I am stucked in collaborative filtering google colab with softmax version

I am working on colab, the link is as the below.

https://colab.research.google.com/github/google/eng-edu/blob/main/ml/recommendation-systems/recommendation-systems.ipynb?utm_source=ss-recommendation-systems&utm_campaign=colab-external&utm_medium=referral&utm_content=recommendation-systems#scrollTo=zAAr73xno4uj

There are two versions each MF and Softmax on the site, and I don't know the Softmax part.

What I want to know is how can I get the recommendation result after finishing Softmax training? ( I know how to compute cosine similarity using movie vectors. )

The user vector ,I think, should be able to include the movie ID that the user has already seen and the corresponding movie category and year information.

Because I trained it like that.

How can I query if I have user information who watched movie with id each 1, 2, 3.

I would appreciate it if you could let me know the example code.