CODEWITHSUNDEEP
pythonpytorchclassificationlstm
annonyos
annonyos

Reputation: 68

Pytorch-> LSTM-> RuntimeError: input must have 3 dimensions, got 2

I'm facing error in LSTM input dimensions with following model code:

class Model(nn.Module):
def __init__(self, input_size, hidden_size, num_layers, num_keys):
    super(Model, self).__init__()
    self.hidden_size = hidden_size
    self.num_layers = num_layers
    self.lstm = nn.LSTM(input_size, hidden_size, num_layers, batch_first=True)
    self.fc = nn.Linear(hidden_size, num_keys)

def forward(self, x):
    h0 = torch.zeros(self.num_layers, x.size(0), self.hidden_size).to(device)
    c0 = torch.zeros(self.num_layers, x.size(0), self.hidden_size).to(device)
    out, _ = self.lstm(x, (h0, c0))
    out = self.fc(out[:, -1, :])
    return out

Input sequence shape: 2048, 10,1 -> (#batch, window_size, #input)

Input label shape: 2048

Following code I have for training model:

for epoch in range(num_epochs):  # Loop over the dataset multiple times
    train_loss = 0
    for step, (seq, label) in enumerate(dataloader):
        # Forward pass
        seq = seq.clone().detach().view(-1, window_size, input_size).to(device)
        
        output = model(seq)
        loss = criterion(output, label.to(device))
        print('step: ',step,'sequence: ', seq.shape, 'Label: ', label.shape, 'model output: ', output.shape)
        
        # Backward and optimize
        optimizer.zero_grad()
        loss.backward()
        train_loss += loss.item()
        optimizer.step()
        writer.add_graph(model, seq)

I get following error:

RuntimeError: input must have 3 dimensions, got 2

Can anyone tell what is the problem should I fixed, I used seq.unsequeeze(-1) but its not working.

Upvotes: 0

Views: 255

Answers (1)

annonyos
annonyos

Reputation: 68

I got it, the input data containing -1 values. I have used map function to convert negative integers into positive as following:

line = tuple(map(int, line.strip().split()))

Please make it sure for classification your data should contain positive number!

Upvotes: 1

Related Questions