How to save weights from training and then test

1 
  for i, (x,y) in enumerate(zip(feature_trainloader,label_trainloader), 0):
                optimizer.zero_grad()   
                output = self.forward(x)
                target = y
                loss = self.CrossEntropyLoss(output, target)
                print(loss)
                loss.backward()
                optimizer.step()            
  print('Finished Training')

I want to now calculate the accuracy by putting feature_testloader through the input layer and calculating accuracy by comparing to label_testloader. How can I do that because when the for loop ends and I call forward again the weights will be reset?

2

The weights should not be reset. You should just be able to do another for loop for your test loaders. The model weights will be saved in memory.