Luo2003/fea-bench

该数据集包含GitHub仓库的拉取请求（PR）相关信息，主要特征包括仓库名称（repo）、拉取请求编号（pull_number）、URL、实例ID（instance_id）、关联的问题编号（issue_numbers）、基础提交（base_commit）、补丁内容（patch）、测试补丁（test_patch）、创建时间（created_at）、README文件内容（readmes）、文件列表（files）、非Python补丁（non_py_patch）、新组件（new_components）、版本信息（version）、测试状态变化（FAIL_TO_PASS和PASS_TO_PASS）、环境设置提交（environment_setup_commit）以及问题描述（problem_statement）。数据集仅包含测试集（test），共有198个样本，总大小为16638032字节。

This dataset contains information related to GitHub pull requests (PRs), with features including repository name (repo), pull request number (pull_number), URL, instance ID (instance_id), associated issue numbers (issue_numbers), base commit (base_commit), patch content (patch), test patch (test_patch), creation time (created_at), README file content (readmes), file list (files), non-Python patches (non_py_patch), new components (new_components), version information (version), test status changes (FAIL_TO_PASS and PASS_TO_PASS), environment setup commit (environment_setup_commit), and problem statement (problem_statement). The dataset includes only a test split with 198 samples and a total size of 16638032 bytes.