search

artem-oppermann / Deep-Autoencoders-For-Collaborative-Filtering

Using Deep Autoencoders for predictions of movie ratings.
Apache License 2.0
112 stars 39 forks source link

How to get train.dat and test.dat? #5

Closed tyutjxs closed 5 years ago

tyutjxs commented 5 years ago

Hi I try to study your demo, but my result is bad, so I guess there is a problem with my data. so How did you generate training sets and test sets from raw data?

tyutjxs commented 5 years ago

emmmm...

I have solved this problem, thank you very much for your code.

good luck！

kalashnlkov commented 5 years ago

@tyutjxs Could you please share your solution?

I have a overfitting result ( nice train_loss but increasing test_loss).

BTW, I randomize the raw data and split it with 9:1 to get train:test set.

Thank you in advance :)

LucRyan commented 5 years ago

@kalashnlkov Please check out this solved issue https://github.com/artem-oppermann/Deep-Autoencoders-For-Collaborative-Filtering/issues/2, I think you're having the same problem. You have to train/test with similar distribution, otherwise you'll see it overfitting at the beginning.

kalashnlkov commented 5 years ago

@LucRyan Thanks for your reply. I've tried the approach #2 mentioned before but still get overfitting. split.py It's this possible sharing your train/test.dat?

tyutjxs commented 5 years ago

@tyutjxs Could you please share your solution?

I have a overfitting result ( nice train_loss but increasing test_loss).

BTW, I randomize the raw data and split it with 9:1 to get train:test set.

Thank you in advance :)

maybe you can reference my responsity https://github.com/tyutjxs/AutoEncoder-for-recommend-system,good luck !