Skip to content

Latest commit

History

History
34 lines (23 loc) 路 1.3 KB

File metadata and controls

34 lines (23 loc) 路 1.3 KB

PCREQ Evaluation

This repository contains the evaluation results of PCREQ, assessing its capability to infer compatible requirements for TPL upgrades in Python projects.

馃搨 Directory Overview

  • Motivating-Study/
    Experimental results of our motivating study.

  • RQ1/
    Evaluates PCREQ's performance on compatible requirements inference of TPL upgrades in Python projects.

    • Uses the REQBench dataset.
  • RQ2/
    Compares PCREQ against PyEGo and ReadPyE on compatible requirements inference of TPL upgrades in Python projects.

    • Uses the REQBench-PyEGo and REQBench-ReadPyE datasets.
  • RQ3/
    Compares PCREQ against DeepSeek (V3, R1) and ChatGPT (GPT-4o) on compatible requirements inference of TPL upgrades in Python projects.

    • Uses the REQBench dataset.
  • RQ4/
    Measures PCREQ's runtime efficiency.

    • Computes the inference time of test cases.

馃搶 Notes

  • Each directory contains the corresponding experiment results and analysis.
  • For detailed evaluation metrics and results, please take a look at our paper.

License

PCREQ-evaluation is licensed under the Creative Commons Attribution 4.0 International License (CC BY 4.0). See LICENSE for details.