face-miner 1.0.0

Extractor for HuggingFace models

Hugging Face Extractor

A web-based tool to extract metadata for models from the Hugging Face Hub.

Features

• Extract information for a single model or all models.
• Fetches metadata, tags, datasets, CO2 emissions, commits, discussions, and file manifests.
• Multi-threaded extraction for performance.
• Download the extracted data as a ZIP file containing multiple CSVs.
• Ability to resume a full crawl if interrupted.

Installation

1. Clone the repository:

   git clone [https://github.com/yourusername/hf-extractor.git](https://github.com/yourusername/hf-extractor.git)
   cd hf-extractor
   

2. Install the required packages and make the project recognizable as a local package:

   pip install -e .
   

Usage

1. Run the Flask application from the project's root directory:

   python run.py
   

2. Open your web browser and navigate to http://127.0.0.1:5050.

3. Use the web interface to start the extraction process.