I have trained a model using Auto-encoder on movielens dataset. Below is how i trained the model.

r = model.fit_generator(
  generator(A, mask),
  validation_data=test_generator(A_copy, mask_copy, A_test_copy, mask_test_copy),
  steps_per_epoch=A.shape[0] // batch_size + 1,
  validation_steps=A_test.shape[0] // batch_size + 1,

It is giving good results but now i am confused how should i get the top 5 recommendation on user input.

Just wanted to print the result on console. Can anyone help me please?


That is not what an auto-encoder is doing. An auto-encoder gives you a compressed representation of the input. It is trained by mapping the input data to itself, with the compressed form in between.

To predict recommendations, you need to train your input data on existing user recommendations.

  • $\begingroup$ Hey, thank you for responding. Can you share any useful links? or examples that is doing this predict recommendation part? $\endgroup$
    – Debugger
    Feb 20 '20 at 18:22
  • $\begingroup$ @Debugger you can try to use recommendation system algorithm, for example matrix factorization $\endgroup$
    – malioboro
    Feb 20 '20 at 22:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.