noamaanMulla-03/datasets-issues

HuggingFace Datasets Library GitHub Issues Dataset

Dataset Description

This dataset contains all issues and pull requests (open and closed) from the huggingface/datasets GitHub repository, along with their comment threads. It was collected on July 20, 2026 via the GitHub REST API and follows the workflow described in the Hugging Face NLP course — Creating your own dataset.

• Repository: huggingface/datasets
• Total rows: 8,084 (issues + pull requests)
• Date range: April 14, 2020 → July 19, 2026
• Format: Parquet (single shard, train split)

Dataset Structure

Splits

Split	Num examples
train	8,084

Composition

Type	Count
Issues	3,293
Pull requests	4,791
Open	1,162
Closed	6,922
With labels	1,791
With comments	6,108
Total comments	20,791

Key Columns

Column	Type	Description
number	int64	GitHub issue/PR number (unique, no duplicates)
title	string	Issue/PR title
body	string	Issue/PR body (Markdown; 417 nulls — normal)
state	string	open or closed
is_pull_request	bool	True if this row is a PR, False if it's an issue
labels	list	List of label objects (name, color, id, ...) — 1,791 rows have labels
comments	int64	Number of comments on the issue
comments_text	list[str]	The body text of each comment (fetched separately via the comments API)
user	struct	The author (login, id, avatar_url, type, ...)
created_at	timestamp	When the issue/PR was created
updated_at	timestamp	When it was last updated
closed_at	timestamp	When it was closed (null for 1,162 open issues)
reactions	struct	Reaction counts (+1, -1, laugh, heart, rocket, eyes, ...)
pull_request	struct	PR metadata (url, diff_url, patch_url, merged_at) — null for issues
milestone	struct	Milestone info (url, title, creator, state, ...)
assignees	list	Assigned users

Top Labels

Label	Count
bug	710
enhancement	516
dataset request	163
dataset-viewer	90
dataset bug	74
dataset contribution	74
documentation	67
good first issue	53
duplicate	30
generic discussion	27
question	24
vision	24
streaming	22
speech	20
maintenance	19

Collection Methodology

The dataset was collected using a custom Python script (prepare_dataset.py) that:

1. Fetched all issues + PRs using the GitHub REST API with since-based pagination (sorted by updated_at ascending) to bypass GitHub's 10,000-record page-number cap. Issues were written incrementally to a JSONL file for crash recovery.
2. Added an is_pull_request flag based on whether the pull_request field is present.
3. Pushed the issues-only dataset to the Hub (as a safety checkpoint before the long comment fetch).
4. Fetched comment bodies for every issue with comments > 0 (1,975 zero-comment issues were skipped as an optimization). Comments were fetched with per_page=100 to minimize pagination requests.
5. Pushed the final dataset (issues + comments) to the Hub.

GitHub's authenticated rate limit (5,000 requests/hour) was respected throughout — the script sleeps until the rate-limit window resets when the budget is exhausted, with a one-hour fallback sleep if the reset header is missing. Transient network errors (ConnectionError, Timeout) and HTTP 5xx responses are retried with exponential backoff.

Known Limitations

• 262 missing issue numbers in the range 1–8,346: these correspond to deleted issues/PRs on GitHub that cannot be retrieved via the API. This is expected and unavoidable.
• 6 minor mismatches between the comments count field and the length of comments_text: these occur when comments were added or deleted between the issue fetch and the comment fetch. The comments field reflects the count at fetch time; comments_text reflects the actual comments retrieved.
• body has 417 nulls — some issues/PRs simply have no body text.
• The dataset includes both issues and pull requests (GitHub's issues endpoint returns both). Use the is_pull_request column to filter.

Potential Use Cases

• Issue classification: Train a model to categorize issues (bug, enhancement, question, dataset request) based on title + body.
• Bug triage: Predict which label an issue should have, or whether it's likely to be closed.
• Duplicate detection: Find semantically similar issues using embeddings of title + body.
• Pull request summarization: Summarize long PR bodies or comment threads.
• Software engineering research: Analyze issue lifecycle, response times, contributor activity, label distributions, etc.
• LLM fine-tuning: Build instruction-tuning datasets for "classify this issue" or "summarize this thread" tasks.

Licensing

The dataset is derived from public GitHub issue data from the huggingface/datasets repository, which is licensed under Apache 2.0. The dataset is provided under the same license.

Citation

If you use this dataset, please cite:

@misc{datasets-issues-dataset,
  author       = {Noamaan Mulla},
  title        = {HuggingFace Datasets Library GitHub Issues Dataset},
  year         = {2026},
  url          = {https://huggingface.co/datasets/noamaanMulla-03/datasets-issues},
  note         = {Collected via the GitHub REST API on July 20, 2026}
}

Also cite the original repository:

@misc{lhoest2021datasets,
  author       = {Quentin Lhoest and Albert Villanova del Moral and Patrick von Platen and Thomas Wolf and Leandro von Werra and Jannis Bulian and Albert Webson and Mario Šimko and Moustafa AlAkhras and Sharif Rahman and Clément Delangue and Lewis Tunstall and Sylvain Gugger and Pedro Cisneros and Anthony Moi and Jonathan Tow and Sara Charte and Marc Sun and Sagnik Ray and Teven Le Scao and Nicholas Broad and Alexander M. Rush and Sasha Luccioni and Maria Lomeli and Patrick Altmeyer and Vilson Rodrigues and Fabio Ciccone and Andrea Sottana},
  title        = {Datasets: A Community Library for Natural Language Processing Datasets},
  year         = {2021},
  publisher    = {GitHub},
  howpublished = {\url{https://github.com/huggingface/datasets}}
}