fastai
Using fastai at BOINC AI
fastai
is an open-source Deep Learning library that leverages PyTorch and Python to provide high-level components to train fast and accurate neural networks with state-of-the-art outputs on text, vision, and tabular data.
Exploring fastai in the Hub
You can find fastai
models by filtering at the left of the models page.
All models on the Hub come up with the following features:
An automatically generated model card with a brief description and metadata tags that help for discoverability.
An interactive widget you can use to play out with the model directly in the browser (for Image Classification)
An Inference API that allows to make inference requests (for Image Classification).
Using existing models
The huggingface_hub
library is a lightweight Python client with utlity functions to download models from the Hub.
Copied
Once you have the library installed, you just need to use the from_pretrained_fastai
method. This method not only loads the model, but also validates the fastai
version when the model was saved, which is important for reproducibility.
Copied
If you want to see how to load a specific model, you can click Use in fastai
and you will be given a working snippet that you can load it!
Sharing your models
You can share your fastai
models by using the push_to_hub_fastai
method.
Copied
Additional resources
fastai course.
fastai website.
Integration with Hub docs.
Integration with Hub announcement.
Last updated