search

hexiangnan / neural_collaborative_filtering

Neural Collaborative Filtering
Apache License 2.0
1.8k stars 655 forks source link

Bad performance on the whole test data with all movies #73

Closed YanXuHappygela closed 3 years ago

YanXuHappygela commented 3 years ago

The model is evaluated on sampled negatives in the example. I can generate similar result as reported in the paper on the sampled dataset. But when I try to evaluate it on the whole test data with all items as candidate, the HR@10 will drop to around 3%. Any advice on that?

YanXuHappygela commented 3 years ago

Turn out fine if I exclude the items in the training for each user. I'll close the issue.