apunkt/MetaGPT
1
0
Fork 0
mirror of https://github.com/FoundationAgents/MetaGPT.git synced 2026-05-12 01:02:37 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

add context and config2

Browse source
This commit is contained in:
geekan 2024-01-04 21:16:23 +08:00
parent 42bb40a0f6
commit e5d11a046c
76 changed files with 922 additions and 495 deletions
Download patch file Download diff file Expand all files Collapse all files

18
metagpt/learn/text_to_image.py
View file

@ -8,33 +8,37 @@
"""
import base64
from metagpt.config import CONFIG
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", model_url="", config: Config = None):
"""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:
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 oai_llm := config.get_openai_llm():
binary_data = await oas3_openai_text_to_image(text, size_type, LLM(oai_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 ""
assert config.s3, "S3 config is required."
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 ""

4
metagpt/learn/text_to_speech.py
View file

@ -48,7 +48,7 @@ async def text_to_speech(
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 ""
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
@ -60,7 +60,7 @@ async def text_to_speech(
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 ""
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