ML Community Day is November 9! Join us for updates from TensorFlow, JAX, and more Learn more

TensorFlow Hub is a repository of trained machine learning models.

  !pip install --upgrade tensorflow_hub

  import tensorflow_hub as hub

  model = hub.KerasLayer("https://tfhub.dev/google/nnlm-en-dim128/2")
  embeddings = model(["The rain in Spain.", "falls",
                      "mainly", "In the plain!"])

  print(embeddings.shape)  #(4,128)
TensorFlow Hub is a repository of trained machine learning models ready for fine-tuning and deployable anywhere. Reuse trained models like BERT and Faster R-CNN with just a few lines of code.



Models

Find trained models from the TensorFlow community on TFHub.dev
Check out BERT for NLP tasks including text classification and question answering.
Use the Faster R-CNN Inception ResNet V2 640x640 model for detecting objects in images.
Transfer the style of one image to another using the image style transfer model.
Use this TFLite model to classify photos of food on a mobile device.



News & announcements

Check out our blog for more announcements and view the latest #TFHub updates on Twitter
Learn how you can use TensorFlow Hub to build ML solutions with real world impact.
To explore ML solutions for your mobile and web apps including TensorFlow Hub, visit the Google on-device machine learning page.
TensorFlow Hub makes BERT simple to use with new preprocessing models.
Learn how to use the SPICE model to automatically transcribe sheet music from live audio.