About train_loss = train_loss_ce + train_loss_mml

[mirrorballty]

Hi, I hava a question. I just want to see the effect when I use train_loss_mml in your code.
like this:
train_loss_mml = criterion_mml(outputs['M'], train_labels_batch)
train_loss = train_loss_ce + train_loss_mml
and the following warning information is generated
[W Resize.cpp:23] Warning: An output with one or more elements was resized since it had shape [], which does not match the required output shape [32, 4].This behavior is deprecated, and in a future PyTorch release outputs will not be resized unless they have zero elements. You can explicitly reuse an out tensor t by resizing it, inplace, to zero elements with t.resize_(0). (function resize_output_check)