_repo, - task_file_repo=task_file_repo, - design_file_repo=design_file_repo, - dependency=dependency, - ) - assert ctx_doc - assert ctx_doc.filename == filename - assert ctx_doc.content - ctx = CodingContext.model_validate_json(ctx_doc.content) - assert ctx.filename == filename - assert ctx.design_doc - assert ctx.design_doc.content - assert ctx.task_doc - assert ctx.task_doc.content - assert ctx.code_doc + try: + filename = "game.py" + engineer = Engineer(context=context) + ctx_doc = await engineer._new_coding_doc( + filename=filename, + dependency=dependency, + ) + assert ctx_doc + assert ctx_doc.filename == filename + assert ctx_doc.content + ctx = CodingContext.model_validate_json(ctx_doc.content) + assert ctx.filename == filename + assert ctx.design_doc + assert ctx.design_doc.content + assert ctx.task_doc + assert ctx.task_doc.content + assert ctx.code_doc - CONFIG.git_repo.add_change({f"{TASK_FILE_REPO}/{rqno}": ChangeType.UNTRACTED}) - CONFIG.git_repo.commit("mock env") - await src_file_repo.save(filename=filename, content="content") - role = Engineer() - assert not role.code_todos - await role._new_code_actions() - assert role.code_todos + context.git_repo.add_change({f"{TASK_FILE_REPO}/{rqno}": ChangeType.UNTRACTED}) + context.git_repo.commit("mock env") + await context.repo.with_src_path(context.src_workspace).srcs.save(filename=filename, content="content") + role = Engineer(context=context) + assert not role.code_todos + await role._new_code_actions() + assert role.code_todos + finally: + context.git_repo.delete_repository() if __name__ == "__main__": diff --git a/tests/metagpt/roles/test_invoice_ocr_assistant.py b/tests/metagpt/roles/test_invoice_ocr_assistant.py index e3a9259da..bedcd6712 100644 --- a/tests/metagpt/roles/test_invoice_ocr_assistant.py +++ b/tests/metagpt/roles/test_invoice_ocr_assistant.py @@ -41,9 +41,11 @@ from metagpt.schema import Message ), ], ) -async def test_invoice_ocr_assistant(query: str, invoice_path: Path, invoice_table_path: Path, expected_result: dict): +async def test_invoice_ocr_assistant( + query: str, invoice_path: Path, invoice_table_path: Path, expected_result: dict, context +): invoice_path = TEST_DATA_PATH / invoice_path - role = InvoiceOCRAssistant() + role = InvoiceOCRAssistant(context=context) await role.run(Message(content=query, instruct_content=InvoicePath(file_path=invoice_path))) invoice_table_path = DATA_PATH / invoice_table_path df = pd.read_excel(invoice_table_path) diff --git a/tests/metagpt/roles/test_ml_engineer.py b/tests/metagpt/roles/test_ml_engineer.py new file mode 100644 index 000000000..1373213a5 --- /dev/null +++ b/tests/metagpt/roles/test_ml_engineer.py @@ -0,0 +1,21 @@ +import pytest + +from metagpt.logs import logger +from metagpt.roles.ml_engineer import MLEngineer + + +def test_mle_init(): + ci = MLEngineer(goal="test", auto_run=True, use_tools=True, tools=["tool1", "tool2"]) + assert ci.tools == [] + + +@pytest.mark.asyncio +async def test_ml_engineer(): + data_path = "tests/data/ml_datasets/titanic" + requirement = f"This is a titanic passenger survival dataset, your goal is to predict passenger survival outcome. The target column is Survived. Perform data analysis, data preprocessing, feature engineering, and modeling to predict the target. Report accuracy on the eval data. Train data path: '{data_path}/split_train.csv', eval data path: '{data_path}/split_eval.csv'." + tools = ["FillMissingValue", "CatCross", "dummy_tool"] + + mle = MLEngineer(goal=requirement, auto_run=True, use_tools=True, tools=tools) + rsp = await mle.run(requirement) + logger.info(rsp) + assert len(rsp.content) > 0 diff --git a/tests/metagpt/roles/test_product_manager.py b/tests/metagpt/roles/test_product_manager.py index 1083e81b0..59b5aa81a 100644 --- a/tests/metagpt/roles/test_product_manager.py +++ b/tests/metagpt/roles/test_product_manager.py @@ -5,17 +5,51 @@ @Author : alexanderwu @File : test_product_manager.py """ +import json + import pytest +from metagpt.actions import WritePRD +from metagpt.actions.prepare_documents import PrepareDocuments +from metagpt.const import REQUIREMENT_FILENAME +from metagpt.context import Context from metagpt.logs import logger from metagpt.roles import ProductManager +from metagpt.utils.common import any_to_str from tests.metagpt.roles.mock import MockMessages @pytest.mark.asyncio async def test_product_manager(new_filename): - product_manager = ProductManager() - rsp = await product_manager.run(MockMessages.req) - logger.info(rsp) - assert len(rsp.content) > 0 - assert rsp.content == MockMessages.req.content + context = Context() + try: + assert context.git_repo is None + assert context.repo is None + product_manager = ProductManager(context=context) + # prepare documents + rsp = await product_manager.run(MockMessages.req) + assert context.git_repo + assert context.repo + assert rsp.cause_by == any_to_str(PrepareDocuments) + assert REQUIREMENT_FILENAME in context.repo.docs.changed_files + + # write prd + rsp = await product_manager.run(rsp) + assert rsp.cause_by == any_to_str(WritePRD) + logger.info(rsp) + assert len(rsp.content) > 0 + doc = list(rsp.instruct_content.docs.values())[0] + m = json.loads(doc.content) + assert m["Original Requirements"] == MockMessages.req.content + + # nothing to do + rsp = await product_manager.run(rsp) + assert rsp is None + except Exception as e: + assert not e + finally: + context.git_repo.delete_repository() + + +if __name__ == "__main__": + pytest.main([__file__, "-s"]) diff --git a/tests/metagpt/roles/test_project_manager.py b/tests/metagpt/roles/test_project_manager.py index 9207623bc..9b016927e 100644 --- a/tests/metagpt/roles/test_project_manager.py +++ b/tests/metagpt/roles/test_project_manager.py @@ -13,7 +13,7 @@ from tests.metagpt.roles.mock import MockMessages @pytest.mark.asyncio -async def test_project_manager(): - project_manager = ProjectManager() +async def test_project_manager(context): + project_manager = ProjectManager(context=context) rsp = await project_manager.run(MockMessages.system_design) logger.info(rsp) diff --git a/tests/metagpt/roles/test_qa_engineer.py b/tests/metagpt/roles/test_qa_engineer.py index 784c26a06..b89e7d5eb 100644 --- a/tests/metagpt/roles/test_qa_engineer.py +++ b/tests/metagpt/roles/test_qa_engineer.py @@ -13,20 +13,19 @@ from pydantic import Field from metagpt.actions import DebugError, RunCode, WriteTest from metagpt.actions.summarize_code import SummarizeCode -from metagpt.config import CONFIG from metagpt.environment import Environment from metagpt.roles import QaEngineer from metagpt.schema import Message from metagpt.utils.common import any_to_str, aread, awrite -async def test_qa(): +async def test_qa(context): # Prerequisites demo_path = Path(__file__).parent / "../../data/demo_project" - CONFIG.src_workspace = Path(CONFIG.git_repo.workdir) / "qa/game_2048" + context.src_workspace = Path(context.repo.workdir) / "qa/game_2048" data = await aread(filename=demo_path / "game.py", encoding="utf-8") - await awrite(filename=CONFIG.src_workspace / "game.py", data=data, encoding="utf-8") - await awrite(filename=Path(CONFIG.git_repo.workdir) / "requirements.txt", data="") + await awrite(filename=context.src_workspace / "game.py", data=data, encoding="utf-8") + await awrite(filename=Path(context.repo.workdir) / "requirements.txt", data="") class MockEnv(Environment): msgs: List[Message] = Field(default_factory=list) @@ -37,7 +36,7 @@ async def test_qa(): env = MockEnv() - role = QaEngineer() + role = QaEngineer(context=context) role.set_env(env) await role.run(with_message=Message(content="", cause_by=SummarizeCode)) assert env.msgs diff --git a/tests/metagpt/roles/test_researcher.py b/tests/metagpt/roles/test_researcher.py index 891befa38..af81777ac 100644 --- a/tests/metagpt/roles/test_researcher.py +++ b/tests/metagpt/roles/test_researcher.py @@ -4,7 +4,10 @@ from tempfile import TemporaryDirectory import pytest +from metagpt.actions.research import CollectLinks from metagpt.roles import researcher +from metagpt.tools import SearchEngineType +from metagpt.tools.search_engine import SearchEngine async def mock_llm_ask(self, prompt: str, system_msgs): @@ -25,16 +28,20 @@ async def mock_llm_ask(self, prompt: str, system_msgs): @pytest.mark.asyncio -async def test_researcher(mocker): +async def test_researcher(mocker, search_engine_mocker, context): with TemporaryDirectory() as dirname: topic = "dataiku vs. datarobot" mocker.patch("metagpt.provider.base_llm.BaseLLM.aask", mock_llm_ask) researcher.RESEARCH_PATH = Path(dirname) - await researcher.Researcher().run(topic) + role = researcher.Researcher(context=context) + for i in role.actions: + if isinstance(i, CollectLinks): + i.search_engine = SearchEngine(SearchEngineType.DUCK_DUCK_GO) + await role.run(topic) assert (researcher.RESEARCH_PATH / f"{topic}.md").read_text().startswith("# Research Report") -def test_write_report(mocker): +def test_write_report(mocker, context): with TemporaryDirectory() as dirname: for i, topic in enumerate( [ @@ -46,7 +53,7 @@ def test_write_report(mocker): ): researcher.RESEARCH_PATH = Path(dirname) content = "# Research Report" - researcher.Researcher().write_report(topic, content) + researcher.Researcher(context=context).write_report(topic, content) assert (researcher.RESEARCH_PATH / f"{i+1}. metagpt.md").read_text().startswith("# Research Report") diff --git a/tests/metagpt/roles/test_role.py b/tests/metagpt/roles/test_role.py index bef71f9a5..8b11e2d4a 100644 --- a/tests/metagpt/roles/test_role.py +++ b/tests/metagpt/roles/test_role.py @@ -3,7 +3,7 @@ # @Desc : unittest of Role import pytest -from metagpt.llm import HumanProvider +from metagpt.provider.human_provider import HumanProvider from metagpt.roles.role import Role @@ -13,8 +13,8 @@ def test_role_desc(): assert role.desc == "Best Seller" -def test_role_human(): - role = Role(is_human=True) +def test_role_human(context): + role = Role(is_human=True, context=context) assert isinstance(role.llm, HumanProvider) diff --git a/tests/metagpt/roles/test_teacher.py b/tests/metagpt/roles/test_teacher.py index 1efc329db..83a7e382a 100644 --- a/tests/metagpt/roles/test_teacher.py +++ b/tests/metagpt/roles/test_teacher.py @@ -5,19 +5,17 @@ @Author : mashenquan @File : test_teacher.py """ -import os from typing import Dict, Optional import pytest -from pydantic import BaseModel +from pydantic import BaseModel, Field -from metagpt.config import CONFIG, Config +from metagpt.context import Context from metagpt.roles.teacher import Teacher from metagpt.schema import Message @pytest.mark.asyncio -@pytest.mark.skip async def test_init(): class Inputs(BaseModel): name: str @@ -31,6 +29,7 @@ async def test_init(): expect_goal: str expect_constraints: str expect_desc: str + exclude: list = Field(default_factory=list) inputs = [ { @@ -45,6 +44,7 @@ async def test_init(): "kwargs": {}, "desc": "aaa{language}", "expect_desc": "aaa{language}", + "exclude": ["language", "key1", "something_big", "teaching_language"], }, { "name": "Lily{language}", @@ -58,20 +58,21 @@ async def test_init(): "kwargs": {"language": "CN", "key1": "HaHa", "something_big": "sleep", "teaching_language": "EN"}, "desc": "aaa{language}", "expect_desc": "aaaCN", + "language": "CN", + "teaching_language": "EN", }, ] - env = os.environ.copy() for i in inputs: seed = Inputs(**i) - os.environ.clear() - os.environ.update(env) - CONFIG = Config() - CONFIG.set_context(seed.kwargs) - print(CONFIG.options) - assert bool("language" in seed.kwargs) == bool("language" in CONFIG.options) + context = Context() + for k in seed.exclude: + context.kwargs.set(k, None) + for k, v in seed.kwargs.items(): + context.kwargs.set(k, v) teacher = Teacher( + context=context, name=seed.name, profile=seed.profile, goal=seed.goal, @@ -105,7 +106,6 @@ async def test_new_file_name(): @pytest.mark.asyncio async def test_run(): - CONFIG.set_context({"language": "Chinese", "teaching_language": "English"}) lesson = """ UNIT 1 Making New Friends TOPIC 1 Welcome to China! @@ -149,7 +149,10 @@ async def test_run(): 3c Match the big letters with the small ones. Then write them on the lines. """ - teacher = Teacher() + context = Context() + context.kwargs.language = "Chinese" + context.kwargs.teaching_language = "English" + teacher = Teacher(context=context) rsp = await teacher.run(Message(content=lesson)) assert rsp diff --git a/tests/metagpt/roles/test_tutorial_assistant.py b/tests/metagpt/roles/test_tutorial_assistant.py index 0e6c1efb9..c12c2b26e 100644 --- a/tests/metagpt/roles/test_tutorial_assistant.py +++ b/tests/metagpt/roles/test_tutorial_assistant.py @@ -15,8 +15,8 @@ from metagpt.roles.tutorial_assistant import TutorialAssistant @pytest.mark.asyncio @pytest.mark.parametrize(("language", "topic"), [("Chinese", "Write a tutorial about pip")]) -async def test_tutorial_assistant(language: str, topic: str): - role = TutorialAssistant(language=language) +async def test_tutorial_assistant(language: str, topic: str, context): + role = TutorialAssistant(language=language, context=context) msg = await role.run(topic) assert TUTORIAL_PATH.exists() filename = msg.content diff --git a/tests/metagpt/serialize_deserialize/test_action.py b/tests/metagpt/serialize_deserialize/test_action.py index 81879e34e..d234a160f 100644 --- a/tests/metagpt/serialize_deserialize/test_action.py +++ b/tests/metagpt/serialize_deserialize/test_action.py @@ -5,28 +5,22 @@ import pytest from metagpt.actions import Action -from metagpt.llm import LLM - - -def test_action_serialize(): - action = Action() - ser_action_dict = action.model_dump() - assert "name" in ser_action_dict - assert "llm" not in ser_action_dict # not export - assert "__module_class_name" not in ser_action_dict - - action = Action(name="test") - ser_action_dict = action.model_dump() - assert "test" in ser_action_dict["name"] @pytest.mark.asyncio -async def test_action_deserialize(): - action = Action() - serialized_data = action.model_dump() +async def test_action_serdeser(context): + action = Action(context=context) + ser_action_dict = action.model_dump() + assert "name" in ser_action_dict + assert "llm" not in ser_action_dict # not export + assert "__module_class_name" in ser_action_dict - new_action = Action(**serialized_data) + action = Action(name="test", context=context) + ser_action_dict = action.model_dump() + assert "test" in ser_action_dict["name"] - assert new_action.name == "Action" - assert isinstance(new_action.llm, type(LLM())) + new_action = Action(**ser_action_dict, context=context) + + assert new_action.name == "test" + assert isinstance(new_action.llm, type(context.llm())) assert len(await new_action._aask("who are you")) > 0 diff --git a/tests/metagpt/serialize_deserialize/test_architect_deserialize.py b/tests/metagpt/serialize_deserialize/test_architect.py similarity index 69% rename from tests/metagpt/serialize_deserialize/test_architect_deserialize.py rename to tests/metagpt/serialize_deserialize/test_architect.py index b113912a7..e3c2703fa 100644 --- a/tests/metagpt/serialize_deserialize/test_architect_deserialize.py +++ b/tests/metagpt/serialize_deserialize/test_architect.py @@ -8,21 +8,21 @@ from metagpt.actions.action import Action from metagpt.roles.architect import Architect -def test_architect_serialize(): - role = Architect() +@pytest.mark.asyncio +async def test_architect_serdeser(context): + role = Architect(context=context) ser_role_dict = role.model_dump(by_alias=True) assert "name" in ser_role_dict assert "states" in ser_role_dict assert "actions" in ser_role_dict - -@pytest.mark.asyncio -async def test_architect_deserialize(): - role = Architect() - ser_role_dict = role.model_dump(by_alias=True) - new_role = Architect(**ser_role_dict) - # new_role = Architect.deserialize(ser_role_dict) + new_role = Architect(**ser_role_dict, context=context) assert new_role.name == "Bob" assert len(new_role.actions) == 1 + assert len(new_role.rc.watch) == 1 assert isinstance(new_role.actions[0], Action) await new_role.actions[0].run(with_messages="write a cli snake game") + + +if __name__ == "__main__": + pytest.main([__file__, "-s"]) diff --git a/tests/metagpt/serialize_deserialize/test_environment.py b/tests/metagpt/serialize_deserialize/test_environment.py index 5a68288a6..4e6ea93b5 100644 --- a/tests/metagpt/serialize_deserialize/test_environment.py +++ b/tests/metagpt/serialize_deserialize/test_environment.py @@ -2,7 +2,6 @@ # -*- coding: utf-8 -*- # @Desc : -import shutil from metagpt.actions.action_node import ActionNode from metagpt.actions.add_requirement import UserRequirement @@ -10,7 +9,7 @@ from metagpt.actions.project_management import WriteTasks from metagpt.environment import Environment from metagpt.roles.project_manager import ProjectManager from metagpt.schema import Message -from metagpt.utils.common import any_to_str +from metagpt.utils.common import any_to_str, read_json_file, write_json_file from tests.metagpt.serialize_deserialize.test_serdeser_base import ( ActionOK, ActionRaise, @@ -19,23 +18,20 @@ from tests.metagpt.serialize_deserialize.test_serdeser_base import ( ) -def test_env_serialize(): - env = Environment() +def test_env_serdeser(context): + env = Environment(context=context) + env.publish_message(message=Message(content="test env serialize")) + ser_env_dict = env.model_dump() assert "roles" in ser_env_dict assert len(ser_env_dict["roles"]) == 0 - -def test_env_deserialize(): - env = Environment() - env.publish_message(message=Message(content="test env serialize")) - ser_env_dict = env.model_dump() - new_env = Environment(**ser_env_dict) + new_env = Environment(**ser_env_dict, context=context) assert len(new_env.roles) == 0 assert len(new_env.history) == 25 -def test_environment_serdeser(): +def test_environment_serdeser(context): out_mapping = {"field1": (list[str], ...)} out_data = {"field1": ["field1 value1", "field1 value2"]} ic_obj = ActionNode.create_model_class("prd", out_mapping) @@ -44,7 +40,7 @@ def test_environment_serdeser(): content="prd", instruct_content=ic_obj(**out_data), role="product manager", cause_by=any_to_str(UserRequirement) ) - environment = Environment() + environment = Environment(context=context) role_c = RoleC() environment.add_role(role_c) environment.publish_message(message) @@ -52,7 +48,7 @@ def test_environment_serdeser(): ser_data = environment.model_dump() assert ser_data["roles"]["Role C"]["name"] == "RoleC" - new_env: Environment = Environment(**ser_data) + new_env: Environment = Environment(**ser_data, context=context) assert len(new_env.roles) == 1 assert list(new_env.roles.values())[0].states == list(environment.roles.values())[0].states @@ -61,30 +57,31 @@ def test_environment_serdeser(): assert type(list(new_env.roles.values())[0].actions[1]) == ActionRaise -def test_environment_serdeser_v2(): - environment = Environment() +def test_environment_serdeser_v2(context): + environment = Environment(context=context) pm = ProjectManager() environment.add_role(pm) ser_data = environment.model_dump() - new_env: Environment = Environment(**ser_data) + new_env: Environment = Environment(**ser_data, context=context) role = new_env.get_role(pm.profile) assert isinstance(role, ProjectManager) assert isinstance(role.actions[0], WriteTasks) assert isinstance(list(new_env.roles.values())[0].actions[0], WriteTasks) -def test_environment_serdeser_save(): - environment = Environment() +def test_environment_serdeser_save(context): + environment = Environment(context=context) role_c = RoleC() - shutil.rmtree(serdeser_path.joinpath("team"), ignore_errors=True) - stg_path = serdeser_path.joinpath("team", "environment") + env_path = stg_path.joinpath("env.json") environment.add_role(role_c) - environment.serialize(stg_path) - new_env: Environment = Environment.deserialize(stg_path) + write_json_file(env_path, environment.model_dump()) + + env_dict = read_json_file(env_path) + new_env: Environment = Environment(**env_dict, context=context) assert len(new_env.roles) == 1 assert type(list(new_env.roles.values())[0].actions[0]) == ActionOK diff --git a/tests/metagpt/serialize_deserialize/test_memory.py b/tests/metagpt/serialize_deserialize/test_memory.py index aa3e2a465..560ae2c51 100644 --- a/tests/metagpt/serialize_deserialize/test_memory.py +++ b/tests/metagpt/serialize_deserialize/test_memory.py @@ -9,11 +9,11 @@ from metagpt.actions.add_requirement import UserRequirement from metagpt.actions.design_api import WriteDesign from metagpt.memory.memory import Memory from metagpt.schema import Message -from metagpt.utils.common import any_to_str +from metagpt.utils.common import any_to_str, read_json_file, write_json_file from tests.metagpt.serialize_deserialize.test_serdeser_base import serdeser_path -def test_memory_serdeser(): +def test_memory_serdeser(context): msg1 = Message(role="Boss", content="write a snake game", cause_by=UserRequirement) out_mapping = {"field2": (list[str], ...)} @@ -39,7 +39,7 @@ def test_memory_serdeser(): assert memory.count() == 2 -def test_memory_serdeser_save(): +def test_memory_serdeser_save(context): msg1 = Message(role="User", content="write a 2048 game", cause_by=UserRequirement) out_mapping = {"field1": (list[str], ...)} @@ -53,14 +53,14 @@ def test_memory_serdeser_save(): memory.add_batch([msg1, msg2]) stg_path = serdeser_path.joinpath("team", "environment") - memory.serialize(stg_path) - assert stg_path.joinpath("memory.json").exists() + memory_path = stg_path.joinpath("memory.json") + write_json_file(memory_path, memory.model_dump()) + assert memory_path.exists() - new_memory = Memory.deserialize(stg_path) + memory_dict = read_json_file(memory_path) + new_memory = Memory(**memory_dict) assert new_memory.count() == 2 new_msg2 = new_memory.get(1)[0] assert new_msg2.instruct_content.field1 == ["field1 value1", "field1 value2"] assert new_msg2.cause_by == any_to_str(WriteDesign) assert len(new_memory.index) == 2 - - stg_path.joinpath("memory.json").unlink() diff --git a/tests/metagpt/serialize_deserialize/test_polymorphic.py b/tests/metagpt/serialize_deserialize/test_polymorphic.py index ed0482c34..e5f8ec8d6 100644 --- a/tests/metagpt/serialize_deserialize/test_polymorphic.py +++ b/tests/metagpt/serialize_deserialize/test_polymorphic.py @@ -1,6 +1,7 @@ #!/usr/bin/env python # -*- coding: utf-8 -*- # @Desc : unittest of polymorphic conditions +import copy from pydantic import BaseModel, ConfigDict, SerializeAsAny @@ -12,6 +13,8 @@ from tests.metagpt.serialize_deserialize.test_serdeser_base import ( class ActionSubClasses(BaseModel): + model_config = ConfigDict(arbitrary_types_allowed=True) + actions: list[SerializeAsAny[Action]] = [] @@ -40,19 +43,21 @@ def test_no_serialize_as_any(): def test_polymorphic(): - _ = ActionOKV2( + ok_v2 = ActionOKV2( **{"name": "ActionOKV2", "context": "", "prefix": "", "desc": "", "extra_field": "ActionOKV2 Extra Info"} ) action_subcls = ActionSubClasses(actions=[ActionOKV2(), ActionPass()]) action_subcls_dict = action_subcls.model_dump() + action_subcls_dict2 = copy.deepcopy(action_subcls_dict) assert "__module_class_name" in action_subcls_dict["actions"][0] new_action_subcls = ActionSubClasses(**action_subcls_dict) assert isinstance(new_action_subcls.actions[0], ActionOKV2) + assert new_action_subcls.actions[0].extra_field == ok_v2.extra_field assert isinstance(new_action_subcls.actions[1], ActionPass) - new_action_subcls = ActionSubClasses.model_validate(action_subcls_dict) + new_action_subcls = ActionSubClasses.model_validate(action_subcls_dict2) assert isinstance(new_action_subcls.actions[0], ActionOKV2) assert isinstance(new_action_subcls.actions[1], ActionPass) diff --git a/tests/metagpt/serialize_deserialize/test_prepare_interview.py b/tests/metagpt/serialize_deserialize/test_prepare_interview.py index cd9912103..a3e3edafc 100644 --- a/tests/metagpt/serialize_deserialize/test_prepare_interview.py +++ b/tests/metagpt/serialize_deserialize/test_prepare_interview.py @@ -8,12 +8,12 @@ from metagpt.actions.prepare_interview import PrepareInterview @pytest.mark.asyncio -async def test_action_deserialize(): - action = PrepareInterview() +async def test_action_serdeser(context): + action = PrepareInterview(context=context) serialized_data = action.model_dump() assert serialized_data["name"] == "PrepareInterview" - new_action = PrepareInterview(**serialized_data) + new_action = PrepareInterview(**serialized_data, context=context) assert new_action.name == "PrepareInterview" assert type(await new_action.run("python developer")) == ActionNode diff --git a/tests/metagpt/serialize_deserialize/test_product_manager.py b/tests/metagpt/serialize_deserialize/test_product_manager.py index 094943900..2338b406d 100644 --- a/tests/metagpt/serialize_deserialize/test_product_manager.py +++ b/tests/metagpt/serialize_deserialize/test_product_manager.py @@ -10,10 +10,10 @@ from metagpt.schema import Message @pytest.mark.asyncio -async def test_product_manager_deserialize(new_filename): - role = ProductManager() +async def test_product_manager_serdeser(new_filename, context): + role = ProductManager(context=context) ser_role_dict = role.model_dump(by_alias=True) - new_role = ProductManager(**ser_role_dict) + new_role = ProductManager(**ser_role_dict, context=context) assert new_role.name == "Alice" assert len(new_role.actions) == 2 diff --git a/tests/metagpt/serialize_deserialize/test_project_manager.py b/tests/metagpt/serialize_deserialize/test_project_manager.py index 1088a4461..fb998ae31 100644 --- a/tests/metagpt/serialize_deserialize/test_project_manager.py +++ b/tests/metagpt/serialize_deserialize/test_project_manager.py @@ -9,20 +9,15 @@ from metagpt.actions.project_management import WriteTasks from metagpt.roles.project_manager import ProjectManager -def test_project_manager_serialize(): - role = ProjectManager() +@pytest.mark.asyncio +async def test_project_manager_serdeser(context): + role = ProjectManager(context=context) ser_role_dict = role.model_dump(by_alias=True) assert "name" in ser_role_dict assert "states" in ser_role_dict assert "actions" in ser_role_dict - -@pytest.mark.asyncio -async def test_project_manager_deserialize(): - role = ProjectManager() - ser_role_dict = role.model_dump(by_alias=True) - - new_role = ProjectManager(**ser_role_dict) + new_role = ProjectManager(**ser_role_dict, context=context) assert new_role.name == "Eve" assert len(new_role.actions) == 1 assert isinstance(new_role.actions[0], Action) diff --git a/tests/metagpt/serialize_deserialize/test_reasearcher.py b/tests/metagpt/serialize_deserialize/test_reasearcher.py index 1b8dbf2c7..67c52e692 100644 --- a/tests/metagpt/serialize_deserialize/test_reasearcher.py +++ b/tests/metagpt/serialize_deserialize/test_reasearcher.py @@ -8,13 +8,13 @@ from metagpt.roles.researcher import Researcher @pytest.mark.asyncio -async def test_tutorial_assistant_deserialize(): - role = Researcher() +async def test_tutorial_assistant_serdeser(context): + role = Researcher(context=context) ser_role_dict = role.model_dump() assert "name" in ser_role_dict assert "language" in ser_role_dict - new_role = Researcher(**ser_role_dict) + new_role = Researcher(**ser_role_dict, context=context) assert new_role.language == "en-us" assert len(new_role.actions) == 3 assert isinstance(new_role.actions[0], CollectLinks) diff --git a/tests/metagpt/serialize_deserialize/test_role.py b/tests/metagpt/serialize_deserialize/test_role.py index d38797baf..aaf7c1935 100644 --- a/tests/metagpt/serialize_deserialize/test_role.py +++ b/tests/metagpt/serialize_deserialize/test_role.py @@ -10,13 +10,12 @@ from pydantic import BaseModel, SerializeAsAny from metagpt.actions import WriteCode from metagpt.actions.add_requirement import UserRequirement -from metagpt.const import SERDESER_PATH from metagpt.logs import logger from metagpt.roles.engineer import Engineer from metagpt.roles.product_manager import ProductManager from metagpt.roles.role import Role from metagpt.schema import Message -from metagpt.utils.common import format_trackback_info +from metagpt.utils.common import format_trackback_info, read_json_file, write_json_file from tests.metagpt.serialize_deserialize.test_serdeser_base import ( ActionOK, RoleA, @@ -27,7 +26,7 @@ from tests.metagpt.serialize_deserialize.test_serdeser_base import ( ) -def test_roles(): +def test_roles(context): role_a = RoleA() assert len(role_a.rc.watch) == 1 role_b = RoleB() @@ -38,7 +37,7 @@ def test_roles(): assert len(role_d.actions) == 1 -def test_role_subclasses(): +def test_role_subclasses(context): """test subclasses of role with same fields in ser&deser""" class RoleSubClasses(BaseModel): @@ -52,7 +51,7 @@ def test_role_subclasses(): assert isinstance(new_role_subcls.roles[1], RoleB) -def test_role_serialize(): +def test_role_serialize(context): role = Role() ser_role_dict = role.model_dump() assert "name" in ser_role_dict @@ -60,60 +59,56 @@ def test_role_serialize(): assert "actions" in ser_role_dict -def test_engineer_serialize(): +def test_engineer_serdeser(context): role = Engineer() ser_role_dict = role.model_dump() assert "name" in ser_role_dict assert "states" in ser_role_dict assert "actions" in ser_role_dict - -@pytest.mark.asyncio -async def test_engineer_deserialize(): - role = Engineer(use_code_review=True) - ser_role_dict = role.model_dump() - new_role = Engineer(**ser_role_dict) assert new_role.name == "Alex" - assert new_role.use_code_review is True + assert new_role.use_code_review is False assert len(new_role.actions) == 1 assert isinstance(new_role.actions[0], WriteCode) - # await new_role.actions[0].run(context="write a cli snake game", filename="test_code") -def test_role_serdeser_save(): - stg_path_prefix = serdeser_path.joinpath("team", "environment", "roles") +def test_role_serdeser_save(context): shutil.rmtree(serdeser_path.joinpath("team"), ignore_errors=True) pm = ProductManager() - role_tag = f"{pm.__class__.__name__}_{pm.name}" - stg_path = stg_path_prefix.joinpath(role_tag) - pm.serialize(stg_path) - new_pm = Role.deserialize(stg_path) + stg_path = serdeser_path.joinpath("team", "environment", "roles", f"{pm.__class__.__name__}_{pm.name}") + role_path = stg_path.joinpath("role.json") + write_json_file(role_path, pm.model_dump()) + + role_dict = read_json_file(role_path) + new_pm = ProductManager(**role_dict) assert new_pm.name == pm.name assert len(new_pm.get_memories(1)) == 0 @pytest.mark.asyncio -async def test_role_serdeser_interrupt(): +async def test_role_serdeser_interrupt(context): role_c = RoleC() - shutil.rmtree(SERDESER_PATH.joinpath("team"), ignore_errors=True) + shutil.rmtree(serdeser_path.joinpath("team"), ignore_errors=True) - stg_path = SERDESER_PATH.joinpath("team", "environment", "roles", f"{role_c.__class__.__name__}_{role_c.name}") + stg_path = serdeser_path.joinpath("team", "environment", "roles", f"{role_c.__class__.__name__}_{role_c.name}") + role_path = stg_path.joinpath("role.json") try: await role_c.run(with_message=Message(content="demo", cause_by=UserRequirement)) except Exception: - logger.error(f"Exception in `role_a.run`, detail: {format_trackback_info()}") - role_c.serialize(stg_path) + logger.error(f"Exception in `role_c.run`, detail: {format_trackback_info()}") + write_json_file(role_path, role_c.model_dump()) assert role_c.rc.memory.count() == 1 - new_role_a: Role = Role.deserialize(stg_path) - assert new_role_a.rc.state == 1 + role_dict = read_json_file(role_path) + new_role_c: Role = RoleC(**role_dict) + assert new_role_c.rc.state == 1 with pytest.raises(Exception): - await new_role_a.run(with_message=Message(content="demo", cause_by=UserRequirement)) + await new_role_c.run(with_message=Message(content="demo", cause_by=UserRequirement)) if __name__ == "__main__": diff --git a/tests/metagpt/serialize_deserialize/test_schema.py b/tests/metagpt/serialize_deserialize/test_schema.py index e793079f0..c5a457a1e 100644 --- a/tests/metagpt/serialize_deserialize/test_schema.py +++ b/tests/metagpt/serialize_deserialize/test_schema.py @@ -1,10 +1,11 @@ #!/usr/bin/env python # -*- coding: utf-8 -*- # @Desc : unittest of schema ser&deser +import pytest from metagpt.actions.action_node import ActionNode from metagpt.actions.write_code import WriteCode -from metagpt.schema import Document, Documents, Message +from metagpt.schema import CodingContext, Document, Documents, Message, TestingContext from metagpt.utils.common import any_to_str from tests.metagpt.serialize_deserialize.test_serdeser_base import ( MockICMessage, @@ -12,12 +13,16 @@ from tests.metagpt.serialize_deserialize.test_serdeser_base import ( ) -def test_message_serdeser(): +def test_message_serdeser_from_create_model(): + with pytest.raises(KeyError): + _ = Message(content="code", instruct_content={"class": "test", "key": "value"}) + out_mapping = {"field3": (str, ...), "field4": (list[str], ...)} out_data = {"field3": "field3 value3", "field4": ["field4 value1", "field4 value2"]} ic_obj = ActionNode.create_model_class("code", out_mapping) + ic_inst = ic_obj(**out_data) - message = Message(content="code", instruct_content=ic_obj(**out_data), role="engineer", cause_by=WriteCode) + message = Message(content="code", instruct_content=ic_inst, role="engineer", cause_by=WriteCode) ser_data = message.model_dump() assert ser_data["cause_by"] == "metagpt.actions.write_code.WriteCode" assert ser_data["instruct_content"]["class"] == "code" @@ -25,28 +30,72 @@ def test_message_serdeser(): new_message = Message(**ser_data) assert new_message.cause_by == any_to_str(WriteCode) assert new_message.cause_by in [any_to_str(WriteCode)] - assert new_message.instruct_content != ic_obj(**out_data) # TODO find why `!=` - assert new_message.instruct_content.model_dump() == ic_obj(**out_data).model_dump() - message = Message(content="test_ic", instruct_content=MockICMessage()) + assert new_message.instruct_content == ic_obj(**out_data) + assert new_message.instruct_content == ic_inst + assert new_message.instruct_content.model_dump() == ic_obj(**out_data).model_dump() + assert new_message == message + + mock_msg = MockMessage() + message = Message(content="test_ic", instruct_content=mock_msg) ser_data = message.model_dump() new_message = Message(**ser_data) - assert new_message.instruct_content != MockICMessage() # TODO - - message = Message(content="test_documents", instruct_content=Documents(docs={"doc1": Document(content="test doc")})) - ser_data = message.model_dump() - assert "class" in ser_data["instruct_content"] + assert new_message.instruct_content == mock_msg + assert new_message == message def test_message_without_postprocess(): - """to explain `instruct_content` should be postprocessed""" + """to explain `instruct_content` from `create_model_class` should be postprocessed""" out_mapping = {"field1": (list[str], ...)} out_data = {"field1": ["field1 value1", "field1 value2"]} ic_obj = ActionNode.create_model_class("code", out_mapping) - message = MockMessage(content="code", instruct_content=ic_obj(**out_data)) + message = MockICMessage(content="code", instruct_content=ic_obj(**out_data)) ser_data = message.model_dump() assert ser_data["instruct_content"] == {} ser_data["instruct_content"] = None - new_message = MockMessage(**ser_data) + new_message = MockICMessage(**ser_data) assert new_message.instruct_content != ic_obj(**out_data) + assert new_message != message + + +def test_message_serdeser_from_basecontext(): + doc_msg = Message(content="test_document", instruct_content=Document(content="test doc")) + ser_data = doc_msg.model_dump() + assert ser_data["instruct_content"]["value"]["content"] == "test doc" + assert ser_data["instruct_content"]["value"]["filename"] == "" + + docs_msg = Message( + content="test_documents", instruct_content=Documents(docs={"doc1": Document(content="test doc")}) + ) + ser_data = docs_msg.model_dump() + assert ser_data["instruct_content"]["class"] == "Documents" + assert ser_data["instruct_content"]["value"]["docs"]["doc1"]["content"] == "test doc" + assert ser_data["instruct_content"]["value"]["docs"]["doc1"]["filename"] == "" + + code_ctxt = CodingContext( + filename="game.py", + design_doc=Document(root_path="docs/system_design", filename="xx.json", content="xxx"), + task_doc=Document(root_path="docs/tasks", filename="xx.json", content="xxx"), + code_doc=Document(root_path="xxx", filename="game.py", content="xxx"), + ) + code_ctxt_msg = Message(content="coding_context", instruct_content=code_ctxt) + ser_data = code_ctxt_msg.model_dump() + assert ser_data["instruct_content"]["class"] == "CodingContext" + + new_code_ctxt_msg = Message(**ser_data) + assert new_code_ctxt_msg.instruct_content == code_ctxt + assert new_code_ctxt_msg.instruct_content.code_doc.filename == "game.py" + assert new_code_ctxt_msg == code_ctxt_msg + + testing_ctxt = TestingContext( + filename="test.py", + code_doc=Document(root_path="xxx", filename="game.py", content="xxx"), + test_doc=Document(root_path="docs/tests", filename="test.py", content="xxx"), + ) + testing_ctxt_msg = Message(content="testing_context", instruct_content=testing_ctxt) + ser_data = testing_ctxt_msg.model_dump() + new_testing_ctxt_msg = Message(**ser_data) + assert new_testing_ctxt_msg.instruct_content == testing_ctxt + assert new_testing_ctxt_msg.instruct_content.test_doc.filename == "test.py" + assert new_testing_ctxt_msg == testing_ctxt_msg diff --git a/tests/metagpt/serialize_deserialize/test_serdeser_base.py b/tests/metagpt/serialize_deserialize/test_serdeser_base.py index daa46c99c..62ab26d72 100644 --- a/tests/metagpt/serialize_deserialize/test_serdeser_base.py +++ b/tests/metagpt/serialize_deserialize/test_serdeser_base.py @@ -16,14 +16,14 @@ from metagpt.roles.role import Role, RoleReactMode serdeser_path = Path(__file__).absolute().parent.joinpath("..", "..", "data", "serdeser_storage") -class MockICMessage(BaseModel): - content: str = "test_ic" - - class MockMessage(BaseModel): + content: str = "test_msg" + + +class MockICMessage(BaseModel): """to test normal dict without postprocess""" - content: str = "" + content: str = "test_ic_msg" instruct_content: Optional[BaseModel] = Field(default=None) @@ -67,7 +67,7 @@ class RoleA(Role): def __init__(self, **kwargs): super(RoleA, self).__init__(**kwargs) - self._init_actions([ActionPass]) + self.set_actions([ActionPass]) self._watch([UserRequirement]) @@ -79,7 +79,7 @@ class RoleB(Role): def __init__(self, **kwargs): super(RoleB, self).__init__(**kwargs) - self._init_actions([ActionOK, ActionRaise]) + self.set_actions([ActionOK, ActionRaise]) self._watch([ActionPass]) self.rc.react_mode = RoleReactMode.BY_ORDER @@ -92,7 +92,7 @@ class RoleC(Role): def __init__(self, **kwargs): super(RoleC, self).__init__(**kwargs) - self._init_actions([ActionOK, ActionRaise]) + self.set_actions([ActionOK, ActionRaise]) self._watch([UserRequirement]) self.rc.react_mode = RoleReactMode.BY_ORDER self.rc.memory.ignore_id = True diff --git a/tests/metagpt/serialize_deserialize/test_sk_agent.py b/tests/metagpt/serialize_deserialize/test_sk_agent.py index 7f287b8f9..97c0ade99 100644 --- a/tests/metagpt/serialize_deserialize/test_sk_agent.py +++ b/tests/metagpt/serialize_deserialize/test_sk_agent.py @@ -5,15 +5,8 @@ import pytest from metagpt.roles.sk_agent import SkAgent -def test_sk_agent_serialize(): - role = SkAgent() - ser_role_dict = role.model_dump(exclude={"import_semantic_skill_from_directory", "import_skill"}) - assert "name" in ser_role_dict - assert "planner" in ser_role_dict - - @pytest.mark.asyncio -async def test_sk_agent_deserialize(): +async def test_sk_agent_serdeser(): role = SkAgent() ser_role_dict = role.model_dump(exclude={"import_semantic_skill_from_directory", "import_skill"}) assert "name" in ser_role_dict diff --git a/tests/metagpt/serialize_deserialize/test_team.py b/tests/metagpt/serialize_deserialize/test_team.py index 566f63c3d..dbd38422d 100644 --- a/tests/metagpt/serialize_deserialize/test_team.py +++ b/tests/metagpt/serialize_deserialize/test_team.py @@ -4,13 +4,14 @@ # @Desc : import shutil +from pathlib import Path import pytest -from metagpt.const import SERDESER_PATH from metagpt.logs import logger from metagpt.roles import Architect, ProductManager, ProjectManager from metagpt.team import Team +from metagpt.utils.common import write_json_file from tests.metagpt.serialize_deserialize.test_serdeser_base import ( ActionOK, RoleA, @@ -20,8 +21,8 @@ from tests.metagpt.serialize_deserialize.test_serdeser_base import ( ) -def test_team_deserialize(): - company = Team() +def test_team_deserialize(context): + company = Team(context=context) pm = ProductManager() arch = Architect() @@ -45,9 +46,16 @@ def test_team_deserialize(): assert new_company.env.get_role(arch.profile) is not None -def test_team_serdeser_save(): - company = Team() +def mock_team_serialize(self, stg_path: Path = serdeser_path.joinpath("team")): + team_info_path = stg_path.joinpath("team.json") + write_json_file(team_info_path, self.model_dump()) + + +def test_team_serdeser_save(mocker, context): + mocker.patch("metagpt.team.Team.serialize", mock_team_serialize) + + company = Team(context=context) company.hire([RoleC()]) stg_path = serdeser_path.joinpath("team") @@ -61,12 +69,14 @@ def test_team_serdeser_save(): @pytest.mark.asyncio -async def test_team_recover(): +async def test_team_recover(mocker, context): + mocker.patch("metagpt.team.Team.serialize", mock_team_serialize) + idea = "write a snake game" - stg_path = SERDESER_PATH.joinpath("team") + stg_path = serdeser_path.joinpath("team") shutil.rmtree(stg_path, ignore_errors=True) - company = Team() + company = Team(context=context) role_c = RoleC() company.hire([role_c]) company.run_project(idea) @@ -75,9 +85,9 @@ async def test_team_recover(): ser_data = company.model_dump() new_company = Team(**ser_data) - new_company.env.get_role(role_c.profile) - # assert new_role_c.rc.memory == role_c.rc.memory # TODO - # assert new_role_c.rc.env != role_c.rc.env # TODO + new_role_c = new_company.env.get_role(role_c.profile) + assert new_role_c.rc.memory == role_c.rc.memory + assert new_role_c.rc.env != role_c.rc.env assert type(list(new_company.env.roles.values())[0].actions[0]) == ActionOK new_company.run_project(idea) @@ -85,12 +95,14 @@ async def test_team_recover(): @pytest.mark.asyncio -async def test_team_recover_save(): +async def test_team_recover_save(mocker, context): + mocker.patch("metagpt.team.Team.serialize", mock_team_serialize) + idea = "write a 2048 web game" - stg_path = SERDESER_PATH.joinpath("team") + stg_path = serdeser_path.joinpath("team") shutil.rmtree(stg_path, ignore_errors=True) - company = Team() + company = Team(context=context) role_c = RoleC() company.hire([role_c]) company.run_project(idea) @@ -98,8 +110,8 @@ async def test_team_recover_save(): new_company = Team.deserialize(stg_path) new_role_c = new_company.env.get_role(role_c.profile) - # assert new_role_c.rc.memory == role_c.rc.memory - # assert new_role_c.rc.env != role_c.rc.env + assert new_role_c.rc.memory == role_c.rc.memory + assert new_role_c.rc.env != role_c.rc.env assert new_role_c.recovered != role_c.recovered # here cause previous ut is `!=` assert new_role_c.rc.todo != role_c.rc.todo # serialize exclude `rc.todo` assert new_role_c.rc.news != role_c.rc.news # serialize exclude `rc.news` @@ -109,15 +121,17 @@ async def test_team_recover_save(): @pytest.mark.asyncio -async def test_team_recover_multi_roles_save(): +async def test_team_recover_multi_roles_save(mocker, context): + mocker.patch("metagpt.team.Team.serialize", mock_team_serialize) + idea = "write a snake game" - stg_path = SERDESER_PATH.joinpath("team") + stg_path = serdeser_path.joinpath("team") shutil.rmtree(stg_path, ignore_errors=True) role_a = RoleA() role_b = RoleB() - company = Team() + company = Team(context=context) company.hire([role_a, role_b]) company.run_project(idea) await company.run(n_round=4) @@ -130,3 +144,7 @@ async def test_team_recover_multi_roles_save(): assert new_company.env.get_role(role_b.profile).rc.state == 1 await new_company.run(n_round=4) + + +if __name__ == "__main__": + pytest.main([__file__, "-s"]) diff --git a/tests/metagpt/serialize_deserialize/test_tutorial_assistant.py b/tests/metagpt/serialize_deserialize/test_tutorial_assistant.py index e642dae54..ab5db4c57 100644 --- a/tests/metagpt/serialize_deserialize/test_tutorial_assistant.py +++ b/tests/metagpt/serialize_deserialize/test_tutorial_assistant.py @@ -7,7 +7,7 @@ from metagpt.roles.tutorial_assistant import TutorialAssistant @pytest.mark.asyncio -async def test_tutorial_assistant_deserialize(): +async def test_tutorial_assistant_serdeser(context): role = TutorialAssistant() ser_role_dict = role.model_dump() assert "name" in ser_role_dict diff --git a/tests/metagpt/serialize_deserialize/test_write_code.py b/tests/metagpt/serialize_deserialize/test_write_code.py index cb262bb45..2f3c08f9b 100644 --- a/tests/metagpt/serialize_deserialize/test_write_code.py +++ b/tests/metagpt/serialize_deserialize/test_write_code.py @@ -9,22 +9,23 @@ from metagpt.actions import WriteCode from metagpt.schema import CodingContext, Document -def test_write_design_serialize(): - action = WriteCode() +def test_write_design_serdeser(context): + action = WriteCode(context=context) ser_action_dict = action.model_dump() assert ser_action_dict["name"] == "WriteCode" assert "llm" not in ser_action_dict # not export @pytest.mark.asyncio -async def test_write_code_deserialize(): - context = CodingContext( +async def test_write_code_serdeser(context): + context.src_workspace = context.repo.workdir / "srcs" + coding_context = CodingContext( filename="test_code.py", design_doc=Document(content="write add function to calculate two numbers") ) - doc = Document(content=context.model_dump_json()) - action = WriteCode(context=doc) + doc = Document(content=coding_context.model_dump_json()) + action = WriteCode(i_context=doc, context=context) serialized_data = action.model_dump() - new_action = WriteCode(**serialized_data) + new_action = WriteCode(**serialized_data, context=context) assert new_action.name == "WriteCode" await action.run() diff --git a/tests/metagpt/serialize_deserialize/test_write_code_review.py b/tests/metagpt/serialize_deserialize/test_write_code_review.py index 991b3c13b..32a017a97 100644 --- a/tests/metagpt/serialize_deserialize/test_write_code_review.py +++ b/tests/metagpt/serialize_deserialize/test_write_code_review.py @@ -9,22 +9,23 @@ from metagpt.schema import CodingContext, Document @pytest.mark.asyncio -async def test_write_code_review_deserialize(): +async def test_write_code_review_serdeser(context): + context.src_workspace = context.repo.workdir / "srcs" code_content = """ def div(a: int, b: int = 0): return a / b """ - context = CodingContext( + coding_context = CodingContext( filename="test_op.py", design_doc=Document(content="divide two numbers"), code_doc=Document(content=code_content), ) - action = WriteCodeReview(context=context) + action = WriteCodeReview(i_context=coding_context) serialized_data = action.model_dump() assert serialized_data["name"] == "WriteCodeReview" - new_action = WriteCodeReview(**serialized_data) + new_action = WriteCodeReview(**serialized_data, context=context) assert new_action.name == "WriteCodeReview" await new_action.run() diff --git a/tests/metagpt/serialize_deserialize/test_write_design.py b/tests/metagpt/serialize_deserialize/test_write_design.py index 7bcba3fc8..6519d8cdc 100644 --- a/tests/metagpt/serialize_deserialize/test_write_design.py +++ b/tests/metagpt/serialize_deserialize/test_write_design.py @@ -7,33 +7,25 @@ import pytest from metagpt.actions import WriteDesign, WriteTasks -def test_write_design_serialize(): - action = WriteDesign() - ser_action_dict = action.model_dump() - assert "name" in ser_action_dict - assert "llm" not in ser_action_dict # not export - - -def test_write_task_serialize(): - action = WriteTasks() - ser_action_dict = action.model_dump() - assert "name" in ser_action_dict - assert "llm" not in ser_action_dict # not export - - @pytest.mark.asyncio -async def test_write_design_deserialize(): - action = WriteDesign() - serialized_data = action.model_dump() - new_action = WriteDesign(**serialized_data) +async def test_write_design_serialize(context): + action = WriteDesign(context=context) + ser_action_dict = action.model_dump() + assert "name" in ser_action_dict + assert "llm" not in ser_action_dict # not export + + new_action = WriteDesign(**ser_action_dict, context=context) assert new_action.name == "WriteDesign" await new_action.run(with_messages="write a cli snake game") @pytest.mark.asyncio -async def test_write_task_deserialize(): - action = WriteTasks() - serialized_data = action.model_dump() - new_action = WriteTasks(**serialized_data) +async def test_write_task_serialize(context): + action = WriteTasks(context=context) + ser_action_dict = action.model_dump() + assert "name" in ser_action_dict + assert "llm" not in ser_action_dict # not export + + new_action = WriteTasks(**ser_action_dict, context=context) assert new_action.name == "WriteTasks" await new_action.run(with_messages="write a cli snake game") diff --git a/tests/metagpt/serialize_deserialize/test_write_docstring.py b/tests/metagpt/serialize_deserialize/test_write_docstring.py index e4116ab30..363bed05e 100644 --- a/tests/metagpt/serialize_deserialize/test_write_docstring.py +++ b/tests/metagpt/serialize_deserialize/test_write_docstring.py @@ -29,14 +29,14 @@ class Person: ], ids=["google", "numpy", "sphinx"], ) -async def test_action_deserialize(style: str, part: str): - action = WriteDocstring() +async def test_action_serdeser(style: str, part: str, context): + action = WriteDocstring(context=context) serialized_data = action.model_dump() assert "name" in serialized_data assert serialized_data["desc"] == "Write docstring for code." - new_action = WriteDocstring(**serialized_data) + new_action = WriteDocstring(**serialized_data, context=context) assert new_action.name == "WriteDocstring" assert new_action.desc == "Write docstring for code." diff --git a/tests/metagpt/serialize_deserialize/test_write_prd.py b/tests/metagpt/serialize_deserialize/test_write_prd.py index b9eff5a19..e4951efb7 100644 --- a/tests/metagpt/serialize_deserialize/test_write_prd.py +++ b/tests/metagpt/serialize_deserialize/test_write_prd.py @@ -9,18 +9,14 @@ from metagpt.actions import WritePRD from metagpt.schema import Message -def test_action_serialize(new_filename): - action = WritePRD() +@pytest.mark.asyncio +async def test_action_serdeser(new_filename, context): + action = WritePRD(context=context) ser_action_dict = action.model_dump() assert "name" in ser_action_dict assert "llm" not in ser_action_dict # not export - -@pytest.mark.asyncio -async def test_action_deserialize(new_filename): - action = WritePRD() - serialized_data = action.model_dump() - new_action = WritePRD(**serialized_data) + new_action = WritePRD(**ser_action_dict, context=context) assert new_action.name == "WritePRD" - action_output = await new_action.run(with_messages=Message(content="write a cli snake game")) - assert len(action_output.content) > 0 + with pytest.raises(FileNotFoundError): + await new_action.run(with_messages=Message(content="write a cli snake game")) diff --git a/tests/metagpt/serialize_deserialize/test_write_review.py b/tests/metagpt/serialize_deserialize/test_write_review.py index f02a01910..de2fd9d7a 100644 --- a/tests/metagpt/serialize_deserialize/test_write_review.py +++ b/tests/metagpt/serialize_deserialize/test_write_review.py @@ -5,7 +5,7 @@ import pytest from metagpt.actions.action_node import ActionNode from metagpt.actions.write_review import WriteReview -CONTEXT = """ +TEMPLATE_CONTEXT = """ { "Language": "zh_cn", "Programming Language": "Python", @@ -42,13 +42,13 @@ CONTEXT = """ @pytest.mark.asyncio -async def test_action_deserialize(): - action = WriteReview() +async def test_action_serdeser(context): + action = WriteReview(context=context) serialized_data = action.model_dump() assert serialized_data["name"] == "WriteReview" - new_action = WriteReview(**serialized_data) - review = await new_action.run(CONTEXT) + new_action = WriteReview(**serialized_data, context=context) + review = await new_action.run(TEMPLATE_CONTEXT) assert new_action.name == "WriteReview" assert type(review) == ActionNode diff --git a/tests/metagpt/serialize_deserialize/test_write_tutorial.py b/tests/metagpt/serialize_deserialize/test_write_tutorial.py index 606a90f8c..d41b7b341 100644 --- a/tests/metagpt/serialize_deserialize/test_write_tutorial.py +++ b/tests/metagpt/serialize_deserialize/test_write_tutorial.py @@ -9,13 +9,13 @@ from metagpt.actions.write_tutorial import WriteContent, WriteDirectory @pytest.mark.asyncio @pytest.mark.parametrize(("language", "topic"), [("English", "Write a tutorial about Python")]) -async def test_write_directory_deserialize(language: str, topic: str): - action = WriteDirectory() +async def test_write_directory_serdeser(language: str, topic: str, context): + action = WriteDirectory(context=context) serialized_data = action.model_dump() assert serialized_data["name"] == "WriteDirectory" assert serialized_data["language"] == "Chinese" - new_action = WriteDirectory(**serialized_data) + new_action = WriteDirectory(**serialized_data, context=context) ret = await new_action.run(topic=topic) assert isinstance(ret, dict) assert "title" in ret @@ -30,12 +30,12 @@ async def test_write_directory_deserialize(language: str, topic: str): ("language", "topic", "directory"), [("English", "Write a tutorial about Python", {"Introduction": ["What is Python?", "Why learn Python?"]})], ) -async def test_write_content_deserialize(language: str, topic: str, directory: Dict): - action = WriteContent(language=language, directory=directory) +async def test_write_content_serdeser(language: str, topic: str, directory: Dict, context): + action = WriteContent(language=language, directory=directory, context=context) serialized_data = action.model_dump() assert serialized_data["name"] == "WriteContent" - new_action = WriteContent(**serialized_data) + new_action = WriteContent(**serialized_data, context=context) ret = await new_action.run(topic=topic) assert isinstance(ret, str) assert list(directory.keys())[0] in ret diff --git a/tests/metagpt/strategy/examples/creative_writing.py b/tests/metagpt/strategy/examples/test_creative_writing.py similarity index 100% rename from tests/metagpt/strategy/examples/creative_writing.py rename to tests/metagpt/strategy/examples/test_creative_writing.py diff --git a/tests/metagpt/strategy/examples/game24.py b/tests/metagpt/strategy/examples/test_game24.py similarity index 100% rename from tests/metagpt/strategy/examples/game24.py rename to tests/metagpt/strategy/examples/test_game24.py diff --git a/tests/metagpt/test_config.py b/tests/metagpt/test_config.py new file mode 100644 index 000000000..7ce5765cf --- /dev/null +++ b/tests/metagpt/test_config.py @@ -0,0 +1,24 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/1/9 15:57 +@Author : alexanderwu +@File : test_config.py +""" + +from metagpt.config2 import Config +from metagpt.configs.llm_config import LLMType +from tests.metagpt.provider.mock_llm_config import mock_llm_config + + +def test_config_1(): + cfg = Config.default() + llm = cfg.get_openai_llm() + assert llm is not None + assert llm.api_type == LLMType.OPENAI + + +def test_config_from_dict(): + cfg = Config(llm=mock_llm_config) + assert cfg + assert cfg.llm.api_key == "mock_api_key" diff --git a/tests/metagpt/test_context.py b/tests/metagpt/test_context.py new file mode 100644 index 000000000..f8218c44d --- /dev/null +++ b/tests/metagpt/test_context.py @@ -0,0 +1,73 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/1/9 13:52 +@Author : alexanderwu +@File : test_context.py +""" +from metagpt.configs.llm_config import LLMType +from metagpt.context import AttrDict, Context + + +def test_attr_dict_1(): + ad = AttrDict(name="John", age=30) + assert ad.name == "John" + assert ad.age == 30 + assert ad.height is None + + +def test_attr_dict_2(): + ad = AttrDict(name="John", age=30) + ad.height = 180 + assert ad.height == 180 + + +def test_attr_dict_3(): + ad = AttrDict(name="John", age=30) + del ad.age + assert ad.age is None + + +def test_attr_dict_4(): + ad = AttrDict(name="John", age=30) + try: + del ad.weight + except AttributeError as e: + assert str(e) == "No such attribute: weight" + + +def test_attr_dict_5(): + ad = AttrDict.model_validate({"name": "John", "age": 30}) + assert ad.name == "John" + assert ad.age == 30 + + +def test_context_1(): + ctx = Context() + assert ctx.config is not None + assert ctx.git_repo is None + assert ctx.src_workspace is None + assert ctx.cost_manager is not None + + +def test_context_2(): + ctx = Context() + llm = ctx.config.get_openai_llm() + assert llm is not None + assert llm.api_type == LLMType.OPENAI + + kwargs = ctx.kwargs + assert kwargs is not None + + kwargs.test_key = "test_value" + assert kwargs.test_key == "test_value" + + +def test_context_3(): + # ctx = Context() + # ctx.use_llm(provider=LLMType.OPENAI) + # assert ctx._llm_config is not None + # assert ctx._llm_config.api_type == LLMType.OPENAI + # assert ctx.llm() is not None + # assert "gpt" in ctx.llm().model + pass diff --git a/tests/metagpt/test_context_mixin.py b/tests/metagpt/test_context_mixin.py new file mode 100644 index 000000000..4389dc251 --- /dev/null +++ b/tests/metagpt/test_context_mixin.py @@ -0,0 +1,134 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/1/11 19:24 +@Author : alexanderwu +@File : test_context_mixin.py +""" +from pathlib import Path + +import pytest +from pydantic import BaseModel + +from metagpt.actions import Action +from metagpt.config2 import Config +from metagpt.const import CONFIG_ROOT +from metagpt.context_mixin import ContextMixin +from metagpt.environment import Environment +from metagpt.roles import Role +from metagpt.team import Team +from tests.metagpt.provider.mock_llm_config import ( + mock_llm_config, + mock_llm_config_proxy, + mock_llm_config_zhipu, +) + + +class ModelX(ContextMixin, BaseModel): + a: str = "a" + b: str = "b" + + +class WTFMixin(BaseModel): + c: str = "c" + d: str = "d" + + +class ModelY(WTFMixin, ModelX): + pass + + +def test_config_mixin_1(): + new_model = ModelX() + assert new_model.a == "a" + assert new_model.b == "b" + + +def test_config_mixin_2(): + i = Config(llm=mock_llm_config) + j = Config(llm=mock_llm_config_proxy) + obj = ModelX(config=i) + assert obj.config == i + assert obj.config.llm == mock_llm_config + + obj.set_config(j) + # obj already has a config, so it will not be set + assert obj.config == i + + +def test_config_mixin_3_multi_inheritance_not_override_config(): + """Test config mixin with multiple inheritance""" + i = Config(llm=mock_llm_config) + j = Config(llm=mock_llm_config_proxy) + obj = ModelY(config=i) + assert obj.config == i + assert obj.config.llm == mock_llm_config + + obj.set_config(j) + # obj already has a config, so it will not be set + assert obj.config == i + assert obj.config.llm == mock_llm_config + + assert obj.a == "a" + assert obj.b == "b" + assert obj.c == "c" + assert obj.d == "d" + + print(obj.__dict__.keys()) + assert "private_config" in obj.__dict__.keys() + + +def test_config_mixin_4_multi_inheritance_override_config(): + """Test config mixin with multiple inheritance""" + i = Config(llm=mock_llm_config) + j = Config(llm=mock_llm_config_zhipu) + obj = ModelY(config=i) + assert obj.config == i + assert obj.config.llm == mock_llm_config + + obj.set_config(j, override=True) + # override obj.config + assert obj.config == j + assert obj.config.llm == mock_llm_config_zhipu + + assert obj.a == "a" + assert obj.b == "b" + assert obj.c == "c" + assert obj.d == "d" + + print(obj.__dict__.keys()) + assert "private_config" in obj.__dict__.keys() + assert obj.config.llm.model == "mock_zhipu_model" + + +@pytest.mark.asyncio +async def test_config_priority(): + """If action's config is set, then its llm will be set, otherwise, it will use the role's llm""" + home_dir = Path.home() / CONFIG_ROOT + gpt4t = Config.from_home("gpt-4-1106-preview.yaml") + if not home_dir.exists(): + assert gpt4t is None + gpt35 = Config.default() + gpt35.llm.model = "gpt-3.5-turbo-1106" + gpt4 = Config.default() + gpt4.llm.model = "gpt-4-0613" + + a1 = Action(config=gpt4t, name="Say", instruction="Say your opinion with emotion and don't repeat it") + a2 = Action(name="Say", instruction="Say your opinion with emotion and don't repeat it") + a3 = Action(name="Vote", instruction="Vote for the candidate, and say why you vote for him/her") + + # it will not work for a1 because the config is already set + A = Role(name="A", profile="Democratic candidate", goal="Win the election", actions=[a1], watch=[a2], config=gpt4) + # it will work for a2 because the config is not set + B = Role(name="B", profile="Republican candidate", goal="Win the election", actions=[a2], watch=[a1], config=gpt4) + # ditto + C = Role(name="C", profile="Voter", goal="Vote for the candidate", actions=[a3], watch=[a1, a2], config=gpt35) + + env = Environment(desc="US election live broadcast") + Team(investment=10.0, env=env, roles=[A, B, C]) + + assert a1.llm.model == "gpt-4-1106-preview" if Path(home_dir / "gpt-4-1106-preview.yaml").exists() else "gpt-4-0613" + assert a2.llm.model == "gpt-4-0613" + assert a3.llm.model == "gpt-3.5-turbo-1106" + + # history = await team.run(idea="Topic: climate change. Under 80 words per message.", send_to="a1", n_round=3) diff --git a/tests/metagpt/test_document.py b/tests/metagpt/test_document.py index 18650e112..9c076f4e6 100644 --- a/tests/metagpt/test_document.py +++ b/tests/metagpt/test_document.py @@ -5,7 +5,7 @@ @Author : alexanderwu @File : test_document.py """ -from metagpt.config import CONFIG +from metagpt.config2 import config from metagpt.document import Repo from metagpt.logs import logger @@ -28,6 +28,6 @@ def load_existing_repo(path): def test_repo_set_load(): - repo_path = CONFIG.workspace_path / "test_repo" + repo_path = config.workspace.path / "test_repo" set_existing_repo(repo_path) load_existing_repo(repo_path) diff --git a/tests/metagpt/test_environment.py b/tests/metagpt/test_environment.py index 90e4b5b42..7559655d3 100644 --- a/tests/metagpt/test_environment.py +++ b/tests/metagpt/test_environment.py @@ -4,8 +4,6 @@ @Time : 2023/5/12 00:47 @Author : alexanderwu @File : test_environment.py -@Modified By: mashenquan, 2023/8/20. Remove global configuration `CONFIG`, enable configuration support for business isolation. - """ from pathlib import Path @@ -13,7 +11,6 @@ from pathlib import Path import pytest from metagpt.actions import UserRequirement -from metagpt.config import CONFIG from metagpt.environment import Environment from metagpt.logs import logger from metagpt.roles import Architect, ProductManager, Role @@ -45,10 +42,10 @@ def test_get_roles(env: Environment): @pytest.mark.asyncio -async def test_publish_and_process_message(env: Environment, new_filename): - if CONFIG.git_repo: - CONFIG.git_repo.delete_repository() - CONFIG.git_repo = None +async def test_publish_and_process_message(env: Environment): + if env.context.git_repo: + env.context.git_repo.delete_repository() + env.context.git_repo = None product_manager = ProductManager(name="Alice", profile="Product Manager", goal="做AI Native产品", constraints="资源有限") architect = Architect( diff --git a/tests/metagpt/test_gpt.py b/tests/metagpt/test_gpt.py deleted file mode 100644 index 2b19f173d..000000000 --- a/tests/metagpt/test_gpt.py +++ /dev/null @@ -1,45 +0,0 @@ -#!/usr/bin/env python -# -*- coding: utf-8 -*- -""" -@Time : 2023/4/29 19:47 -@Author : alexanderwu -@File : test_gpt.py -""" -import openai -import pytest - -from metagpt.config import CONFIG -from metagpt.logs import logger - - -@pytest.mark.usefixtures("llm_api") -class TestGPT: - @pytest.mark.asyncio - async def test_llm_api_aask(self, llm_api): - answer = await llm_api.aask("hello chatgpt", stream=False) - logger.info(answer) - assert len(answer) > 0 - - answer = await llm_api.aask("hello chatgpt", stream=True) - logger.info(answer) - assert len(answer) > 0 - - @pytest.mark.asyncio - async def test_llm_api_aask_code(self, llm_api): - try: - answer = await llm_api.aask_code(["请扮演一个Google Python专家工程师，如果理解，回复明白", "写一个hello world"], timeout=60) - logger.info(answer) - assert len(answer) > 0 - except openai.BadRequestError: - assert CONFIG.OPENAI_API_TYPE == "azure" - - @pytest.mark.asyncio - async def test_llm_api_costs(self, llm_api): - await llm_api.aask("hello chatgpt", stream=False) - costs = llm_api.get_costs() - logger.info(costs) - assert costs.total_cost > 0 - - -if __name__ == "__main__": - pytest.main([__file__, "-s"]) diff --git a/tests/metagpt/test_incremental_dev.py b/tests/metagpt/test_incremental_dev.py new file mode 100644 index 000000000..3e4a1b901 --- /dev/null +++ b/tests/metagpt/test_incremental_dev.py @@ -0,0 +1,189 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/01/03 +@Author : mannaandpoem +@File : test_incremental_dev.py +""" +import os +import subprocess +import time + +import pytest +from typer.testing import CliRunner + +from metagpt.const import TEST_DATA_PATH +from metagpt.logs import logger +from metagpt.startup import app + +runner = CliRunner() + +IDEAS = [ + "Add subtraction, multiplication and division operations to the calculator. The current calculator can only perform basic addition operations, and it is necessary to introduce subtraction, multiplication, division operation into the calculator", + "Adding graphical interface functionality to enhance the user experience in the number-guessing game. The existing number-guessing game currently relies on command-line input for numbers. The goal is to introduce a graphical interface to improve the game's usability and visual appeal", + "Add a feature to remove deprecated words from the word cloud. The current word cloud generator does not support removing deprecated words. Now, The word cloud generator should support removing deprecated words. Customize deactivated words to exclude them from word cloud. Let users see all the words in the text file, and allow users to select the words they want to remove.", + "Add an AI opponent with fixed difficulty levels. Currently, the game only allows players to compete against themselves. Implement an AI algorithm that can playing with player. This will provide a more engaging and challenging experience for players.", + "Add functionality to view the history of scores. The original dice rolling game could only display the current game result, but the new requirement allows players to view the history of scores", + "Add functionality to view the history of scores and perform statistical analysis on them. The original dice rolling game could only display the current game result, but the new requirement allows players to view the history of scores and display the statistical analysis results of the current score", + "Changed score target for 2048 game from 2048 to 4096. Please change the game's score target from 2048 to 4096, and change the interface size from 4*4 to 8*8", + "Display the history score of the player in the 2048 game. Add a record board that can display players' historical score records so that players can trace their scores", + "Incremental Idea Gradually increase the speed of the snake as the game progresses. In the current version of the game, the snake’s speed remains constant throughout the gameplay. Implement a feature where the snake’s speed gradually increases over time, making the game more challenging and intense as the player progresses.", + "Introduce power-ups and obstacles to the game. The current version of the game only involves eating food and growing the snake. Add new elements such as power-ups that can enhance the snake’s speed or make it invincible for a short duration. At the same time, introduce obstacles like walls or enemies that the snake must avoid or overcome to continue growing.", +] + +PROJECT_NAMES = [ + "simple_add_calculator", + "number_guessing_game", + "word_cloud", + "Gomoku", + "dice_simulator_new", + "dice_simulator_new", + "pygame_2048", + "pygame_2048", + "snake_game", + "snake_game", +] + + +def test_simple_add_calculator(): + result = get_incremental_dev_result(IDEAS[0], PROJECT_NAMES[0]) + log_and_check_result(result) + + +@pytest.mark.skip +def test_number_guessing_game(): + result = get_incremental_dev_result(IDEAS[1], PROJECT_NAMES[1]) + log_and_check_result(result) + + +@pytest.mark.skip +def test_word_cloud(): + result = get_incremental_dev_result(IDEAS[2], PROJECT_NAMES[2]) + log_and_check_result(result) + + +@pytest.mark.skip +def test_gomoku(): + result = get_incremental_dev_result(IDEAS[3], PROJECT_NAMES[3]) + log_and_check_result(result) + + +@pytest.mark.skip +def test_dice_simulator_new(): + for i, (idea, project_name) in enumerate(zip(IDEAS[4:6], PROJECT_NAMES[4:6]), start=1): + result = get_incremental_dev_result(idea, project_name) + log_and_check_result(result, "refine_" + str(i)) + + +@pytest.mark.skip +def test_refined_pygame_2048(): + for i, (idea, project_name) in enumerate(zip(IDEAS[6:8], PROJECT_NAMES[6:8]), start=1): + result = get_incremental_dev_result(idea, project_name) + log_and_check_result(result, "refine_" + str(i)) + + +@pytest.mark.skip +def test_refined_snake_game(): + for i, (idea, project_name) in enumerate(zip(IDEAS[8:10], PROJECT_NAMES[8:10]), start=1): + result = get_incremental_dev_result(idea, project_name) + log_and_check_result(result, "refine_" + str(i)) + + +def log_and_check_result(result, tag_name="refine"): + logger.info(result) + logger.info(result.output) + if "Aborting" in result.output: + assert False + else: + # After running, there will be new commit + cur_tag = subprocess.run(["git", "describe", "--tags"], capture_output=True, text=True).stdout.strip() + if cur_tag == "base": + assert False + else: + assert True + if subprocess.run(["git", "show-ref", "--verify", "--quiet", f"refs/tags/{tag_name}"]).returncode == 0: + tag_name += str(int(time.time())) + try: + subprocess.run(["git", "tag", tag_name], check=True) + except subprocess.CalledProcessError as e: + raise e + + +def get_incremental_dev_result(idea, project_name, use_review=True): + project_path = TEST_DATA_PATH / "incremental_dev_project" / project_name + # Check if the project path exists + if not project_path.exists(): + # If the project does not exist, extract the project file + try: + # Use the tar command to extract the .zip file + subprocess.run(["tar", "-xf", f"{project_path}.zip", "-C", str(project_path.parent)], check=True) + except subprocess.CalledProcessError as e: + # If the extraction fails, throw an exception + raise Exception(f"Failed to extract project {project_name}. Error: {e}") + + check_or_create_base_tag(project_path) + args = [idea, "--inc", "--project-path", project_path, "--n-round", "20"] + if not use_review: + args.append("--no-code-review") + result = runner.invoke(app, args) + return result + + +def check_or_create_base_tag(project_path): + # Change the current working directory to the specified project path + os.chdir(project_path) + + # Initialize a Git repository + subprocess.run(["git", "init"], check=True) + + # Check if the 'base' tag exists + check_base_tag_cmd = ["git", "show-ref", "--verify", "--quiet", "refs/tags/base"] + if subprocess.run(check_base_tag_cmd).returncode == 0: + has_base_tag = True + else: + has_base_tag = False + + if has_base_tag: + logger.info("Base tag exists") + # Switch to the 'base' branch if it exists + try: + status = subprocess.run(["git", "status", "-s"], capture_output=True, text=True).stdout.strip() + if status: + subprocess.run(["git", "clean", "-df"]) + subprocess.run(["git", "checkout", "-f", "base"], check=True) + logger.info("Switched to base branch") + except Exception as e: + logger.error("Failed to switch to base branch") + raise e + + else: + logger.info("Base tag doesn't exist.") + # Add and commit the current code if 'base' tag doesn't exist + add_cmd = ["git", "add", "."] + try: + subprocess.run(add_cmd, check=True) + logger.info("Files added successfully.") + except subprocess.CalledProcessError as e: + logger.error(f"Failed to add files: {e}") + + commit_cmd = ["git", "commit", "-m", "Initial commit"] + try: + subprocess.run(commit_cmd, check=True) + logger.info("Committed all files with the message 'Initial commit'.") + except subprocess.CalledProcessError as e: + logger.error(f"Failed to commit: {e.stderr}") + + # Add 'base' tag + add_base_tag_cmd = ["git", "tag", "base"] + + # Check if the 'git tag' command was successful + try: + subprocess.run(add_base_tag_cmd, check=True) + logger.info("Added 'base' tag.") + except Exception as e: + logger.error("Failed to add 'base' tag.") + raise e + + +if __name__ == "__main__": + pytest.main([__file__, "-s"]) diff --git a/tests/metagpt/test_llm.py b/tests/metagpt/test_llm.py index 247f043e2..d46a29c7f 100644 --- a/tests/metagpt/test_llm.py +++ b/tests/metagpt/test_llm.py @@ -4,12 +4,11 @@ @Time : 2023/5/11 14:45 @Author : alexanderwu @File : test_llm.py -@Modified By: mashenquan, 2023/8/20. Remove global configuration `CONFIG`, enable configuration support for business isolation. """ import pytest -from metagpt.provider.openai_api import OpenAILLM as LLM +from metagpt.llm import LLM @pytest.fixture() @@ -23,6 +22,12 @@ async def test_llm_aask(llm): assert len(rsp) > 0 +@pytest.mark.asyncio +async def test_llm_aask_stream(llm): + rsp = await llm.aask("hello world", stream=True) + assert len(rsp) > 0 + + @pytest.mark.asyncio async def test_llm_acompletion(llm): hello_msg = [{"role": "user", "content": "hello"}] diff --git a/tests/metagpt/test_manager.py b/tests/metagpt/test_manager.py deleted file mode 100644 index 5c2a2c795..000000000 --- a/tests/metagpt/test_manager.py +++ /dev/null @@ -1,7 +0,0 @@ -#!/usr/bin/env python -# -*- coding: utf-8 -*- -""" -@Time : 2023/5/11 14:45 -@Author : alexanderwu -@File : test_manager.py -""" diff --git a/tests/metagpt/test_role.py b/tests/metagpt/test_role.py index 52d08e92e..7e707803b 100644 --- a/tests/metagpt/test_role.py +++ b/tests/metagpt/test_role.py @@ -33,16 +33,16 @@ class MockAction(Action): class MockRole(Role): def __init__(self, name="", profile="", goal="", constraints="", desc=""): super().__init__(name=name, profile=profile, goal=goal, constraints=constraints, desc=desc) - self._init_actions([MockAction()]) + self.set_actions([MockAction()]) def test_basic(): mock_role = MockRole() - assert mock_role.subscription == {"tests.metagpt.test_role.MockRole"} + assert mock_role.addresses == ({"tests.metagpt.test_role.MockRole"}) assert mock_role.rc.watch == {"metagpt.actions.add_requirement.UserRequirement"} mock_role = MockRole(name="mock_role") - assert mock_role.subscription == {"tests.metagpt.test_role.MockRole", "mock_role"} + assert mock_role.addresses == {"tests.metagpt.test_role.MockRole", "mock_role"} @pytest.mark.asyncio @@ -53,7 +53,7 @@ async def test_react(): goal: str constraints: str desc: str - subscription: str + address: str inputs = [ { @@ -62,7 +62,7 @@ async def test_react(): "goal": "Test", "constraints": "constraints", "desc": "desc", - "subscription": "start", + "address": "start", } ] @@ -71,7 +71,7 @@ async def test_react(): role = MockRole( name=seed.name, profile=seed.profile, goal=seed.goal, constraints=seed.constraints, desc=seed.desc ) - role.subscribe({seed.subscription}) + role.set_addresses({seed.address}) assert role.rc.watch == {any_to_str(UserRequirement)} assert role.name == seed.name assert role.profile == seed.profile @@ -81,20 +81,20 @@ async def test_react(): assert role.is_idle env = Environment() env.add_role(role) - assert env.get_subscription(role) == {seed.subscription} - env.publish_message(Message(content="test", msg_to=seed.subscription)) + assert env.get_addresses(role) == {seed.address} + env.publish_message(Message(content="test", msg_to=seed.address)) assert not role.is_idle while not env.is_idle: await env.run() assert role.is_idle - env.publish_message(Message(content="test", cause_by=seed.subscription)) + env.publish_message(Message(content="test", cause_by=seed.address)) assert not role.is_idle while not env.is_idle: await env.run() assert role.is_idle tag = uuid.uuid4().hex - role.subscribe({tag}) - assert env.get_subscription(role) == {tag} + role.set_addresses({tag}) + assert env.get_addresses(role) == {tag} @pytest.mark.asyncio @@ -111,8 +111,8 @@ async def test_send_to(): def test_init_action(): role = Role() - role.init_actions([MockAction, MockAction]) - assert role.action_count == 2 + role.set_actions([MockAction, MockAction]) + assert len(role.actions) == 2 @pytest.mark.asyncio @@ -127,11 +127,11 @@ async def test_recover(): role.publish_message(None) role.llm = mock_llm - role.init_actions([MockAction, MockAction]) + role.set_actions([MockAction, MockAction]) role.recovered = True role.latest_observed_msg = Message(content="recover_test") role.rc.state = 0 - assert role.todo == any_to_name(MockAction) + assert role.action_description == any_to_name(MockAction) rsp = await role.run() assert rsp.cause_by == any_to_str(MockAction) @@ -144,7 +144,7 @@ async def test_think_act(): mock_llm.aask.side_effect = ["ok"] role = Role() - role.init_actions([MockAction]) + role.set_actions([MockAction]) await role.think() role.rc.memory.add(Message("run")) assert len(role.get_memories()) == 1 diff --git a/tests/metagpt/test_schema.py b/tests/metagpt/test_schema.py index ab2e206a4..a8fa27151 100644 --- a/tests/metagpt/test_schema.py +++ b/tests/metagpt/test_schema.py @@ -15,7 +15,6 @@ import pytest from metagpt.actions import Action from metagpt.actions.action_node import ActionNode from metagpt.actions.write_code import WriteCode -from metagpt.config import CONFIG from metagpt.const import SYSTEM_DESIGN_FILE_REPO, TASK_FILE_REPO from metagpt.schema import ( AIMessage, @@ -47,6 +46,143 @@ def test_messages(): assert all([i in text for i in roles]) +def test_message(): + Message("a", role="v1") + + m = Message(content="a", role="v1") + v = m.dump() + d = json.loads(v) + assert d + assert d.get("content") == "a" + assert d.get("role") == "v1" + m.role = "v2" + v = m.dump() + assert v + m = Message.load(v) + assert m.content == "a" + assert m.role == "v2" + + m = Message(content="a", role="b", cause_by="c", x="d", send_to="c") + assert m.content == "a" + assert m.role == "b" + assert m.send_to == {"c"} + assert m.cause_by == "c" + m.sent_from = "e" + assert m.sent_from == "e" + + m.cause_by = "Message" + assert m.cause_by == "Message" + m.cause_by = Action + assert m.cause_by == any_to_str(Action) + m.cause_by = Action() + assert m.cause_by == any_to_str(Action) + m.content = "b" + assert m.content == "b" + + +def test_routes(): + m = Message(content="a", role="b", cause_by="c", x="d", send_to="c") + m.send_to = "b" + assert m.send_to == {"b"} + m.send_to = {"e", Action} + assert m.send_to == {"e", any_to_str(Action)} + + +def test_message_serdeser(): + out_mapping = {"field3": (str, ...), "field4": (list[str], ...)} + out_data = {"field3": "field3 value3", "field4": ["field4 value1", "field4 value2"]} + ic_obj = ActionNode.create_model_class("code", out_mapping) + + message = Message(content="code", instruct_content=ic_obj(**out_data), role="engineer", cause_by=WriteCode) + message_dict = message.model_dump() + assert message_dict["cause_by"] == "metagpt.actions.write_code.WriteCode" + assert message_dict["instruct_content"] == { + "class": "code", + "mapping": {"field3": "(, Ellipsis)", "field4": "(list[str], Ellipsis)"}, + "value": {"field3": "field3 value3", "field4": ["field4 value1", "field4 value2"]}, + } + new_message = Message.model_validate(message_dict) + assert new_message.content == message.content + assert new_message.instruct_content.model_dump() == message.instruct_content.model_dump() + assert new_message.instruct_content == message.instruct_content # TODO + assert new_message.cause_by == message.cause_by + assert new_message.instruct_content.field3 == out_data["field3"] + + message = Message(content="code") + message_dict = message.model_dump() + new_message = Message(**message_dict) + assert new_message.instruct_content is None + assert new_message.cause_by == "metagpt.actions.add_requirement.UserRequirement" + assert not Message.load("{") + + +def test_document(): + doc = Document(root_path="a", filename="b", content="c") + meta_doc = doc.get_meta() + assert doc.root_path == meta_doc.root_path + assert doc.filename == meta_doc.filename + assert meta_doc.content == "" + + +@pytest.mark.asyncio +async def test_message_queue(): + mq = MessageQueue() + val = await mq.dump() + assert val == "[]" + mq.push(Message(content="1")) + mq.push(Message(content="2中文测试aaa")) + msg = mq.pop() + assert msg.content == "1" + + val = await mq.dump() + assert val + new_mq = MessageQueue.load(val) + assert new_mq.pop_all() == mq.pop_all() + + +@pytest.mark.parametrize( + ("file_list", "want"), + [ + ( + [f"{SYSTEM_DESIGN_FILE_REPO}/a.txt", f"{TASK_FILE_REPO}/b.txt"], + CodeSummarizeContext( + design_filename=f"{SYSTEM_DESIGN_FILE_REPO}/a.txt", task_filename=f"{TASK_FILE_REPO}/b.txt" + ), + ) + ], +) +def test_CodeSummarizeContext(file_list, want): + ctx = CodeSummarizeContext.loads(file_list) + assert ctx == want + m = {ctx: ctx} + assert want in m + + +def test_class_view(): + attr_a = ClassAttribute(name="a", value_type="int", default_value="0", visibility="+", abstraction=True) + assert attr_a.get_mermaid(align=1) == "\t+int a=0*" + attr_b = ClassAttribute(name="b", value_type="str", default_value="0", visibility="#", static=True) + assert attr_b.get_mermaid(align=0) == '#str b="0"$' + class_view = ClassView(name="A") + class_view.attributes = [attr_a, attr_b] + + method_a = ClassMethod(name="run", visibility="+", abstraction=True) + assert method_a.get_mermaid(align=1) == "\t+run()*" + method_b = ClassMethod( + name="_test", + visibility="#", + static=True, + args=[ClassAttribute(name="a", value_type="str"), ClassAttribute(name="b", value_type="int")], + return_type="str", + ) + assert method_b.get_mermaid(align=0) == "#_test(str a,int b):str$" + class_view.methods = [method_a, method_b] + assert ( + class_view.get_mermaid(align=0) + == 'class A{\n\t+int a=0*\n\t#str b="0"$\n\t+run()*\n\t#_test(str a,int b):str$\n}\n' + ) + + class TestPlan: def test_add_tasks_ordering(self): plan = Plan(goal="") @@ -215,144 +351,5 @@ class TestPlan: assert plan.current_task_id == "2" -def test_message(): - Message("a", role="v1") - - m = Message(content="a", role="v1") - v = m.dump() - d = json.loads(v) - assert d - assert d.get("content") == "a" - assert d.get("role") == "v1" - m.role = "v2" - v = m.dump() - assert v - m = Message.load(v) - assert m.content == "a" - assert m.role == "v2" - - m = Message(content="a", role="b", cause_by="c", x="d", send_to="c") - assert m.content == "a" - assert m.role == "b" - assert m.send_to == {"c"} - assert m.cause_by == "c" - m.sent_from = "e" - assert m.sent_from == "e" - - m.cause_by = "Message" - assert m.cause_by == "Message" - m.cause_by = Action - assert m.cause_by == any_to_str(Action) - m.cause_by = Action() - assert m.cause_by == any_to_str(Action) - m.content = "b" - assert m.content == "b" - - -def test_routes(): - m = Message(content="a", role="b", cause_by="c", x="d", send_to="c") - m.send_to = "b" - assert m.send_to == {"b"} - m.send_to = {"e", Action} - assert m.send_to == {"e", any_to_str(Action)} - - -def test_message_serdeser(): - out_mapping = {"field3": (str, ...), "field4": (list[str], ...)} - out_data = {"field3": "field3 value3", "field4": ["field4 value1", "field4 value2"]} - ic_obj = ActionNode.create_model_class("code", out_mapping) - - message = Message(content="code", instruct_content=ic_obj(**out_data), role="engineer", cause_by=WriteCode) - message_dict = message.model_dump() - assert message_dict["cause_by"] == "metagpt.actions.write_code.WriteCode" - assert message_dict["instruct_content"] == { - "class": "code", - "mapping": {"field3": "(, Ellipsis)", "field4": "(list[str], Ellipsis)"}, - "value": {"field3": "field3 value3", "field4": ["field4 value1", "field4 value2"]}, - } - new_message = Message.model_validate(message_dict) - assert new_message.content == message.content - assert new_message.instruct_content.model_dump() == message.instruct_content.model_dump() - assert new_message.instruct_content != message.instruct_content # TODO - assert new_message.cause_by == message.cause_by - assert new_message.instruct_content.field3 == out_data["field3"] - - message = Message(content="code") - message_dict = message.model_dump() - new_message = Message(**message_dict) - assert new_message.instruct_content is None - assert new_message.cause_by == "metagpt.actions.add_requirement.UserRequirement" - assert not Message.load("{") - - -def test_document(): - doc = Document(root_path="a", filename="b", content="c") - meta_doc = doc.get_meta() - assert doc.root_path == meta_doc.root_path - assert doc.filename == meta_doc.filename - assert meta_doc.content == "" - - assert doc.full_path == str(CONFIG.git_repo.workdir / doc.root_path / doc.filename) - - -@pytest.mark.asyncio -async def test_message_queue(): - mq = MessageQueue() - val = await mq.dump() - assert val == "[]" - mq.push(Message(content="1")) - mq.push(Message(content="2中文测试aaa")) - msg = mq.pop() - assert msg.content == "1" - - val = await mq.dump() - assert val - new_mq = MessageQueue.load(val) - assert new_mq.pop_all() == mq.pop_all() - - -@pytest.mark.parametrize( - ("file_list", "want"), - [ - ( - [f"{SYSTEM_DESIGN_FILE_REPO}/a.txt", f"{TASK_FILE_REPO}/b.txt"], - CodeSummarizeContext( - design_filename=f"{SYSTEM_DESIGN_FILE_REPO}/a.txt", task_filename=f"{TASK_FILE_REPO}/b.txt" - ), - ) - ], -) -def test_CodeSummarizeContext(file_list, want): - ctx = CodeSummarizeContext.loads(file_list) - assert ctx == want - m = {ctx: ctx} - assert want in m - - -def test_class_view(): - attr_a = ClassAttribute(name="a", value_type="int", default_value="0", visibility="+", abstraction=True) - assert attr_a.get_mermaid(align=1) == "\t+int a=0*" - attr_b = ClassAttribute(name="b", value_type="str", default_value="0", visibility="#", static=True) - assert attr_b.get_mermaid(align=0) == '#str b="0"$' - class_view = ClassView(name="A") - class_view.attributes = [attr_a, attr_b] - - method_a = ClassMethod(name="run", visibility="+", abstraction=True) - assert method_a.get_mermaid(align=1) == "\t+run()*" - method_b = ClassMethod( - name="_test", - visibility="#", - static=True, - args=[ClassAttribute(name="a", value_type="str"), ClassAttribute(name="b", value_type="int")], - return_type="str", - ) - assert method_b.get_mermaid(align=0) == "#_test(str a,int b):str$" - class_view.methods = [method_a, method_b] - assert ( - class_view.get_mermaid(align=0) - == 'class A{\n\t+int a=0*\n\t#str b="0"$\n\t+run()*\n\t#_test(str a,int b):str$\n}\n' - ) - - if __name__ == "__main__": pytest.main([__file__, "-s"]) diff --git a/tests/metagpt/tools/libs/test_udf.py b/tests/metagpt/tools/libs/test_udf.py deleted file mode 100644 index 19e523448..000000000 --- a/tests/metagpt/tools/libs/test_udf.py +++ /dev/null @@ -1,49 +0,0 @@ -import json - -import yaml - -from metagpt.logs import logger -from metagpt.tools.libs.udf import UDFS, UDFS_YAML, docstring_to_yaml - - -def test_udfs(): - assert len(UDFS) > 0 - assert "udf_name" in UDFS[0] - assert "udf_doc" in UDFS[0] - logger.info(UDFS) - - -def test_docstring2yaml(): - docstring = """Calculate the duration in hours between two datetime columns. - - Args: - dataframe (pd.DataFrame): The dataframe containing the datetime columns. - - Returns: - pd.DataFrame: The dataframe with an additional column 'duration_hour' added. - """ - - yaml_result = docstring_to_yaml(docstring, return_vars="dataframe") - assert "parameters" in yaml_result - assert "properties" in yaml_result["parameters"] - assert "dataframe" in yaml_result["parameters"]["properties"] - - -def test_UDFS_YAML(): - assert len(UDFS_YAML) > 0 - logger.info(f"\n\n{json.dumps(UDFS_YAML, indent=2, ensure_ascii=False)}") - function_schema = UDFS_YAML - assert "description" in function_schema[list(function_schema.keys())[0]] - assert "type" in function_schema[list(function_schema.keys())[0]] - assert "parameters" in function_schema[list(function_schema.keys())[0]] - assert "properties" in function_schema[list(function_schema.keys())[0]]["parameters"] - assert "required" in function_schema[list(function_schema.keys())[0]]["parameters"] - assert "returns" in function_schema[list(function_schema.keys())[0]] - # 指定要保存的文件路径 - file_path = "./tests/data/function_schema.yaml" - - # 使用 PyYAML 将字典保存为 YAML 文件 - with open(file_path, "w") as file: - yaml.dump(function_schema, file, default_flow_style=False) - - print(f"Data has been saved to {file_path}") diff --git a/tests/metagpt/tools/libs/test_web_scraping.py b/tests/metagpt/tools/libs/test_web_scraping.py new file mode 100644 index 000000000..c11960e68 --- /dev/null +++ b/tests/metagpt/tools/libs/test_web_scraping.py @@ -0,0 +1,23 @@ +import pytest + +from metagpt.tools.libs.web_scraping import scrape_web_playwright + + +@pytest.mark.asyncio +async def test_scrape_web_playwright(): + test_url = "https://www.deepwisdom.ai" + + result = await scrape_web_playwright(test_url) + + # Assert that the result is a dictionary + assert isinstance(result, dict) + + # Assert that the result contains 'inner_text' and 'html' keys + assert "inner_text" in result + assert "html" in result + + # Assert startswith and endswith + assert not result["inner_text"].startswith(" ") + assert not result["inner_text"].endswith(" ") + assert not result["html"].startswith(" ") + assert not result["html"].endswith(" ") diff --git a/tests/metagpt/tools/test_azure_tts.py b/tests/metagpt/tools/test_azure_tts.py index 38fef557e..74d23e439 100644 --- a/tests/metagpt/tools/test_azure_tts.py +++ b/tests/metagpt/tools/test_azure_tts.py @@ -7,21 +7,31 @@ @Modified By: mashenquan, 2023-8-9, add more text formatting options @Modified By: mashenquan, 2023-8-17, move to `tools` folder. """ +from pathlib import Path import pytest -from azure.cognitiveservices.speech import ResultReason +from azure.cognitiveservices.speech import ResultReason, SpeechSynthesizer -from metagpt.config import CONFIG +from metagpt.config2 import config from metagpt.tools.azure_tts import AzureTTS @pytest.mark.asyncio -async def test_azure_tts(): - # Prerequisites - assert CONFIG.AZURE_TTS_SUBSCRIPTION_KEY and CONFIG.AZURE_TTS_SUBSCRIPTION_KEY != "YOUR_API_KEY" - assert CONFIG.AZURE_TTS_REGION +async def test_azure_tts(mocker): + # mock + mock_result = mocker.Mock() + mock_result.audio_data = b"mock audio data" + mock_result.reason = ResultReason.SynthesizingAudioCompleted + mock_data = mocker.Mock() + mock_data.get.return_value = mock_result + mocker.patch.object(SpeechSynthesizer, "speak_ssml_async", return_value=mock_data) + mocker.patch.object(Path, "exists", return_value=True) - azure_tts = AzureTTS(subscription_key="", region="") + # Prerequisites + assert config.AZURE_TTS_SUBSCRIPTION_KEY and config.AZURE_TTS_SUBSCRIPTION_KEY != "YOUR_API_KEY" + assert config.AZURE_TTS_REGION + + azure_tts = AzureTTS(subscription_key=config.AZURE_TTS_SUBSCRIPTION_KEY, region=config.AZURE_TTS_REGION) text = """ 女儿看见父亲走了进来，问道： @@ -32,7 +42,7 @@ async def test_azure_tts(): “Writing a binary file in Python is similar to writing a regular text file, but you'll work with bytes instead of strings.” """ - path = CONFIG.workspace_path / "tts" + path = config.workspace.path / "tts" path.mkdir(exist_ok=True, parents=True) filename = path / "girl.wav" filename.unlink(missing_ok=True) diff --git a/tests/metagpt/tools/test_iflytek_tts.py b/tests/metagpt/tools/test_iflytek_tts.py index 58d8a83ce..8e4c0cf54 100644 --- a/tests/metagpt/tools/test_iflytek_tts.py +++ b/tests/metagpt/tools/test_iflytek_tts.py @@ -7,22 +7,32 @@ """ import pytest -from metagpt.config import CONFIG -from metagpt.tools.iflytek_tts import oas3_iflytek_tts +from metagpt.config2 import Config +from metagpt.tools.iflytek_tts import IFlyTekTTS, oas3_iflytek_tts @pytest.mark.asyncio -async def test_tts(): +async def test_iflytek_tts(mocker): + # mock + config = Config.default() + config.AZURE_TTS_SUBSCRIPTION_KEY = None + config.AZURE_TTS_REGION = None + mocker.patch.object(IFlyTekTTS, "synthesize_speech", return_value=None) + mock_data = mocker.AsyncMock() + mock_data.read.return_value = b"mock iflytek" + mock_reader = mocker.patch("aiofiles.open") + mock_reader.return_value.__aenter__.return_value = mock_data + # Prerequisites - assert CONFIG.IFLYTEK_APP_ID - assert CONFIG.IFLYTEK_API_KEY - assert CONFIG.IFLYTEK_API_SECRET + assert config.IFLYTEK_APP_ID + assert config.IFLYTEK_API_KEY + assert config.IFLYTEK_API_SECRET result = await oas3_iflytek_tts( text="你好，hello", - app_id=CONFIG.IFLYTEK_APP_ID, - api_key=CONFIG.IFLYTEK_API_KEY, - api_secret=CONFIG.IFLYTEK_API_SECRET, + app_id=config.IFLYTEK_APP_ID, + api_key=config.IFLYTEK_API_KEY, + api_secret=config.IFLYTEK_API_SECRET, ) assert result diff --git a/tests/metagpt/tools/test_metagpt_oas3_api_svc.py b/tests/metagpt/tools/test_metagpt_oas3_api_svc.py index 5f52b28cc..5be139106 100644 --- a/tests/metagpt/tools/test_metagpt_oas3_api_svc.py +++ b/tests/metagpt/tools/test_metagpt_oas3_api_svc.py @@ -12,14 +12,12 @@ from pathlib import Path import pytest import requests -from metagpt.config import CONFIG - @pytest.mark.asyncio -async def test_oas2_svc(): +async def test_oas2_svc(context): workdir = Path(__file__).parent.parent.parent.parent script_pathname = workdir / "metagpt/tools/metagpt_oas3_api_svc.py" - env = CONFIG.new_environ() + env = context.new_environ() env["PYTHONPATH"] = str(workdir) + ":" + env.get("PYTHONPATH", "") process = subprocess.Popen(["python", str(script_pathname)], cwd=str(workdir), env=env) await asyncio.sleep(5) diff --git a/tests/metagpt/tools/test_metagpt_text_to_image.py b/tests/metagpt/tools/test_metagpt_text_to_image.py index b765119f0..0dcad20d2 100644 --- a/tests/metagpt/tools/test_metagpt_text_to_image.py +++ b/tests/metagpt/tools/test_metagpt_text_to_image.py @@ -10,7 +10,7 @@ from unittest.mock import AsyncMock import pytest -from metagpt.config import CONFIG +from metagpt.config2 import config from metagpt.tools.metagpt_text_to_image import oas3_metagpt_text_to_image @@ -24,7 +24,7 @@ async def test_draw(mocker): mock_post.return_value.__aenter__.return_value = mock_response # Prerequisites - assert CONFIG.METAGPT_TEXT_TO_IMAGE_MODEL_URL + assert config.METAGPT_TEXT_TO_IMAGE_MODEL_URL binary_data = await oas3_metagpt_text_to_image("Panda emoji") assert binary_data diff --git a/tests/metagpt/tools/test_moderation.py b/tests/metagpt/tools/test_moderation.py index 534fe812a..8dc9e9d5e 100644 --- a/tests/metagpt/tools/test_moderation.py +++ b/tests/metagpt/tools/test_moderation.py @@ -8,7 +8,8 @@ import pytest -from metagpt.config import CONFIG +from metagpt.config2 import config +from metagpt.llm import LLM from metagpt.tools.moderation import Moderation @@ -23,11 +24,9 @@ from metagpt.tools.moderation import Moderation ) async def test_amoderation(content): # Prerequisites - assert CONFIG.OPENAI_API_KEY and CONFIG.OPENAI_API_KEY != "YOUR_API_KEY" - assert not CONFIG.OPENAI_API_TYPE - assert CONFIG.OPENAI_API_MODEL + assert config.get_openai_llm() - moderation = Moderation() + moderation = Moderation(LLM()) results = await moderation.amoderation(content=content) assert isinstance(results, list) assert len(results) == len(content) diff --git a/tests/metagpt/tools/test_openai_text_to_embedding.py b/tests/metagpt/tools/test_openai_text_to_embedding.py index 086c9d45b..81b3895c3 100644 --- a/tests/metagpt/tools/test_openai_text_to_embedding.py +++ b/tests/metagpt/tools/test_openai_text_to_embedding.py @@ -5,21 +5,36 @@ @Author : mashenquan @File : test_openai_text_to_embedding.py """ +import json +from pathlib import Path import pytest -from metagpt.config import CONFIG +from metagpt.config2 import Config from metagpt.tools.openai_text_to_embedding import oas3_openai_text_to_embedding +from metagpt.utils.common import aread @pytest.mark.asyncio -async def test_embedding(): - # Prerequisites - assert CONFIG.OPENAI_API_KEY and CONFIG.OPENAI_API_KEY != "YOUR_API_KEY" - assert not CONFIG.OPENAI_API_TYPE - assert CONFIG.OPENAI_API_MODEL +async def test_embedding(mocker): + # mock + config = Config.default() + mock_post = mocker.patch("aiohttp.ClientSession.post") + mock_response = mocker.AsyncMock() + mock_response.status = 200 + data = await aread(Path(__file__).parent / "../../data/openai/embedding.json") + mock_response.json.return_value = json.loads(data) + mock_post.return_value.__aenter__.return_value = mock_response + type(config.get_openai_llm()).proxy = mocker.PropertyMock(return_value="http://mock.proxy") - result = await oas3_openai_text_to_embedding("Panda emoji") + # Prerequisites + llm_config = config.get_openai_llm() + assert llm_config + assert llm_config.proxy + + result = await oas3_openai_text_to_embedding( + "Panda emoji", openai_api_key=llm_config.api_key, proxy=llm_config.proxy + ) assert result assert result.model assert len(result.data) > 0 diff --git a/tests/metagpt/tools/test_openai_text_to_image.py b/tests/metagpt/tools/test_openai_text_to_image.py index e560da798..3f9169ddd 100644 --- a/tests/metagpt/tools/test_openai_text_to_image.py +++ b/tests/metagpt/tools/test_openai_text_to_image.py @@ -5,24 +5,44 @@ @Author : mashenquan @File : test_openai_text_to_image.py """ +import base64 +import openai import pytest +from pydantic import BaseModel -from metagpt.config import CONFIG +from metagpt.config2 import config +from metagpt.llm import LLM from metagpt.tools.openai_text_to_image import ( OpenAIText2Image, oas3_openai_text_to_image, ) +from metagpt.utils.s3 import S3 @pytest.mark.asyncio -async def test_draw(): - # Prerequisites - assert CONFIG.OPENAI_API_KEY and CONFIG.OPENAI_API_KEY != "YOUR_API_KEY" - assert not CONFIG.OPENAI_API_TYPE - assert CONFIG.OPENAI_API_MODEL +async def test_draw(mocker): + # mock + mock_url = mocker.Mock() + mock_url.url.return_value = "http://mock.com/0.png" - binary_data = await oas3_openai_text_to_image("Panda emoji") + class _MockData(BaseModel): + data: list + + mock_data = _MockData(data=[mock_url]) + mocker.patch.object(openai.resources.images.AsyncImages, "generate", return_value=mock_data) + mock_post = mocker.patch("aiohttp.ClientSession.get") + mock_response = mocker.AsyncMock() + mock_response.status = 200 + mock_response.read.return_value = base64.b64encode(b"success") + mock_post.return_value.__aenter__.return_value = mock_response + mocker.patch.object(S3, "cache", return_value="http://mock.s3.com/0.png") + + # Prerequisites + llm_config = config.get_openai_llm() + assert llm_config + + binary_data = await oas3_openai_text_to_image("Panda emoji", llm=LLM(llm_config=llm_config)) assert binary_data diff --git a/tests/metagpt/tools/test_openapi_v3_hello.py b/tests/metagpt/tools/test_openapi_v3_hello.py index 5726cf8e0..f49b8412a 100644 --- a/tests/metagpt/tools/test_openapi_v3_hello.py +++ b/tests/metagpt/tools/test_openapi_v3_hello.py @@ -12,14 +12,12 @@ from pathlib import Path import pytest import requests -from metagpt.config import CONFIG - @pytest.mark.asyncio -async def test_hello(): +async def test_hello(context): workdir = Path(__file__).parent.parent.parent.parent script_pathname = workdir / "metagpt/tools/openapi_v3_hello.py" - env = CONFIG.new_environ() + env = context.new_environ() env["PYTHONPATH"] = str(workdir) + ":" + env.get("PYTHONPATH", "") process = subprocess.Popen(["python", str(script_pathname)], cwd=workdir, env=env) await asyncio.sleep(5) diff --git a/tests/metagpt/tools/test_search_engine.py b/tests/metagpt/tools/test_search_engine.py index dab466af7..966f53a38 100644 --- a/tests/metagpt/tools/test_search_engine.py +++ b/tests/metagpt/tools/test_search_engine.py @@ -7,20 +7,15 @@ """ from __future__ import annotations -import json -from pathlib import Path from typing import Callable import pytest -import tests.data.search -from metagpt.config import CONFIG +from metagpt.config2 import config from metagpt.logs import logger from metagpt.tools import SearchEngineType from metagpt.tools.search_engine import SearchEngine -search_cache_path = Path(tests.data.search.__path__[0]) - class MockSearchEnine: async def run(self, query: str, max_results: int = 8, as_string: bool = True) -> str | list[dict[str, str]]: @@ -46,24 +41,28 @@ class MockSearchEnine: (SearchEngineType.CUSTOM_ENGINE, MockSearchEnine().run, 6, False), ], ) -async def test_search_engine(search_engine_type, run_func: Callable, max_results: int, as_string: bool, aiohttp_mocker): +async def test_search_engine( + search_engine_type, + run_func: Callable, + max_results: int, + as_string: bool, + search_engine_mocker, +): # Prerequisites - cache_json_path = None - if search_engine_type is SearchEngineType.SERPAPI_GOOGLE: - assert CONFIG.SERPAPI_API_KEY and CONFIG.SERPAPI_API_KEY != "YOUR_API_KEY" - cache_json_path = search_cache_path / f"serpapi-metagpt-{max_results}.json" - elif search_engine_type is SearchEngineType.DIRECT_GOOGLE: - assert CONFIG.GOOGLE_API_KEY and CONFIG.GOOGLE_API_KEY != "YOUR_API_KEY" - assert CONFIG.GOOGLE_CSE_ID and CONFIG.GOOGLE_CSE_ID != "YOUR_CSE_ID" - elif search_engine_type is SearchEngineType.SERPER_GOOGLE: - assert CONFIG.SERPER_API_KEY and CONFIG.SERPER_API_KEY != "YOUR_API_KEY" - cache_json_path = search_cache_path / f"serper-metagpt-{max_results}.json" + search_engine_config = {} - if cache_json_path: - with open(cache_json_path) as f: - data = json.load(f) - aiohttp_mocker.set_json(data) - search_engine = SearchEngine(search_engine_type, run_func) + if search_engine_type is SearchEngineType.SERPAPI_GOOGLE: + assert config.search + search_engine_config["serpapi_api_key"] = "mock-serpapi-key" + elif search_engine_type is SearchEngineType.DIRECT_GOOGLE: + assert config.search + search_engine_config["google_api_key"] = "mock-google-key" + search_engine_config["google_cse_id"] = "mock-google-cse" + elif search_engine_type is SearchEngineType.SERPER_GOOGLE: + assert config.search + search_engine_config["serper_api_key"] = "mock-serper-key" + + search_engine = SearchEngine(search_engine_type, run_func, **search_engine_config) rsp = await search_engine.run("metagpt", max_results, as_string) logger.info(rsp) if as_string: diff --git a/tests/metagpt/tools/test_ut_writer.py b/tests/metagpt/tools/test_ut_writer.py index eac28d56f..29b6572c2 100644 --- a/tests/metagpt/tools/test_ut_writer.py +++ b/tests/metagpt/tools/test_ut_writer.py @@ -9,7 +9,7 @@ from pathlib import Path import pytest -from metagpt.config import CONFIG +from metagpt.config2 import config from metagpt.const import API_QUESTIONS_PATH, UT_PY_PATH from metagpt.tools.ut_writer import YFT_PROMPT_PREFIX, UTGenerator @@ -20,9 +20,7 @@ class TestUTWriter: # Prerequisites swagger_file = Path(__file__).parent / "../../data/ut_writer/yft_swaggerApi.json" assert swagger_file.exists() - assert CONFIG.OPENAI_API_KEY and CONFIG.OPENAI_API_KEY != "YOUR_API_KEY" - assert not CONFIG.OPENAI_API_TYPE - assert CONFIG.OPENAI_API_MODEL + assert config.get_openai_llm() tags = ["测试", "作业"] # 这里在文件中手动加入了两个测试标签的API diff --git a/tests/metagpt/tools/test_web_browser_engine.py b/tests/metagpt/tools/test_web_browser_engine.py index 289edda2f..ceebd67fc 100644 --- a/tests/metagpt/tools/test_web_browser_engine.py +++ b/tests/metagpt/tools/test_web_browser_engine.py @@ -1,6 +1,5 @@ -""" -@Modified By: mashenquan, 2023/8/20. Remove global configuration `CONFIG`, enable configuration support for business isolation. -""" +#!/usr/bin/env python +# -*- coding: utf-8 -*- import pytest diff --git a/tests/metagpt/tools/test_web_browser_engine_playwright.py b/tests/metagpt/tools/test_web_browser_engine_playwright.py index 0f2679531..0e838a2f8 100644 --- a/tests/metagpt/tools/test_web_browser_engine_playwright.py +++ b/tests/metagpt/tools/test_web_browser_engine_playwright.py @@ -1,10 +1,9 @@ -""" -@Modified By: mashenquan, 2023/8/20. Remove global configuration `CONFIG`, enable configuration support for business isolation. -""" +#!/usr/bin/env python +# -*- coding: utf-8 -*- import pytest -from metagpt.config import CONFIG +from metagpt.config2 import config from metagpt.tools import web_browser_engine_playwright from metagpt.utils.parse_html import WebPage @@ -20,11 +19,11 @@ from metagpt.utils.parse_html import WebPage ids=["chromium-normal", "firefox-normal", "webkit-normal"], ) async def test_scrape_web_page(browser_type, use_proxy, kwagrs, url, urls, proxy, capfd): - global_proxy = CONFIG.global_proxy + global_proxy = config.proxy try: if use_proxy: - server, proxy = await proxy - CONFIG.global_proxy = proxy + server, proxy_url = await proxy() + config.proxy = proxy_url browser = web_browser_engine_playwright.PlaywrightWrapper(browser_type=browser_type, **kwagrs) result = await browser.run(url) assert isinstance(result, WebPage) @@ -39,7 +38,7 @@ async def test_scrape_web_page(browser_type, use_proxy, kwagrs, url, urls, proxy server.close() assert "Proxy:" in capfd.readouterr().out finally: - CONFIG.global_proxy = global_proxy + config.proxy = global_proxy if __name__ == "__main__": diff --git a/tests/metagpt/tools/test_web_browser_engine_selenium.py b/tests/metagpt/tools/test_web_browser_engine_selenium.py index 8fe365352..1b1439d29 100644 --- a/tests/metagpt/tools/test_web_browser_engine_selenium.py +++ b/tests/metagpt/tools/test_web_browser_engine_selenium.py @@ -1,10 +1,10 @@ -""" -@Modified By: mashenquan, 2023/8/20. Remove global configuration `CONFIG`, enable configuration support for business isolation. -""" +#!/usr/bin/env python +# -*- coding: utf-8 -*- +import browsers import pytest -from metagpt.config import CONFIG +from metagpt.config2 import config from metagpt.tools import web_browser_engine_selenium from metagpt.utils.parse_html import WebPage @@ -13,9 +13,27 @@ from metagpt.utils.parse_html import WebPage @pytest.mark.parametrize( "browser_type, use_proxy, url, urls", [ - ("chrome", True, "https://deepwisdom.ai", ("https://deepwisdom.ai",)), - ("firefox", False, "https://deepwisdom.ai", ("https://deepwisdom.ai",)), - ("edge", False, "https://deepwisdom.ai", ("https://deepwisdom.ai",)), + pytest.param( + "chrome", + True, + "https://deepwisdom.ai", + ("https://deepwisdom.ai",), + marks=pytest.mark.skipif(not browsers.get("chrome"), reason="chrome browser not found"), + ), + pytest.param( + "firefox", + False, + "https://deepwisdom.ai", + ("https://deepwisdom.ai",), + marks=pytest.mark.skipif(not browsers.get("firefox"), reason="firefox browser not found"), + ), + pytest.param( + "edge", + False, + "https://deepwisdom.ai", + ("https://deepwisdom.ai",), + marks=pytest.mark.skipif(not browsers.get("msedge"), reason="edge browser not found"), + ), ], ids=["chrome-normal", "firefox-normal", "edge-normal"], ) @@ -23,11 +41,11 @@ async def test_scrape_web_page(browser_type, use_proxy, url, urls, proxy, capfd) # Prerequisites # firefox, chrome, Microsoft Edge - global_proxy = CONFIG.global_proxy + global_proxy = config.proxy try: if use_proxy: - server, proxy = await proxy - CONFIG.global_proxy = proxy + server, proxy_url = await proxy() + config.proxy = proxy_url browser = web_browser_engine_selenium.SeleniumWrapper(browser_type=browser_type) result = await browser.run(url) assert isinstance(result, WebPage) @@ -42,7 +60,7 @@ async def test_scrape_web_page(browser_type, use_proxy, url, urls, proxy, capfd) server.close() assert "Proxy:" in capfd.readouterr().out finally: - CONFIG.global_proxy = global_proxy + config.proxy = global_proxy if __name__ == "__main__": diff --git a/tests/metagpt/utils/test_config.py b/tests/metagpt/utils/test_config.py deleted file mode 100644 index 4ca7a225c..000000000 --- a/tests/metagpt/utils/test_config.py +++ /dev/null @@ -1,38 +0,0 @@ -#!/usr/bin/env python -# -*- coding: utf-8 -*- -""" -@Time : 2023/5/1 11:19 -@Author : alexanderwu -@File : test_config.py -@Modified By: mashenquan, 2013/8/20, Add `test_options`; remove global configuration `CONFIG`, enable configuration support for business isolation. -""" -from pathlib import Path - -import pytest - -from metagpt.config import Config - - -def test_config_class_get_key_exception(): - with pytest.raises(Exception) as exc_info: - config = Config() - config.get("wtf") - assert str(exc_info.value) == "Key 'wtf' not found in environment variables or in the YAML file" - - -def test_config_yaml_file_not_exists(): - # FIXME: 由于这里是单例，所以会导致Config重新创建失效。后续要将Config改为非单例模式。 - _ = Config("wtf.yaml") - # with pytest.raises(Exception) as exc_info: - # config.get("OPENAI_BASE_URL") - # assert str(exc_info.value) == "Set OPENAI_API_KEY or Anthropic_API_KEY first" - - -def test_options(): - filename = Path(__file__).resolve().parent.parent.parent.parent / "config/config.yaml" - config = Config(filename) - assert config.options - - -if __name__ == "__main__": - test_options() diff --git a/tests/metagpt/utils/test_human_interaction.py b/tests/metagpt/utils/test_human_interaction.py new file mode 100644 index 000000000..24dbac61c --- /dev/null +++ b/tests/metagpt/utils/test_human_interaction.py @@ -0,0 +1,66 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +# @Desc : unittest of human_interaction + +from pydantic import BaseModel + +from metagpt.utils.human_interaction import HumanInteraction + + +class InstructContent(BaseModel): + test_field1: str = "" + test_field2: list[str] = [] + + +data_mapping = {"test_field1": (str, ...), "test_field2": (list[str], ...)} + +human_interaction = HumanInteraction() + + +def test_input_num(mocker): + mocker.patch("builtins.input", lambda _: "quit") + + interact_contents = human_interaction.interact_with_instruct_content(InstructContent(), data_mapping) + assert len(interact_contents) == 0 + + mocker.patch("builtins.input", lambda _: "1") + input_num = human_interaction.input_num_until_valid(2) + assert input_num == 1 + + +def test_check_input_type(): + ret, _ = human_interaction.check_input_type(input_str="test string", req_type=str) + assert ret + + ret, _ = human_interaction.check_input_type(input_str='["test string"]', req_type=list[str]) + assert ret + + ret, _ = human_interaction.check_input_type(input_str='{"key", "value"}', req_type=list[str]) + assert not ret + + +global_index = 0 + + +def mock_input(*args, **kwargs): + """there are multi input call, return it by global_index""" + arr = ["1", '["test"]', "ignore", "quit"] + global global_index + global_index += 1 + if global_index == 3: + raise EOFError() + val = arr[global_index - 1] + return val + + +def test_human_interact_valid_content(mocker): + mocker.patch("builtins.input", mock_input) + input_contents = HumanInteraction().interact_with_instruct_content(InstructContent(), data_mapping, "review") + assert len(input_contents) == 1 + assert input_contents["test_field2"] == '["test"]' + + global global_index + global_index = 0 + input_contents = HumanInteraction().interact_with_instruct_content(InstructContent(), data_mapping, "revise") + assert len(input_contents) == 1 + assert input_contents["test_field2"] == ["test"] diff --git a/tests/metagpt/utils/test_mermaid.py b/tests/metagpt/utils/test_mermaid.py index b7b97a3f1..7e9129314 100644 --- a/tests/metagpt/utils/test_mermaid.py +++ b/tests/metagpt/utils/test_mermaid.py @@ -8,23 +8,20 @@ import pytest -from metagpt.config import CONFIG from metagpt.utils.common import check_cmd_exists from metagpt.utils.mermaid import MMC1, mermaid_to_file @pytest.mark.asyncio @pytest.mark.parametrize("engine", ["nodejs", "ink"]) # TODO: playwright and pyppeteer -async def test_mermaid(engine): +async def test_mermaid(engine, context): # nodejs prerequisites: npm install -g @mermaid-js/mermaid-cli # ink prerequisites: connected to internet # playwright prerequisites: playwright install --with-deps chromium assert check_cmd_exists("npm") == 0 - assert CONFIG.PYPPETEER_EXECUTABLE_PATH - CONFIG.mermaid_engine = engine - save_to = CONFIG.git_repo.workdir / f"{CONFIG.mermaid_engine}/1" - await mermaid_to_file(MMC1, save_to) + save_to = context.git_repo.workdir / f"{engine}/1" + await mermaid_to_file(engine, MMC1, save_to) # ink does not support pdf if engine == "ink": diff --git a/tests/metagpt/utils/test_project_repo.py b/tests/metagpt/utils/test_project_repo.py new file mode 100644 index 000000000..667927a1d --- /dev/null +++ b/tests/metagpt/utils/test_project_repo.py @@ -0,0 +1,64 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +@Time : 2024/1/8 +@Author : mashenquan +""" +import uuid +from pathlib import Path + +import pytest + +from metagpt.const import ( + BUGFIX_FILENAME, + PACKAGE_REQUIREMENTS_FILENAME, + PRDS_FILE_REPO, + REQUIREMENT_FILENAME, +) +from metagpt.utils.project_repo import ProjectRepo + + +async def test_project_repo(): + root = Path(__file__).parent / f"../../../workspace/unittest/{uuid.uuid4().hex}" + root = root.resolve() + + pr = ProjectRepo(root=str(root)) + assert pr.git_repo.workdir == root + assert pr.workdir == pr.git_repo.workdir + + await pr.save(filename=REQUIREMENT_FILENAME, content=REQUIREMENT_FILENAME) + doc = await pr.get(filename=REQUIREMENT_FILENAME) + assert doc.content == REQUIREMENT_FILENAME + await pr.save(filename=BUGFIX_FILENAME, content=BUGFIX_FILENAME) + doc = await pr.get(filename=BUGFIX_FILENAME) + assert doc.content == BUGFIX_FILENAME + await pr.save(filename=PACKAGE_REQUIREMENTS_FILENAME, content=PACKAGE_REQUIREMENTS_FILENAME) + doc = await pr.get(filename=PACKAGE_REQUIREMENTS_FILENAME) + assert doc.content == PACKAGE_REQUIREMENTS_FILENAME + await pr.docs.prd.save(filename="1.prd", content="1.prd", dependencies=[REQUIREMENT_FILENAME]) + doc = await pr.docs.prd.get(filename="1.prd") + assert doc.content == "1.prd" + await pr.resources.prd.save( + filename="1.prd", + content="1.prd", + dependencies=[REQUIREMENT_FILENAME, f"{PRDS_FILE_REPO}/1.prd"], + ) + doc = await pr.resources.prd.get(filename="1.prd") + assert doc.content == "1.prd" + dependencies = await pr.resources.prd.get_dependency(filename="1.prd") + assert len(dependencies) == 2 + + assert pr.changed_files + assert pr.docs.prd.changed_files + assert not pr.tests.changed_files + + with pytest.raises(ValueError): + pr.srcs + assert pr.with_src_path("test_src").srcs.root_path == Path("test_src") + assert pr.src_relative_path == Path("test_src") + + pr.git_repo.delete_repository() + + +if __name__ == "__main__": + pytest.main([__file__, "-s"]) diff --git a/tests/metagpt/utils/test_redis.py b/tests/metagpt/utils/test_redis.py index 140c04f6b..6fd4250a6 100644 --- a/tests/metagpt/utils/test_redis.py +++ b/tests/metagpt/utils/test_redis.py @@ -9,47 +9,29 @@ from unittest.mock import AsyncMock import pytest -from metagpt.config import CONFIG from metagpt.utils.redis import Redis -async def async_mock_from_url(*args, **kwargs): - mock_client = AsyncMock() - mock_client.set.return_value = None - mock_client.get.side_effect = [b"test", b""] - return mock_client - - @pytest.mark.asyncio async def test_redis(mocker): - # Mock + async def async_mock_from_url(*args, **kwargs): + mock_client = AsyncMock() + mock_client.set.return_value = None + mock_client.get.return_value = b"test" + return mock_client + mocker.patch("aioredis.from_url", return_value=async_mock_from_url()) + mock_config = mocker.Mock() + mock_config.to_url.return_value = "http://mock.com" + mock_config.username = "mockusername" + mock_config.password = "mockpwd" + mock_config.db = "0" - # Prerequisites - CONFIG.REDIS_HOST = "MOCK_REDIS_HOST" - CONFIG.REDIS_PORT = "MOCK_REDIS_PORT" - CONFIG.REDIS_PASSWORD = "MOCK_REDIS_PASSWORD" - CONFIG.REDIS_DB = 0 - - conn = Redis() - assert not conn.is_valid + conn = Redis(mock_config) await conn.set("test", "test", timeout_sec=0) assert await conn.get("test") == b"test" await conn.close() - # Mock session env - old_options = CONFIG.options.copy() - new_options = old_options.copy() - new_options["REDIS_HOST"] = "YOUR_REDIS_HOST" - CONFIG.set_context(new_options) - try: - conn = Redis() - await conn.set("test", "test", timeout_sec=0) - assert not await conn.get("test") == b"test" - await conn.close() - finally: - CONFIG.set_context(old_options) - if __name__ == "__main__": pytest.main([__file__, "-s"]) diff --git a/tests/metagpt/utils/test_repair_llm_raw_output.py b/tests/metagpt/utils/test_repair_llm_raw_output.py index 1970c6443..3ccca3e06 100644 --- a/tests/metagpt/utils/test_repair_llm_raw_output.py +++ b/tests/metagpt/utils/test_repair_llm_raw_output.py @@ -2,13 +2,13 @@ # -*- coding: utf-8 -*- # @Desc : unittest of repair_llm_raw_output -from metagpt.config import CONFIG +from metagpt.config2 import config """ CONFIG.repair_llm_output should be True before retry_parse_json_text imported. so we move `from ... impot ...` into each `test_xx` to avoid `Module level import not at top of file` format warning. """ -CONFIG.repair_llm_output = True +config.repair_llm_output = True def test_repair_case_sensitivity(): @@ -128,6 +128,19 @@ def test_repair_json_format(): output = repair_llm_raw_output(output=raw_output, req_keys=[None], repair_type=RepairType.JSON) assert output == target_output + raw_output = """ +{ + "Language": "en_us", # define language + "Programming Language": "Python" +} +""" + target_output = """{ + "Language": "en_us", + "Programming Language": "Python" +}""" + output = repair_llm_raw_output(output=raw_output, req_keys=[None], repair_type=RepairType.JSON) + assert output == target_output + def test_repair_invalid_json(): from metagpt.utils.repair_llm_raw_output import repair_invalid_json @@ -204,6 +217,25 @@ def test_retry_parse_json_text(): output = retry_parse_json_text(output=invalid_json_text) assert output == target_json + invalid_json_text = '''{ + "Data structures and interfaces": """ + class UI: + - game_engine: GameEngine + + __init__(engine: GameEngine) -> None + + display_board() -> None + + display_score() -> None + + prompt_move() -> str + + reset_game() -> None + """ + "Anything UNCLEAR": "no" +}''' + target_json = { + "Data structures and interfaces": "\n class UI:\n - game_engine: GameEngine\n + __init__(engine: GameEngine) -> None\n + display_board() -> None\n + display_score() -> None\n + prompt_move() -> str\n + reset_game() -> None\n ", + "Anything UNCLEAR": "no", + } + output = retry_parse_json_text(output=invalid_json_text) + assert output == target_json + def test_extract_content_from_output(): """ diff --git a/tests/metagpt/utils/test_s3.py b/tests/metagpt/utils/test_s3.py index 132aa0635..b26ebe94d 100644 --- a/tests/metagpt/utils/test_s3.py +++ b/tests/metagpt/utils/test_s3.py @@ -8,49 +8,42 @@ import uuid from pathlib import Path +import aioboto3 import aiofiles -import mock import pytest -from metagpt.config import CONFIG +from metagpt.config2 import Config from metagpt.utils.common import aread from metagpt.utils.s3 import S3 @pytest.mark.asyncio -@mock.patch("aioboto3.Session") -async def test_s3(mock_session_class): +async def test_s3(mocker): # Set up the mock response data = await aread(__file__, "utf-8") - mock_session_object = mock.Mock() - reader_mock = mock.AsyncMock() + reader_mock = mocker.AsyncMock() reader_mock.read.side_effect = [data.encode("utf-8"), b"", data.encode("utf-8")] - type(reader_mock).url = mock.PropertyMock(return_value="https://mock") - mock_client = mock.AsyncMock() + type(reader_mock).url = mocker.PropertyMock(return_value="https://mock") + mock_client = mocker.AsyncMock() mock_client.put_object.return_value = None mock_client.get_object.return_value = {"Body": reader_mock} mock_client.__aenter__.return_value = mock_client mock_client.__aexit__.return_value = None - mock_session_object.client.return_value = mock_client - mock_session_class.return_value = mock_session_object + mocker.patch.object(aioboto3.Session, "client", return_value=mock_client) # Prerequisites - # assert CONFIG.S3_ACCESS_KEY and CONFIG.S3_ACCESS_KEY != "YOUR_S3_ACCESS_KEY" - # assert CONFIG.S3_SECRET_KEY and CONFIG.S3_SECRET_KEY != "YOUR_S3_SECRET_KEY" - # assert CONFIG.S3_ENDPOINT_URL and CONFIG.S3_ENDPOINT_URL != "YOUR_S3_ENDPOINT_URL" - # assert CONFIG.S3_BUCKET and CONFIG.S3_BUCKET != "YOUR_S3_BUCKET" - - conn = S3() - assert conn.is_valid + s3 = Config.default().s3 + assert s3 + conn = S3(s3) object_name = "unittest.bak" - await conn.upload_file(bucket=CONFIG.S3_BUCKET, local_path=__file__, object_name=object_name) + await conn.upload_file(bucket=s3.bucket, local_path=__file__, object_name=object_name) pathname = (Path(__file__).parent / uuid.uuid4().hex).with_suffix(".bak") pathname.unlink(missing_ok=True) - await conn.download_file(bucket=CONFIG.S3_BUCKET, object_name=object_name, local_path=str(pathname)) + await conn.download_file(bucket=s3.bucket, object_name=object_name, local_path=str(pathname)) assert pathname.exists() - url = await conn.get_object_url(bucket=CONFIG.S3_BUCKET, object_name=object_name) + url = await conn.get_object_url(bucket=s3.bucket, object_name=object_name) assert url - bin_data = await conn.get_object(bucket=CONFIG.S3_BUCKET, object_name=object_name) + bin_data = await conn.get_object(bucket=s3.bucket, object_name=object_name) assert bin_data async with aiofiles.open(__file__, mode="r", encoding="utf-8") as reader: data = await reader.read() @@ -58,18 +51,14 @@ async def test_s3(mock_session_class): assert "http" in res # Mock session env - type(reader_mock).url = mock.PropertyMock(return_value="") - old_options = CONFIG.options.copy() - new_options = old_options.copy() - new_options["S3_ACCESS_KEY"] = "YOUR_S3_ACCESS_KEY" - CONFIG.set_context(new_options) + s3.access_key = "ABC" + type(reader_mock).url = mocker.PropertyMock(return_value="") try: - conn = S3() - assert not conn.is_valid + conn = S3(s3) res = await conn.cache("ABC", ".bak", "script") assert not res - finally: - CONFIG.set_context(old_options) + except Exception: + pass await reader.close() diff --git a/tests/metagpt/utils/test_save_code.py b/tests/metagpt/utils/test_save_code.py index 278d9a539..0674315d0 100644 --- a/tests/metagpt/utils/test_save_code.py +++ b/tests/metagpt/utils/test_save_code.py @@ -9,7 +9,6 @@ import nbformat import pytest from metagpt.actions.execute_code import ExecutePyCode -from metagpt.actions.write_analysis_code import WriteCodeByGenerate from metagpt.utils.save_code import DATA_PATH, save_code_file @@ -17,11 +16,6 @@ def test_save_code_file_python(): save_code_file("example", "print('Hello, World!')") file_path = DATA_PATH / "output" / "example" / "code.py" assert os.path.exists(file_path), f"File does not exist: {file_path}" - - -def test_save_code_file_python(): - save_code_file("example", "print('Hello, World!')") - file_path = DATA_PATH / "output" / "example" / "code.py" with open(file_path, "r", encoding="utf-8") as fp: content = fp.read() assert "print('Hello, World!')" in content, "File content does not match" @@ -38,7 +32,7 @@ def test_save_code_file_json(): @pytest.mark.asyncio async def test_save_code_file_notebook(): - code = await WriteCodeByGenerate().run(context="basic python, hello world", plan="", code_steps="", temperature=0.0) + code = "print('Hello, World!')" executor = ExecutePyCode() await executor.run(code) # Save as a Notebook file diff --git a/tests/mock/mock_aiohttp.py b/tests/mock/mock_aiohttp.py new file mode 100644 index 000000000..4690bf4b5 --- /dev/null +++ b/tests/mock/mock_aiohttp.py @@ -0,0 +1,41 @@ +import json +from typing import Callable + +from aiohttp.client import ClientSession + +origin_request = ClientSession.request + + +class MockAioResponse: + check_funcs: dict[tuple[str, str], Callable[[dict], str]] = {} + rsp_cache: dict[str, str] = {} + name = "aiohttp" + + def __init__(self, session, method, url, **kwargs) -> None: + fn = self.check_funcs.get((method, url)) + self.key = f"{self.name}-{method}-{url}-{fn(kwargs) if fn else json.dumps(kwargs, sort_keys=True)}" + self.mng = self.response = None + if self.key not in self.rsp_cache: + self.mng = origin_request(session, method, url, **kwargs) + + async def __aenter__(self): + if self.response: + await self.response.__aenter__() + elif self.mng: + self.response = await self.mng.__aenter__() + return self + + async def __aexit__(self, *args, **kwargs): + if self.response: + await self.response.__aexit__(*args, **kwargs) + self.response = None + elif self.mng: + await self.mng.__aexit__(*args, **kwargs) + self.mng = None + + async def json(self, *args, **kwargs): + if self.key in self.rsp_cache: + return self.rsp_cache[self.key] + data = await self.response.json(*args, **kwargs) + self.rsp_cache[self.key] = data + return data diff --git a/tests/mock/mock_curl_cffi.py b/tests/mock/mock_curl_cffi.py new file mode 100644 index 000000000..3f2bea4a7 --- /dev/null +++ b/tests/mock/mock_curl_cffi.py @@ -0,0 +1,22 @@ +import json +from typing import Callable + +from curl_cffi import requests + +origin_request = requests.Session.request + + +class MockCurlCffiResponse(requests.Response): + check_funcs: dict[tuple[str, str], Callable[[dict], str]] = {} + rsp_cache: dict[str, str] = {} + name = "curl-cffi" + + def __init__(self, session, method, url, **kwargs) -> None: + super().__init__() + fn = self.check_funcs.get((method, url)) + self.key = f"{self.name}-{method}-{url}-{fn(kwargs) if fn else json.dumps(kwargs, sort_keys=True)}" + self.response = None + if self.key not in self.rsp_cache: + response = origin_request(session, method, url, **kwargs) + self.rsp_cache[self.key] = response.content.decode() + self.content = self.rsp_cache[self.key].encode() diff --git a/tests/mock/mock_httplib2.py b/tests/mock/mock_httplib2.py new file mode 100644 index 000000000..b6dd0b77b --- /dev/null +++ b/tests/mock/mock_httplib2.py @@ -0,0 +1,29 @@ +import json +from typing import Callable +from urllib.parse import parse_qsl, urlparse + +import httplib2 + +origin_request = httplib2.Http.request + + +class MockHttplib2Response(httplib2.Response): + check_funcs: dict[tuple[str, str], Callable[[dict], str]] = {} + rsp_cache: dict[str, str] = {} + name = "httplib2" + + def __init__(self, http, uri, method="GET", **kwargs) -> None: + url = uri.split("?")[0] + result = urlparse(uri) + params = dict(parse_qsl(result.query)) + fn = self.check_funcs.get((method, uri)) + new_kwargs = {"params": params} + key = f"{self.name}-{method}-{url}-{fn(new_kwargs) if fn else json.dumps(new_kwargs)}" + if key not in self.rsp_cache: + _, self.content = origin_request(http, uri, method, **kwargs) + self.rsp_cache[key] = self.content.decode() + self.content = self.rsp_cache[key] + + def __iter__(self): + yield self + yield self.content.encode() diff --git a/tests/mock/mock_llm.py b/tests/mock/mock_llm.py index a52aeed09..e2fff214f 100644 --- a/tests/mock/mock_llm.py +++ b/tests/mock/mock_llm.py @@ -1,18 +1,19 @@ import json from typing import Optional, Union -from metagpt.config import CONFIG +from metagpt.config2 import config +from metagpt.configs.llm_config import LLMType from metagpt.logs import log_llm_stream, logger from metagpt.provider.azure_openai_api import AzureOpenAILLM from metagpt.provider.openai_api import OpenAILLM from metagpt.schema import Message -OriginalLLM = OpenAILLM if not CONFIG.openai_api_type else AzureOpenAILLM +OriginalLLM = OpenAILLM if config.llm.api_type == LLMType.OPENAI else AzureOpenAILLM class MockLLM(OriginalLLM): def __init__(self, allow_open_api_call): - super().__init__() + super().__init__(config.get_openai_llm()) self.allow_open_api_call = allow_open_api_call self.rsp_cache: dict = {} self.rsp_candidates: list[dict] = [] # a test can have multiple calls with the same llm, thus a list @@ -47,7 +48,9 @@ class MockLLM(OriginalLLM): if system_msgs: message = self._system_msgs(system_msgs) else: - message = [self._default_system_msg()] if self.use_system_prompt else [] + message = [self._default_system_msg()] + if not self.use_system_prompt: + message = [] if format_msgs: message.extend(format_msgs) message.append(self._user_msg(msg))