What can you find in this repo?
huggingface_hub, a client library to download and publish on the Hugging Face Hub as well as extracting useful information from there.
api-inference-community, the Inference API for open source machine learning libraries.
widgets, the open sourced widgets that allow people to try out the models in the browser.
interfaces, Typescript definition files for the Hugging Face Hub.
docs, containing the official Hugging Face Hub documentation.
This library allows anyone to work with the Hub repositories: you can clone them, create them and upload your models to them. On top of this, the library also offers methods to access information from the Hub. For example, listing all models that meet specific criteria or get all the files from a specific repo. You can find the library implementation here.
We're partnering with cool open source ML libraries to provide free model hosting and versioning. You can find the existing integrations here.
The advantages are:
If you would like to integrate your library, feel free to open an issue to begin the discussion. We wrote a step-by-step guide with ❤️ showing how to do this integration.
In order to get a functional Inference API on the Hub for your models (and thus, cool working widgets!) check out this doc. There is a docker image for each library. Within the image, you can find the implementation for supported pipelines for the given library.
All our widgets are open-sourced. Feel free to propose and implement widgets. You can try all of them out here.
We'll implement a few tweaks to improve the UX for your models on the website – let's use Asteroid as an example.
Model authors add an
asteroid tag to their model card and they get the advantages of model versioning built-in
We add a custom "Use in Asteroid" button. When clicked, you get a library-specific code sample that you'll be able to specify. 🔥