We’re excited to announce improved support for Jupyter notebooks hosted on the Hugging Face Hub!
From serving as a vital learning resource to being a key tool used for model development, Jupyter notebooks have develop into a key component across many areas of machine learning. Notebooks’ interactive and visual nature enables you to get feedback quickly as you develop models, datasets, and demos. For a lot of, their first exposure to training machine learning models is via a Jupyter notebook, and lots of practitioners use notebooks as a critical tool for developing and communicating their work.
Hugging Face is a collaborative Machine Learning platform during which the community has shared over 150,000 models, 25,000 datasets, and 30,000 ML apps. The Hub has model and dataset versioning tools, including model cards and client-side libraries to automate the versioning process. Nonetheless, only including a model card with hyperparameters is just not enough to offer one of the best reproducibility; that is where notebooks may also help. Alongside these models, datasets, and demos, the Hub hosts over 7,000 notebooks. These notebooks often document the event strategy of a model or a dataset and may provide guidance and tutorials showing how others can use these resources. We’re subsequently enthusiastic about our improved support for notebook hosting on the Hub.
What have we modified?
Under the hood, Jupyter notebook files (normally shared with an ipynb extension) are JSON files. While viewing these files directly is feasible, it isn’t a format intended to be read by humans. We’ve now added rendering support for notebooks hosted on the Hub. Which means that notebooks will now be displayed in a human-readable format.
Why are we excited to host more notebooks on the Hub?
- Notebooks help document how people can use your models and datasets; sharing notebooks in the identical place as your models and datasets makes it easier for others to make use of the resources you will have created and shared on the Hub.
- Many individuals use the Hub to develop a Machine Learning portfolio. You may now complement this portfolio with Jupyter Notebooks too.
- Support for one-click direct opening notebooks hosted on the Hub in Google Colab, making notebooks on the Hub an excellent more powerful experience. Look out for future announcements!
