HumanEval
https://github.com/openai/human-eval
https://huggingface.co/datasets/openai/openai_humaneval
https://huggingface.co/datasets/openai_humaneval でもいける
task_id: identifier for the data sample
prompt: input for the model containing function header and docstrings
"def return1():\n"
canonical_solution: solution for the problem in the prompt
" return 1"
test: contains function to test generated code for correctness
entry_point: entry point for test
IMO：どうやって評価しているんだろう？（文字列一致でよい？）