SWE-bench/SWE-bench_Verified
收藏Hugging Face2026-02-27 更新2025-07-05 收录
下载链接：
https://hf-mirror.com/datasets/SWE-bench/SWE-bench_Verified
下载链接
链接失效反馈官方服务：
资源简介：
SWE-bench Verified是一个包含500个样本的SWE-bench测试集的子集，这些样本已经过人工验证以确保质量。该数据集用于测试系统自动解决GitHub问题的能力，包含了来自流行Python仓库的500个测试性问题Issue-Pull Request对。数据集主要包含问题陈述和基准提交信息，用于在给定完整仓库和GitHub问题时进行问题解决任务的评估。
SWE-bench Verified is a subset of 500 samples from the SWE-bench test set, which have been human-validated for quality. It is used to test the ability of systems to automatically resolve GitHub issues, containing 500 test Issue-Pull Request pairs from popular Python repositories. The dataset primarily includes problem statements and base commits for evaluation in the issue resolution task given a full repository and GitHub issue.
提供机构：
SWE-bench
