metadata

title: code_eval_outputs
datasets:
  - giulio98/xlcost-single-prompt
tags:
  - evaluate
  - metric
description:
  - >-
    This metric implements the evaluation harness for the HumanEval problem
    solving dataset described in the paper "Evaluating Large Language Models
    Trained on Code" (https://arxiv.org/abs/2107.03374). But instead of
    evaluating the assertions it compares the output of the generated codes with
    the expected output
sdk: gradio
sdk_version: 3.0.2
app_file: app.py
pinned: false

Metric Card for code_eval_outputs

Metric Description

This metric is based on code_eval but instead of evaluating the functional correctness of the generated program through assertions in the form of unit tests, it compares the output of the generated program with the expected output, for more details please refer to code_eval.

How to Use

The Code Eval metric calculates how good are predictions given a set of references. Its arguments are:

predictions: a list of candidates to evaluate. Each candidate should be a list of strings with several code candidates to solve the problem.

references: a list with a function call for each prediction. Each function call should output a string in stdout.

output: a list of the expected output for each prediction.

k: number of code candidates to consider in the evaluation. The default value is [1, 10, 100].

num_workers: the number of workers used to evaluate the candidate programs (The default value is 4).

timeout: The maximum time taken to produce a prediction before it is considered a "timeout". The default value is 30.0 (i.e. 30 seconds).

from evaluate import load
code_eval_outputs = load("giulio98/code_eval_outputs")
references = ["if __name__ == \"__main__\":\n    print(add(2, 3))"]
expected_outputs = ["5"]
candidates = [["def add(a,b):\n    return a*b", "def add(a, b):\n    return a+b"]]
pass_at_k, results = code_eval_outputs.compute(references=references, predictions=candidates, output=expected_outputs, k=[1, 2])
print(pass_at_k)
print(results)

Output:

{'pass@1': 0.5, 'pass@2': 1.0}
defaultdict(list,
            {0: [(0,
               {'task_id': 0,
                'passed': False,
                'result': 'not passed',
                'completion_id': 0}),
              (1,
               {'task_id': 0,
                'passed': True,
                'result': 'passed',
                'completion_id': 1})]})

N.B. This metric exists to run untrusted model-generated code. Users are strongly encouraged not to do so outside of a robust security sandbox. Before running this metric and once you've taken the necessary precautions, you will need to set the HF_ALLOW_CODE_EVAL environment variable. Use it at your own risk:

import os
os.environ["HF_ALLOW_CODE_EVAL"] = "1"

Output Values

The Code Eval metric outputs two things:

pass_at_k: a dictionary with the pass rates for each k value defined in the arguments.

results: a dictionary with granular results of each unit test.

Limitations and Bias

Refer to code_eval

Citation

@misc{chen2021evaluating,
      title={Evaluating Large Language Models Trained on Code},
      author={Mark Chen and Jerry Tworek and Heewoo Jun and Qiming Yuan \
and Henrique Ponde de Oliveira Pinto and Jared Kaplan and Harri Edwards \
and Yuri Burda and Nicholas Joseph and Greg Brockman and Alex Ray \
and Raul Puri and Gretchen Krueger and Michael Petrov and Heidy Khlaaf \
and Girish Sastry and Pamela Mishkin and Brooke Chan and Scott Gray \
and Nick Ryder and Mikhail Pavlov and Alethea Power and Lukasz Kaiser \
and Mohammad Bavarian and Clemens Winter and Philippe Tillet \
and Felipe Petroski Such and Dave Cummings and Matthias Plappert \
and Fotios Chantzis and Elizabeth Barnes and Ariel Herbert-Voss \
and William Hebgen Guss and Alex Nichol and Alex Paino and Nikolas Tezak \
and Jie Tang and Igor Babuschkin and Suchir Balaji and Shantanu Jain \
and William Saunders and Christopher Hesse and Andrew N. Carr \
and Jan Leike and Josh Achiam and Vedant Misra and Evan Morikawa \
and Alec Radford and Matthew Knight and Miles Brundage and Mira Murati \
and Katie Mayer and Peter Welinder and Bob McGrew and Dario Amodei \
and Sam McCandlish and Ilya Sutskever and Wojciech Zaremba},
      year={2021},
      eprint={2107.03374},
      archivePrefix={arXiv},
      primaryClass={cs.LG}
}

Further References

Refer to code_eval