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

About replace in ReLU and Dropout

Open TomMao23 opened this issue 3 years ago • 0 comments

I noticed that a lot of your code is written like this

nn.ReLU(True),
nn.Dropout(0.1, False),

why you use inplace operator in activation fuction but not in dropout? Does this have any special meaning?

TomMao23 avatar Jun 23 '21 07:06 TomMao23