juanpablomf/deep-learning-platzi

Materiales del curso de Deep Learning con Pytorch de Platzi

deep-learning-platzi's Issues

NotImplementedError:

Just before trining the net me and a couple of students had this error:
`for epoch in range(num_epochs):
print('Epoch {}/{}'.format(epoch+1, num_epochs))
print('-'*10)

running_loss = RunningMetric() # perdida
running_acc = RunningMetric() #precision

for inputs, targets in dataloader:
inputs, targets = inputs.to(device), targets.to(device)
optimizer.zero_grad()

outputs = net(inputs)
_, preds = torch.max(outputs,1)
loss = loss_fn(outputs, targets)

loss.backward() #magias: gradientes calculados automaticamente
optimizer.step() #magia2: actualiza las perillas o los parametros

batch_size = inputs.size()[0]
running_loss.update(loss.item()*batch_size,
                   batch_size)
running_acc.update(torch.sum(preds == targets).float(),
                   batch_size)

print("Loss: {:.4f} Acc: {:.4f} ".format(running_loss(),
running_acc()))`

Output:
`Epoch 1/100

---

NotImplementedError Traceback (most recent call last)
in ()
10 optimizer.zero_grad()
11
---> 12 outputs = net(inputs)
13 _, preds = torch.max(outputs,1)
14 loss = loss_fn(outputs, targets)

1 frames
/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py in _forward_unimplemented(self, *input)
199 registered hooks while the latter silently ignores them.
200 """
--> 201 raise NotImplementedError
202
203

NotImplementedError: `

I tried simple things to fix it, like restarting and re-install some packages but without any good results