# training loop
for epoch in range(2):
running_loss = 0.0
i = 0
with tqdm(trainloader) as tqdm_iterator:
for data in tqdm_iterator:
# get the inputs; data is a list of [inputs, labels]
inputs, labels = data
# zero the parameter gradients
optimizer.zero_grad()
# forward + backward + optimize
outputs = net(inputs)
target = net.get_target(labels)
loss = loss_fn(outputs, target)
loss.backward()
optimizer.step()
i += 1
running_loss += loss.item()
if i > 1000:
i = 0
tqdm_iterator.set_description(f"{running_loss:5f}")
running_loss = 0.0
print('Finished Training')
# torch.save(net.state_dict(), PATH)
