Igor L.R. Azevedo

Thanks @you-n-g for the replies! I'll work on the suggestions and commit the updated code

@you-n-g and @SunsetWolf if it is of any help I can fix this documentation. I just need to undersand what exactly is wrong. Is it the IC definition? Or the...

@you-n-g I can fix this error. But do you know what does `lr_decay` and `lr_decay_steps` mean?

@pGit1 and @zhangfaen If possible I would like to ask you two questions regarding that "supervised" part. I read your discussion but two questions still remains for me. 1) _What...

Thank you for the response @pGit1 !

@jackmpcollins, thank you very much for the update on this. I’ve been trying to get it working with an open-source/local model, and I’ve been following your discussions across different issues....

Thank you for your response @andreeaiana! I just finished collecting the prints I mentioned before. I added them inside the `validation_step` as follows: ``` def validation_step(self, batch: RecommendationBatch, batch_idx: int):...

I'll add some print statements during the training part of epoch 2 to see what happens with the `preds` and `loss`

Apparently some predictions are being returned as `nan`. I've added the print statements inside `model_step` method to better visualized during training. Here is one example: ``` ########### targets ########### torch.Size([50])...

I'm struggling to understand why this is happening 🧐 The same error happened with `MINS` model as well on the 2nd epoch. I found this [issue](https://github.com/pytorch/pytorch/issues/41508) regarding `nn.MultiheadAttention`. Maybe that's...