evalplus 0.1.2

"EvalPlus for rigourous evaluation of LLM-synthesized code"

EvalPlus(📖) => 📚

🔥Quick Start • 📜Papers • 🔨Useful tools • 👷Development • 🙏Acknowledgement

Warning

🚨 Evaluating LLM-generated code over datasets with "3 test-cases" is **NOT** enough! 🚨

To address this, we started the EvalPlus project -- a rigourous evaluation framework for LLM4Code that:

• ✨ improves code benchmarks by adding up to thousands of new tests! (81x new tests for HumanEval!)
• ✨ crafts a set utility tools to sanitize, visualize and inspect LLM-generated code and evaluation results!
• ✨ accelerates LLM4Code research by open-sourcing LLM-generated samples for 14+ models -- no need to re-run the expensive benchmarks!

🔥 Quick Start

To get started, please first setup the environment:

pip install evalplus --upgrade

...Or you can try out the latest developing version:

pip install "git+https://github.com/evalplus/evalplus.git" --upgrade

🤔 Want to use local GitHub repo? :: click to expand ::

git clone https://github.com/evalplus/evalplus.git
cd evalplus
export PYTHONPATH=$PYTHONPATH:$(pwd)
pip install -r requirements.txt

HumanEval+

The usage is just like the original HumanEval where you just need to implement the generate_one_completion function!

from evalplus.data import get_human_eval_plus, write_jsonl

problems = get_human_eval_plus()

num_samples_per_task = 200
samples = [
    dict(task_id=task_id, completion=generate_one_completion(problems[task_id]["prompt"]))
    for task_id in problems
    for _ in range(num_samples_per_task)
]
write_jsonl("samples.jsonl", samples)

🤔 What is in a `problem`? :: click to expand ::

• task_id is the identifier string for the task
• entry_point is name of the function
• prompt is the function signature with docstring

• canonical_solution is the ground-truth implementation (re-implemented to fix bugs in HumanEval)
• base_input is the test inputs in original HumanEval
• plus_input is the test inputs brought by EvalPlus

To evaluate the samples:

You are strongly recommended to use a sandbox such as docker:

docker run -v $(pwd):/app ganler/evalplus:v0.1.1 --dataset humaneval --samples samples.jsonl

...Or if you want to try it locally regardless of the risks ⚠️:

evalplus.evaluate --dataset humaneval --samples samples.jsonl

🤔 Want to use local GitHub repo? :: click to expand ::

python evalplus/evaluate.py --dataset humaneval --samples samples.jsonl

⌨️ More command-line flags :: click to expand ::

• --parallel: by default half of the cores
• --base-only (store_ture): only run base HumanEval tests
• --i-just-wanna-run: force a re-run

The output should be like (below is GPT-4 greedy decoding example):

Computing expected output...
Expected outputs computed in 15.18s
Reading samples...
164it [00:04, 37.79it/s]
Evaluating samples...
100%|██████████████████████████████████████████| 164/164 [00:03<00:00, 44.75it/s]
Base
{'pass@1': 0.8841463414634146}
Base + Extra
{'pass@1': 0.75}

• Base is the pass@k for the original HumanEval
• Base + Extra is the pass@k for the our HumanEval+ (with extra tests)
• The "k" includes [1, 10, 100] where k values <= the sample size will be used
• A cache file named like samples_eval_results.jsonl will be cached. Remove it to re-run the evaluation

MBPP+ (TBD)

📜 Papers

Read our paper for more detailed findings!

@article{evalplus,
  title={Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generation},
  author={Jiawei Liu and Chunqiu Steven Xia and Yuyao Wang and Lingming Zhang},
  journal={arXiv preprint arXiv:2305.01210},
  year={2023},
}

🔨 Useful tools

To use these tools, please first install the repository from GitHub:

git clone https://github.com/evalplus/evalplus.git
cd evalplus
pip install -r requirements-tools.txt

Syntax checker for LLM-generated code

Check LLM-produced code and answer the following questions:

1. Is the generation entirely done for all samples / all problems in the dataset?
2. Are LLM-generated code compilable? (if no, something could be wrong and you'd better check)

python tools/checker.py --folder /path/to/[model]-[??]b_temp_[??] --dataset humaneval

Post code sanitizer

LLM-generated code may contain some syntax errors. But some of them can be easily fixable by doing simple post-processing. This tool will make the LLM-generated code more clean/compilable by doing certain post-processing such as trimming with more magical EOFs and some garbage non-code tokens.

python tools/sanitize.py --eof --folder /path/to/vicuna-[??]b_temp_[??]
# Sanitized code will be produced to `/path/to/vicuna-[??]b_temp_[??]-sanitized`

Render pass@k results to rich and LaTeX tables

python tools/render.py --type /path/to/[model]-[??]b # NOTE: no `_temp_[??]`

Perform test input generation from scratch (TBD)

👷 Development

Before you start:

pip install pre-commit
pre-commit install
export PYTHONPATH=$PYTHONPATH:$(pwd)

Name convention

• evalplus is the package name.
• ${DATASET}_plus is the name of dataset applied with evalplus.

🙏 Acknowledgement

• HumanEval