About CrossEntroyLoss

[SmartAI]

The pytorch document shows:

class torch.nn.CrossEntropyLoss(weight=None, size_average=True)[source]
This criterion combines LogSoftMax and NLLLoss in one single class.

SO could I remove softmax layer from the model sequential?

model.add_module("softmax", torch.nn.Softmax())

[vinhkhuc]

I've checked CrossEntropyLoss and found that it computes softmax internally. So, the softmax layer is unnecessary. Thanks for pointing that out.

[brando90]

do you have a line reference for this in the pytorch documentation?