TypeError: .__init__() takes 1 positional argument but 3 were given

I am practicing pytorch nowdays, actually shifting from keras.

But I can not understand why I am getting this error:
TypeError: NeuralNet02.init() takes 1 positional argument but 3 were given

Here’s my code:

‘’'class NeuralNet02(nn.Module):
def int(self, in_size, out_size, num_hidden_layer=10, hidden_layer_size=128):

    self.in_size = in_size
    self.out_size = out_size
    self.num_hidden_layer = num_hidden_layer
    self.hidden_layer_size = hidden_layer_size

    # In PyTorch, the terms "module" and "layer" are often used interchangeably to refer to a
    # building block of a neural network that performs a specific computation on the input data.

    self.layers = nn.Sequential()  # initialization

    for i in range(num_hidden_layer):
        self.layers.add_module(f'fc{i}', nn.Linear(in_size, hidden_layer_size))   # adding layers/modules
        self.layers.add_module('activation', nn.ReLU())
        in_size = hidden_layer_size
    self.layers.add_module('classifier', nn.Sigmoid(hidden_layer_size, out_size))

def forward(self,inputs):
    out = self.layers(inputs)
    return out

x = torch.randn(8,14)
model = NeuralNet02(14,4)
output = model(x)

You have a typo in your code and are defining __int__ instead of __init__ (note the missing i).
Note that you can post code snippets by wrapping them into three backticks ```, which makes debugging easier as your current format is also broken.

1 Like