fastai adds official Hugging Face Hub integration

An official integration between the fastai library and the Hugging Face Hub has been announced. Now fastai users can upload trained models to the Hub with just a single line of Python code to share them with the global community.

Key features:

• Model sharing and utilization: You can publish models on the Hub to maximize exposure, and bring in other people's fastai models to use for Transfer Learning.
• Version control and reproducibility: Through the Hub's Git-LFS-based version control and Model Card features, you can manage a model's history and ensure reproducibility.
• Easy installation: You can instantly install the related dependencies with the pip install huggingface_hub["fastai"] command.

By applying the library=fastai filter on the Hub's model page, you can easily explore various models trained with fastai.