Tuan Truong

Looking at the way they do evaluation in the paper, I don't think it can be simply accommodated. After each step, they don't retrain the model but only fine-tune with...

@lthoang what do think about the suggestion by @darrylong? Do you think it’s a good idea to only keep generic examples here and move most of the examples into their...

@hieuddo I think it’s also a good idea to restructure the models folder. However, it’s a bit more complicated and involved than this one. How about we focus on the...

I think we can remove "query" from the response to minimize the bandwidth because "data" could be quite significant.

Also, the returned `user_result` contains mapped user indices. Shall we try to map them back to the original user IDs, using the mapping in `train_set`, for it to be consistent...

You can set `num_threads=1` to disable multiprocessing. We're also curious about the cause of this issue. Please give us more information regarding this issue. Thanks

In order for us to better localize the issue, you should provide us with all the code to reproduce the error.

One way to speed up evaluation is to do sampling for negative items. This usually makes sense when the set of items is huge and we can't afford to rank...

Most of the GPU-based models only utilize GPU during training but not evaluation. Their predictive functions tend to be dot-product between user and item embedding which is quite efficient with...

@lthoang @hieuddo @darrylong it seems that we can have an option for negative sampling during evaluation. Anyone wants to take a lead on this improvement? 😃