ylongqi/openrec

How to get user/item embeddings ?

justram opened this issue · 2 comments

I've some problems to get the user/item embeddings(or representations) from this framework.
I'm very interested in your CML model. And I want to use it with movielens dataset, and evaluate with other recommendation algorithms like libMF. Is there an option or any plan to implement this feature?

Thanks for the suggestion! We will implement this feature within the next week.

This feature has been implemented and included in the master branch.

Use the function compute_module_outputs in the recommender to get user/item embeddings, e.g.,

model.compute_module_outputs('user_vec', batch_data, train=True)