Expected Object Type Variable[torch.DoubleTensor] but found is Variable[torch.FloatTensor]

21

It then just says

ValueError: too many values to unpack (expected 2)

22

Iā€™m sorry, I currently have no computer to test this but from your description this might work:

for e in range(epochs):
    loss_values = [] 
    for data in np.dstack((x_train, y_train)):
        _x = data[0]
        _y = data[1]
        inps = Variable(torch.from_numpy(_x).to(torch.float())
        outs = Variable(torch.from_numpy(_y).to(torch.float())
    
        optm.zero_grad()
        outputs = model.forward(inps)
        loss = criterion(outputs, outs)
        loss.backward()
        optm.step()
        
        scalar_loss = no.asscalar(loss.detach().cpu().numpy()) 
        loss_values.append(scalar_loss) 

print('epoch {}, loss {}'.format(e,np.mean(np.asarray(loss_values))