ACGAN-PyTorch icon indicating copy to clipboard operation
ACGAN-PyTorch copied to clipboard

Published 20 hours ago • verified publisher icon clvrai

Softmax or logsoftmax for auxiliary prediction

Open haoyangz opened this issue 7 years ago • 1 comments

The current code uses softmax for auxiliary prediction and during training uses NLL Loss. But according to here it seems that the right last layer to go with NLL is LogSoftmax instead?

haoyangz avatar Mar 01 '18 21:03 haoyangz

However,I found only Softmax rather than LogSoftmax works.Could you verify the differences between these?

zhangrong1722 avatar Oct 17 '18 02:10 zhangrong1722