apunkt/MetaGPT
1
0
Fork 0
mirror of https://github.com/FoundationAgents/MetaGPT.git synced 2026-06-05 14:55:18 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

Update GitNore

Browse source
This commit is contained in:
didi 2024-07-27 01:57:06 +08:00
parent 772d2aea56
commit eac4b6c3e6
30 changed files with 939 additions and 9 deletions
Download patch file Download diff file Expand all files Collapse all files

18
he_test.py
View file

@ -6,9 +6,14 @@ from examples.ags.benchmark.humaneval import sample_generate, samples_generate,
from examples.ags.w_action_node.utils import jsonl_ranker, llm_extract_test_case
from examples.ags.w_action_node.graph import HumanEvalGraph
# 132 141 136 80 73
# asyncio.run(sample_generate('HumanEval/118',result_path="llm_based_8.jsonl",mode="llm"))
asyncio.run(samples_generate(mode='llm',result_path="llm_based_100.jsonl"))
# jsonl_ranker("samples.jsonl", "samples.jsonl")
# asyncio.run(sample_generate('HumanEval/140',result_path="llm_based_1000.jsonl",mode="llm"))
# asyncio.run(sample_generate('HumanEval/140',result_path="llm_based_1000.jsonl",mode="llm"))
# asyncio.run(sample_generate('HumanEval/140',result_path="llm_based_1000.jsonl",mode="llm"))
# asyncio.run(sample_generate('HumanEval/67',result_path="llm_based_1000.jsonl",mode="llm"))
# asyncio.run(sample_generate('HumanEval/108',result_path="llm_based_1000.jsonl",mode="llm"))
# asyncio.run(sample_generate('HumanEval/110',result_path="llm_based_1000.jsonl",mode="llm"))
# asyncio.run(samples_generate(mode='alpha',result_path="alpha_based_100.jsonl"))
# jsonl_ranker("llm_based_137.jsonl", "llm_based_137.jsonl")
# result_path = "ags_based_6.jsonl"
# if automatic_evalplus(result_path):
@ -41,6 +46,7 @@ asyncio.run(samples_generate(mode='llm',result_path="llm_based_100.jsonl"))
# [72, 80, 82, 87, 90, 95, 107, 109, 112, 124, 126, 127, 128, 132, 134, 136, 137, 138, 148, 154, 155]
# case_prompt= get_human_eval_plus()["HumanEval/136"]['prompt']
# solver = HumanEvalGraph(name="solver", llm=LLM(), criteria='correctness, efficiency, readability', vote_count=1)
# result = asyncio.run(solver.alpha_codium(problem_id="HumanEval/136", problem=case_prompt, ensemble_count=1))
# TODO 代码问题，改动了一个地方导致Solution 没有了
case_prompt= get_human_eval_plus()["HumanEval/76"]['prompt']
solver = HumanEvalGraph(name="solver", llm=LLM(), criteria='correctness, efficiency, readability', vote_count=1)
result = asyncio.run(solver.alpha_codium(problem_id="HumanEval/136", problem=case_prompt, ensemble_count=1))