Issue with SAKT model

[ghodai]

I noticed a weird way of invoking the TranformerLayers in SAKT model as per the below code snippet, you are only getting the output of the last layer applied on the seq_data input, you should either combine the pool or add y as input per each call!

y = seq_data for block in self.attn_blocks: y = block(mask=1, query=q_data, key=seq_data, values=seq_data)

[THUwangcy]

Thanks for pointing out this issue! It is indeed a mistake and has been fixed in the latest commit. Meanwhile, considering that the original paper only uses a single transformer layer, our previous implementation just works the same. So the experimental results are still reliable.