Hub documentation

Using Keras at Hugging Face

Hugging Face's logo
Join the Hugging Face community

and get access to the augmented documentation experience

to get started

Using Keras at Hugging Face

keras is an open-source machine learning library that uses a consistent and simple API to build models leveraging TensorFlow and its ecosystem.

Exploring Keras in the Hub

You can find over 200 keras models by filtering at the left of the models page.

All models on the Hub come up with useful feature:

  1. An automatically generated model card with a description, a plot of the model, and more.
  2. Metadata tags that help for discoverability and contain information such as license.
  3. If provided by the model owner, TensorBoard logs are hosted on the Keras repositories.

Using existing models

The huggingface_hub library is a lightweight Python client with utility functions to download models from the Hub.

pip install huggingface_hub["tensorflow"]

Once you have the library installed, you just need to use the from_pretrained_keras method. Read more about from_pretrained_keras here.

from huggingface_hub import from_pretrained_keras

model = from_pretrained_keras("keras-io/mobile-vit-xxs")
prediction = model.predict(image)
prediction = tf.squeeze(tf.round(prediction))
print(f'The image is a {classes[(np.argmax(prediction))]}!')

# The image is a sunflower!

If you want to see how to load a specific model, you can click Use in keras and you will be given a working snippet that you can load it!

Sharing your models

You can share your keras models by using the push_to_hub_keras method. This will generate a model card that includes your model’s hyperparameters, plot of your model and couple of sections related to the usage purpose of your model, model biases and limitations about putting the model in production. This saves the metrics of your model in a JSON file as well. Read more about push_to_hub_keras here.

from huggingface_hub import push_to_hub_keras

    tags = ["object-detection", "some_other_tag"],

The repository will host your TensorBoard traces like below.

Additional resources

< > Update on GitHub