MaskZeroCriterion

[sdalaman (ŞABAN DALAMAN)]

Hi,
İ am trying to convert a lua/torch program to pytorch. I have used nn.MaskZeroCriterion method in lua. Is there any method in pytoch i can use?If not, can you suggest ways how i can implement in pytorch?
thanks

[DiffEverything]

I dont think there is any implementation of maskedcriterion. The best way is to mask the score vector/logits yourself as per this thread, using masked_select. gather doesn’t works for variable length sequences.

Having said that, i’m also interested to know if there’s a better approach to this

How can i compute seq2seq loss using mask?

I am working on image captioning task with PyTorch. In seq2seq, padding is used to handle the variable-length sequence problems. Additionally, mask is multiplied by the calculated loss (vector not scalar) so that the padding does not affect the loss. In TensorFlow, i can do this as below. # targets is an int64 tensor of shape (batch_size, padded_length) which contains word indices. # masks is a tensor of shape (batch_size, padded_length) which contains 0 or 1 (0 if pad otherwise 1). out…