Residual.py

class Residual(nn.Module):
    """
    Residual module from "Deep Residual Learning for Image Recognition"
    (https://arxiv.org/pdf/1512.03385.pdf)
    """
    def __init__(self, inp_C, out_C, stride = 1, batch_norm=True):
        super(Residual, self).__init__()
        self.conv1 = nn.Conv2d(inp_C, out_C, kernel_size=3, stride=stride, padding=1)
        self.bn1 = nn.BatchNorm2d(out_C) if batch_norm else nn.Sequential()
        self.a1 = nn.ReLU(inplace=True)
        
        self.conv2 = nn.Conv2d(out_C, out_C, kernel_size=3, padding=1)
        self.bn2 = nn.BatchNorm2d(out_C) if batch_norm else nn.Sequential()
        
        changed_shape = (stride != 1) or (inp_C != out_C)
        
        self.skip = nn.Sequential(nn.Conv2d(inp_C, out_C, kernel_size=1, stride=stride), 
                                   nn.BatchNorm2d(out_C)) if changed_shape else nn.Sequential()

        self.a2 = nn.ReLU(inplace=True)
            
    def forward(self, inp):
        x = self.conv1(inp)
        x = self.bn1(x)
        x = self.a1(x)
        x = self.conv2(x)
        x = self.bn2(x)
        
        return self.a2(x + self.skip(inp))