Collaborative Filtering use_nn=True example?

I have been spending my morning trying to get a collaborative filtering model that has use_nn=True to recreate the steps notebook 8 goes through with the EmbeddingDotBias class. I am not sure why it is not included in the chapter, but there is no explanation of how to actually use the EmbeddingNN version of the model. I would like to be able to find the movie_bias list and calculate movie similarity for starters. Additionally I would like to be able to pass in a user ID, movie title (or ID) and find what rating my model predicts. I have not figured out how to do this yet.

Does anyone have a good example of how to apply a EmbeddingNN model?

Thanks in advance!