CODEWITHSUNDEEP
pythonpytorch
Gabriel A.
Gabriel A.

Reputation: 129

TypeError: forward() takes 2 positional arguments but 3 were given in pytorch

I have the following error in my training loop and I don't really understand what the issue is. I am currently in the process of writing this code so stuff isn't final but I cannot figure out what this problem is.

I have tried googling the error and read some of the answers but still couldn't seem to understand the crux of the issue.

Dataset and Dataloader (X and Y are already given to me, they are both [2000, 40, 1] tensors)

class TrainingDataset(data.Dataset):
  def __init__(self, X, y):
    self.X = X
    self.y = y

  def __len__(self):
    return Nf
    
  # returns corresponding input/output pairs
  def __getitem__(self, t):
    X = self.X[t]
    y = self.y[t]

    #print(X.shape, y.shape)

    return X, y

# prints torch.Size([2000, 40, 1]) torch.Size([2000, 40, 1])
print(x.size(), y.size())

dataset = TrainingDataset(x,y)
batchSize = 20
dataIter = data.DataLoader(dataset, batchSize)

Model:

class Encoder(nn.Module):
  def __init__(self, num_inputs = 40, num_outputs = 40):
    super(Encoder, self).__init__()
    
    self.num_inputs = num_inputs
    self.num_hidden = num_hidden
    self.num_outputs = num_outputs

    self.layers = nn.Sequential(
        nn.Linear(num_inputs, num_outputs), 
        nn.ReLU(),
        nn.Linear(num_outputs, num_outputs),
        nn.ReLU(),
        nn.Linear(num_outputs, num_outputs) 
    )

  def forward(self, x_c, y_c):
    return self.layers(x_c, y_c)

Training Loop:

for epoch in range(epochs): 
  for batch in dataIter:
    optimiser.zero_grad()
    l = loss(encoder(x_c=batch[0], y_c=batch[1]), batch[1])
    l.backward()
    optimiser.step()

Error:

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-15-aa1c60616d82> in <module>()
      6   for batch in dataIter:
      7     optimiser.zero_grad()
----> 8     l = loss(encoder(x_c=batch[0], y_c=batch[1]), batch[1])
      9     l.backward()
     10     optimiser.step()

2 frames
/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py in _call_impl(self, *input, **kwargs)
    887             result = self._slow_forward(*input, **kwargs)
    888         else:
--> 889             result = self.forward(*input, **kwargs)
    890         for hook in itertools.chain(
    891                 _global_forward_hooks.values(),

TypeError: forward() takes 2 positional arguments but 3 were given

Can anyone point me in the right direction? I have just started to learn and do pytorch so I am not good at any of this yet.

Upvotes: 4

Views: 18652

Answers (2)

Guillermo Alfaro
Guillermo Alfaro

Reputation: 1

I had a similar issue and solved it with this:

class SequentialDecoder(nn.Sequential):
    def forward(self, *inputs):
        x, y = inputs
        for module in self._modules.values():
            x = module(x, y)
        return x

Upvotes: 0

Vulwsztyn
Vulwsztyn

Reputation: 2271

  def forward(self, x_c, y_c):
    return self.layers(x_c, y_c)

Your error lies here, this function should have only 1 argument apart from self.

Upvotes: 1

Related Questions