AFAIK there are 2 ways to express ResNet Block in pytorch:

• Copy the input in the beginning, modify the input in the process, add the copy in the end.
• Preserve the input in the beginning, create new variable in the process, add the input in the end.

Which leads to 2 kinds of code:

def forward(self, x):
    y = x
    x = self.conv1(x)
    x = self.norm1(x)
    x = self.act1(x)
    x = self.conv2(x)
    x = self.norm2(x)
    x += y
    x = self.act2(x)
    return x

def forward(self, x):
    y = self.conv1(x)
    y = self.norm1(y)
    y = self.act1(y)
    y = self.conv2(y)
    y = self.norm2(y)
    y += x
    y = self.act2(y)
    return y

Are they identical? Which one is preferred? Why?