HELP! Can we implement parallel computing in the 'forward' function?

please tell me how to implement parallel running of two pieces of code on the GPU, just like multithreading on cpu?

def forward(self, x):

y=self.conv2(x)
z=self.conv2(x)
return y+z

For example, the above code can we execute y=self.conv2(x), z=self.conv2(x) at the same time,?Looking forward to your reply , thanks!!!