rIsHu009/github-issues-updated

1---2annotations_creators:3- unknown4language:5- en6language_creators:7- found8license: mit9multilinguality: monolingual10pretty_name: GitHub Issues Dataset - HuggingFace/datasets11size_categories:12- 10K<n<100K13source_datasets: []14task_categories:15- text-classification16task_ids:17- topic-classification18- extractive-qa19configs:20- config_name: default21  data_files:22  - split: train23    path: data/train-*24dataset_info:25  features:26  - name: id27    dtype: int6428  - name: number29    dtype: int6430  - name: title31    dtype: string32  - name: state33    dtype: string34  - name: body35    dtype: string36  - name: is_pull_request37    dtype: bool38  - name: created_at39    dtype: string40  - name: updated_at41    dtype: string42  - name: closed_at43    dtype: string44  - name: user_login45    dtype: string46  - name: author_association47    dtype: string48  - name: pr_url49    dtype: string50  - name: pr_merged_at51    dtype: string52  - name: comments_count53    dtype: int6454  - name: reactions_total55    dtype: int6456  - name: reactions_plus157    dtype: int6458  - name: reactions_heart59    dtype: int6460  - name: draft61    dtype: bool62  - name: locked63    dtype: bool64  - name: labels65    sequence: string66  - name: html_url67    dtype: string68  - name: is_pr_url69    dtype: bool70  - name: comments71    sequence: string72  splits:73  - name: train74    num_bytes: 2858558575    num_examples: 747476  download_size: 1074993577  dataset_size: 2858558578---79 80# 📊 GitHub Issues Dataset (HuggingFace/datasets Repository)81 82This dataset contains structured GitHub issues scraped from the [`huggingface/datasets`](https://api.github.com/repos/huggingface/datasets/issues?page=1&per_page=1) repository. It is intended for **NLP tasks**, **topic modeling**, **issue classification**, and **software engineering research**.83 84---85 86## 📌 Dataset Summary87 88- **Repository Source**: `huggingface/datasets`89- **Scraped via**: GitHub REST API v390- **Total Issues**: ~7,46591- **Collected On**: June 25, 202592- **Format**: `JSONL` → loaded via Arrow for Hugging Face93- **Language**: English94 95---96 97## 📂 Fields Included98 99Each entry contains:100 101- `title` — Issue title  102- `body` — Issue body text  103- `state` — Open or closed  104- `created_at`, `updated_at`, `closed_at` — Timestamps  105- `is_pull_request` — Boolean flag  106- `user.login` — Author  107- `comments` — Count of comments  108- `html_url` — URL to the issue  109- `pull_request.*` — Nested fields when issue is a PR110 111---112 113## ⚙️ How It Was Created114 115Issues were scraped using Python’s `requests` library with pagination (`per_page=100`). The dataset was flattened to remove deep nesting and improve usability with machine learning models.116 117All entries were saved to `.jsonl`, loaded into a Hugging Face DatasetDict using `load_dataset`, and then pushed to the 🤗 Hub.118 119---120 121## 💡 Example Usage122 123```python124from datasets import load_dataset125 126# Load the dataset from the Hugging Face Hub127ds = load_dataset("rIsHu009/github-issues-updated", split="train")128 129# Access a sample record130print(ds[0])131 132