familyld / deepcf Goto Github PK

The code stays in the initial result and can't run down.

test pinterest dataset ，raise cannot create a tensor proto whose content is larger than 2GB，do you have some idea on this error? thanks

According to the paper, MLP doesn't involve the concat operation and DMF does?

The user whose ID is equal to 21 doesn't exist in the train set while exists in the test set.

According the original paper,

The ml-1m dataset has been preprocessed by the provider. Each user has at least 20 ratings and each item has been rated by at least 5 users. We process the other three datasets in the same way.

while the user whose ID is equal to 20 just has 2 ratings in the train set.

Hi~
LastFM dataset in your repo has no timestamps, so how do you segment the last item of a user's interaction？

Hello there,
I tried to run the code with the AMusic dataset. The results of the CFNet are not correct with the current settings.

How to process the datasets? I mean, how to construct the negative, test and rating files from a dataset.