Datasets:
File size: 3,726 Bytes
48a1dc7 042ec47 48a1dc7 042ec47 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 |
---
license: mit
task_categories:
- text-generation
- question-answering
language:
- en
tags:
- pytorch
- forum
- community
- deep-learning
- machine-learning
size_categories:
- 10K<n<100K
---
# PyTorch Forum Topics Dataset
This dataset contains topic metadata scraped from the [PyTorch Community Forum](https://discuss.pytorch.org). It includes comprehensive information about forum topics that can be used for various NLP tasks related to PyTorch and deep learning discussions.
## Dataset Description
- **Total Topics**: 73,380
- **Source**: PyTorch Community Forum (discuss.pytorch.org)
- **Format**: JSONL (JSON Lines)
- **Language**: English
- **License**: MIT
## Dataset Structure
Each record in the dataset contains the following fields:
- `id`: Unique topic identifier
- `title`: Topic title
- `slug`: URL-friendly version of the title
- `posts_count`: Number of posts in the topic
- `reply_count`: Number of replies
- `highest_post_number`: Highest post number in the topic
- `image_url`: Featured image URL (if any)
- `created_at`: Topic creation timestamp
- `last_posted_at`: Last activity timestamp
- `bumped`: Whether the topic was bumped
- `bumped_at`: Bump timestamp
- `archetype`: Topic type (e.g., "regular")
- `unseen`: Whether the topic is unseen
- `last_read_post_number`: Last read post number
- `unread`: Number of unread posts
- `new_posts`: Number of new posts
- `pinned`: Whether the topic is pinned
- `unpinned`: Unpinned timestamp
- `visible`: Whether the topic is visible
- `closed`: Whether the topic is closed
- `archived`: Whether the topic is archived
- `notification_level`: Notification level
- `bookmarked`: Whether the topic is bookmarked
- `liked`: Whether the topic is liked
- `views`: Number of views
- `like_count`: Number of likes
- `has_summary`: Whether the topic has a summary
- `last_poster_username`: Username of the last poster
- `category_id`: Category identifier
- `op_like_count`: Original post like count
- `pinned_globally`: Whether pinned globally
- `featured_link`: Featured link (if any)
- `posters`: List of users who posted in the topic
## Use Cases
This dataset can be used for:
1. **Topic Analysis**: Understanding discussion patterns in the PyTorch community
2. **Text Classification**: Categorizing forum topics by content
3. **Trend Analysis**: Analyzing the evolution of PyTorch-related discussions
4. **Community Insights**: Understanding user engagement and popular topics
5. **Question-Answer Extraction**: Identifying potential Q&A pairs for further processing
6. **Language Modeling**: Training models on PyTorch-specific terminology and discussions
## Data Collection
The data was collected using an automated scraper that:
- Respects the forum's rate limits
- Uses proper HTTP headers and session management
- Implements exponential backoff for failed requests
- Focuses on publicly available topic metadata
## Ethical Considerations
- This dataset contains only publicly available information from the PyTorch forum
- No private or sensitive information is included
- The data collection process was designed to be respectful of the forum's resources
- Users should respect the original forum's terms of service when using this data
## Citation
If you use this dataset in your research, please cite:
```bibtex
@dataset{pytorch_forum_topics_2025,
title={PyTorch Forum Topics Dataset},
author={Community Contributor},
year={2025},
url={https://huggingface.co/datasets/your-username/pytorch-forum-topics},
note={Scraped from discuss.pytorch.org}
}
```
## Updates
This dataset was created on July 24, 2025, and contains topics available at that time. The PyTorch forum is continuously active, so newer topics may not be included.
|