You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi - part of the challenge here is that we do not yet have publicly available large-scale datasets with long sequence lengths, therefore we only ran SL on our internal dataset. The current repo is intended for reproducing public experiments so far.
In case you have suggestions on datasets to use -- ideally with several thousands of (user, item) interactions on average -- we would love to help make related implementations available!
Great Work! However, there doesn't appear to be any implement of the Stochastic Length and Sampling Technique Γ(S, L) in the paper.
The text was updated successfully, but these errors were encountered: