farid678/github-issues

1---2dataset_info:3  features:4    - name: id5      dtype: string6 7    - name: number8      dtype: int649 10    - name: title11      dtype: string12 13    - name: body14      dtype: string15 16    - name: state17      dtype: string18 19    - name: createdAt20      dtype: timestamp[s]21 22    - name: updatedAt23      dtype: timestamp[s]24 25    - name: closedAt26      dtype: timestamp[s]27 28    - name: url29      dtype: string30 31    - name: author32      struct:33        - name: login34          dtype: string35 36    - name: labels37      struct:38        - name: nodes39          list:40            - name: name41              dtype: string42 43    - name: comments44      list: string45 46  splits:47    - name: train48      num_bytes: 1246761549      num_examples: 334350 51  configs:52    - config_name: default53      data_files:54        - split: train55          path: data/train-*56---57 58# GitHub Issues Dataset59 60## Dataset Description61 62The **GitHub Issues Dataset** is a collection of publicly63available GitHub issues collected from the64[Hugging Face `datasets` repository](https://github.com/huggingface/datasets).65 66Each example represents a GitHub issue and contains information67such as the issue identifier, issue number, title, description,68state, timestamps, author, labels, original URL, and comments.69 70The dataset was collected using the GitHub GraphQL API with71cursor-based pagination.72 73The dataset is intended for research and experimentation in74Natural Language Processing (NLP), software engineering,75information retrieval, semantic search, issue analysis,76question answering, and Large Language Models (LLMs).77 78---79 80## Dataset Details81 82- **Source:** Hugging Face `datasets` repository83- **Source platform:** GitHub84- **Domain:** Software development85- **Language:** Primarily English86- **Data type:** Text and structured metadata87- **Number of examples:** 3,34388- **Format:** Parquet / Arrow89- **Split:** `train`90 91---92 93## Source94 95The original data was collected from the public GitHub repository:96 97https://github.com/huggingface/datasets98 99The original issues are available through the GitHub Issues100section of the repository.101 102---103 104## Data Collection105 106The dataset was collected through the GitHub GraphQL API.107 108Cursor-based pagination was used to retrieve issues from the109repository.110 111For each issue, the following information was collected:112 113- Issue ID114- Issue number115- Title116- Body117- State118- Creation timestamp119- Update timestamp120- Closing timestamp121- Original GitHub URL122- Author login123- Labels124- Comments125 126Only publicly accessible information was collected.127 128---129 130## Dataset Structure131 132Each example contains the following fields:133 134| Field | Type | Description |135|---|---|---|136| `id` | string | Unique GitHub identifier for the issue |137| `number` | int64 | GitHub issue number |138| `title` | string | Title of the issue |139| `body` | string | Main description and content of the issue |140| `state` | string | Current issue state, such as `OPEN` or `CLOSED` |141| `createdAt` | timestamp | Date and time when the issue was created |142| `updatedAt` | timestamp | Date and time when the issue was last updated |143| `closedAt` | timestamp | Date and time when the issue was closed, if applicable |144| `url` | string | Original GitHub issue URL |145| `author` | struct | Information about the issue author |146| `labels` | struct | Labels assigned to the issue |147| `comments` | list[string] | Comments associated with the issue |148 149---150 151## Example152 153A simplified example of a dataset record:154 155```json156{157  "id": "I_kwDODunzps8AAAABMjWDaQ",158  "number": 8475,159  "title": "Dataset.select raises a raw OverflowError on in-range negative indices",160  "body": "Describe the bug ...",161  "state": "OPEN",162  "createdAt": "2026-08-13T03:02:44",163  "updatedAt": "2026-08-13T03:02:44",164  "closedAt": null,165  "url": "https://github.com/huggingface/datasets/issues/8475",166  "author": {167    "login": "codeAnqiang-ma"168  },169  "labels": {170    "nodes": []171  },172  "comments": []173}