Separate last fully connected layer weights

1

Is there any way of separating the final fully connected layer weights after a few local epochs of training? How could I append them into a vector?

2 
def train(self, is_print = False):
        self.net.to(self.device)
        self.net.train()
        
        optimizer = torch.optim.SGD(self.net.parameters(), lr=self.lr, momentum=self.momentum, weight_decay=0)
        
        epoch_loss = []
        for iteration in range(self.local_ep):
            batch_loss = []
            for batch_idx, (images, labels) in enumerate(self.ldr_train):
                images, labels = images.to(self.device), labels.to(self.device)
                self.net.zero_grad()
                #optimizer.zero_grad()
                log_probs = self.net(images)
                loss = self.loss_func(log_probs, labels)
                loss.backward() 
                        
                optimizer.step()
                batch_loss.append(loss.item())
            


            epoch_loss.append(sum(batch_loss)/len(batch_loss))
         
        
        return sum(epoch_loss) / len(epoch_loss)