I have defined a convolutional neural network as follows:
class CNN(nn.Module):
def __init__(self):
super(CNN, self).__init__()
self.layer1 = nn.Sequential(
nn.Conv2d(1, 20, kernel_size=5, padding=2),
nn.BatchNorm2d(20),
nn.ReLU(),
nn.MaxPool2d(2))
self.layer2 = nn.Sequential(
nn.Conv2d(20, 20, kernel_size=5, padding=2),
nn.BatchNorm2d(20),
nn.ReLU(),
nn.MaxPool2d(2))
self.fc = nn.Linear(MxNx20, 10)
def forward(self, x):
out = self.layer1(x)
out = self.layer2(out)
out = out.view(out.size(0), -1)
out = self.fc(out)
return out
cnn = CNN()
In self.fc layer of nn.module i have used M and N in order to define weight matrix size. M and N can take values like 5x5, 7x7, 10x10 and so on. I can not define M and N in terms of self.layer2. Is there any way to predefine size of the weight matrix of self.fc layer