Got: TypeError: 'bool' object is not callable when building CNN using PyTorch

Question

I got the following error when using PyTorch to build a convolutional neural network
TypeError: 'bool' object is not callable.

Attached is the related code:

class alpha(nn.Module):
    '''
    This is the alpha class
    '''

    def __init__(self, alpha_val=0, minus=False, train=True):
        super(alpha, self).__init__()
        if train:
            self.alpha = nn.Parameter(torch.Tensor([alpha_val]).to(device))
            # This sentence defines that alpha is a parameter that to be optimized
            self.alpha.requires_grad = True
        else:
            self.alpha = torch.Tensor([alpha_val]).to(device)
            self.alpha.requires_grad = False
        self.minus = minus

    def forward(self, x):
        out = torch.mul(self.alpha, x)
        if self.minus:
            out = torch.mul(out, -1)
        return out

class InterMedium_Layer(nn.Module):
    def __init__(self, train=False, alpha_threshold=0.9) -> object:
        super(InterMedium_Layer, self).__init__()
        self.alpha1 = alpha()
        self.train = train
        self.alpha_threshold = alpha_threshold

    def forward(self, x):
        if (not self.train) and self.alpha1.alpha.item() < self.alpha_threshold:
            return x
        else:
            out = self.alpha1(x)
            out += F.relu(out)
            out += self.alpha1(x, minus=True)
            return out


class BN_Conv2d_f(nn.Module):
    def __init__(self, in_channels: object, out_channels: object, kernel_size: object, stride: object, padding: object,
                 dilation=1, groups=1, bias=False, activation=True) -> object:
        super(BN_Conv2d_f, self).__init__()
        layers = [nn.Conv2d(in_channels, out_channels, kernel_size=kernel_size, stride=stride,
                            padding=padding, dilation=dilation, groups=groups, bias=bias),
                  nn.BatchNorm2d(out_channels)]
        self.InterMedium_Layer = InterMedium_Layer()
        if activation:
            # Error: bool obj not callable
            layers.append(self.InterMedium_Layer)
        self.seq = nn.Sequential(*layers)

I guess I called a bool var somewhere but I didn't find one.
Attached is the traceback and debug detail:
Debug page

Traceback Page

I find that the model becomes a bool type var, not a cnn network. Is there anything wrong with my code?

Answer 1

It seems quite odd that You are calling self.alpha = alpha() as a function, can You provide the code to that? (Or perhaps this is the issue)

Answer 2

"def init(self, alpha_val=0, minus=False, train=True)", define the 'train' paramater to an other word.

Answer 3

It seems necessary to change requires_grad to requires_grad_. maybe "_" is missing