train.py

import torch
import torch.optim as optim
import torch.nn.functional as F
from model import CNN 
from dataset import train_loader, test_loader
from config import device

def train(model, device, train_loader, optimizer, epoch):
    losses = []
    model.train()
    for batch_idx, (data, target) in enumerate(train_loader):
        data, target = data.to(device), target.to(device)
        optimizer.zero_grad()
        output = model(data)
        loss = F.nll_loss(output, target)
        loss.backward()
        optimizer.step()
        losses.append(loss.item())
        if batch_idx % 10 == 0:
            print('Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.6f}'.format(
                epoch, batch_idx * len(data), len(train_loader.dataset),
                100. * batch_idx / len(train_loader), loss.item()))
    return losses

def test(model, device, test_loader):
    model.eval()
    test_loss = 0
    correct = 0
    with torch.no_grad():
        for data, target in test_loader:
            data, target = data.to(device), target.to(device)
            output = model(data)
            test_loss += F.nll_loss(output, target, reduction='sum').item() # sum up batch loss
            pred = output.argmax(dim=1, keepdim=True) # get the index of the max log-probability
            correct += pred.eq(target.view_as(pred)).sum().item()
    test_loss /= len(test_loader.dataset)
    print('\nTest set: Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)\n'.format(
        test_loss, correct, len(test_loader.dataset),
        100. * correct / len(test_loader.dataset)))
    return (float(correct) / len(test_loader.dataset))


model = CNN()
optimizer = optim.SGD(model.parameters(), lr=0.01, momentum=0.5)

losses = []
accuracies = []
for epoch in range(0, 1000):
    losses.extend(train(model, device, train_loader, optimizer, epoch))
    accuracies.append(test(model, device, test_loader))
    print(accuracies[-1])
# losses = [item for sublist in losses for item in sublist]

torch.save({'epoch': 1000,
              'model_state_dict': model.state_dict(),
              'optimizer_state_dict': optimizer.state_dict(),
              'loss': losses[-1],
              'acc': accuracies[-1]}, 
        './CNN_afterAImodel_checkpoint.pth')