Zhongwei Yang
Zhongwei Yang
> Hi, I calculated the "real hit rate", which is to predict the probability of all items appearing in the next-item position. The training hit rate@10 of the movielen-1m dataset...
https://github.com/pmixer/TiSASRec.pytorch/blob/059c6a6193ad6b93ac318f43ddbdcf5d428c9d42/model.py#L66 hi, pximer: Referring to your full implementation of transformer encoder, i got a RuntimeError in [ model.py:66 ] when i tried to modify args.num_heads==1 to args.num_heads==4. I modify the...
yeah, i find hidden size=50 does not support num_heads=4, so i adjust hidden_size = 64. for this problem, adding a FC layer behind the multihead layer might be one way,...
well-done job!!!