help wanted
仓库指标
- Star
- (17,706 star)
- PR 合并指标
- (平均合并 6天 16小时) (30 天内合并 10 个 PR)
描述
I've noticed some differences in training time and performance between this tensorflow 2 implementation and the original version:
- lower ratings should be removed
user-to-movie interactions with rating <=3.5 are filtered out but they are used to generate test_data_te_ratings, val_data_te_ratings that are used to compute the metrics for the model. Could I ask why?
- huge spike in memory consumption, using dataset 20m when the training is started will gobble up ~25gb RAM while the initial version will only use ~4gb RAM. Is this is a bug?
Thank you