How to get recommendations for a new user (who rates some items) using a tensor flow recommenders model


I am using the following Tensor flow tutorial from Google on building a collaborative filter based recommendation model.

I want to use the model in the following way

  1. A model is trained on the movie lens dataset using the tutorial.
  2. Model is used to power a move recommendation service
  3. A new user joins the service and gives ratings on 10 movies from the dataset.

How do I use the model to get predictions on the top 5 movie recommendations for this user based on the already trained model. I want to do this without training a new model that incorporates the user information.

Any pointers on how to do this will be greatly appreciated.

1 Like