zhaijy
zhaijy
您好，能分享一下您训练ferplus数据集的代码吗？非常感谢！[email protected]
78行后，模型的梯度不更新了，但是，执行79行之后，再查看模型梯度，会发现token embedding的required_grad==True，我看了下源码，resize_token_embedding新建立了一个nn.embedding，所以后面更新了user, item和token三个呀
I conducted experiments by incorporating IDs, and the results are as follows. What confuses me even more is that I did not pre-train either of the methods; I only utilized...
I'm sorry, I misspoke earlier. UniSRec actually performed worse after pretraining. 