net = nn.Linear(out_, in_, bias=False)
def forward(ctx, input, net):
output = net(input)
def backward(ctx, grad_output):...
How to use a net inside autograd.Function? Is it possible? What are my options?
What are you trying to achieve?
While it is possible (you’d want to assign the net to ctx._net instead of using
save_for_backward, probably), it certainly is not advisable to do so.
A more natural way would be to pass the net.weight as an argument to the forward (and, if you don’t want to differentiate with respect to that, return None).