training not converge on Amazon Beauty dataset according to the paper's hyperparameter setting

[lipingcoding]

[pmixer]

@lipingcoding thx for reporting the issue, I also observed some problems and still wondering what's wrong with pytorch version compared with tf implementation. Sorry to say but I still haven't figured it out yet. The only thing I can be sure currently is that original paper's hyperparameter setting could be be directly used for this codebase, as I fixed some leaky attention issue by using PyTorch's MHA, the parameter initialization issue still need to be elaborated but I haven't done it yet.

[pmixer]

btw @lipingcoding could u pls provide bit more information like log etc. to describe the observation that the model do not converge? Like uncomment

SASRec.pytorch/main.py

Line 93 in 30c43cf

# print("loss in epoch {} iteration {}: {}".format(epoch, step, loss.item())) # expected 0.4~0.6 after init few epochs

would print loss for iteration which should be informative.

[pmixer]

@lipingcoding try the newly updated code?