kyung-wook

Thanks reditya ! I wonder, if I don't have negative samples, I can't train this model?

Thanks a lot reditya! So, that means 'negative samples' are not literally negative, but just sample from 'unobserved' items. And when I run this code using my own data, I...

Thanks so much! @reditya

In 'neural collaborative filtering' paper (https://www.comp.nus.edu.sg/~xiangnan/papers/ncf.pdf), there are graph about number of negatives. 4 negatives instances showed good result and as my experience, when you select 10 or 20 neg,...

@fjz15056311771 good luck :)

@fjz15056311771 Sorry for late reply Right. Existing matrix factorization use only confirmed relationship between user and item. Yet, it can cause limitation of matrix factorization. In implicit data, we only...

@fjz15056311771 hi again brief explanation about data is at bottom of main page. and I think data structure like this -train data- user_id, work_id, rating, timestamp (but here, this model...

I had a similar issue. How about check getting updated 'page_source' ? I fixed little after check that.

I fixed it little but its not perfect I think Instagram changed something.

I changed it to parse the html (maybe '_driver.page_source' in code) everytime after scroll down. You find where the code stucks? my answer won't be the answer because i used...