gistfile1.txt

num_epochs = 1


import wandb
entity = "mohamedinspace"
project = "Recaptcha-Solver"
name = "ResNet10" # you can change the name of your runs
wb = wandb . init (
  entity = entity ,
  project = project ,
  name = name ,
)


for epoch in range(num_epochs):
    # Training phase
    model.train()
    running_loss = 0.0
    correct = 0
    total = 0

    train_loader_tqdm = tqdm(
        train_loader, desc=f"Epoch {epoch + 1}/{num_epochs} [Training]"
    )

    for inputs, labels in train_loader_tqdm:
        inputs, labels = inputs.to(device), labels.to(device)

        # Zero the parameter gradients
        optimizer.zero_grad()

        # Forward pass
        outputs = model(inputs)
        loss = criterion(outputs, labels)

        # Backward pass and optimization
        loss.backward()
        optimizer.step()

        # Statistics
        running_loss += loss.item()
        _, preds = torch.max(outputs, 1)
        correct += (preds == labels).sum().item()
        total += labels.size(0)

        train_loader_tqdm.set_postfix(
            loss=loss.item(), accuracy=correct / total
        )

    train_loss = running_loss / len(train_loader)
    train_accuracy = correct / total

    # Validation phase
    model.eval()
    val_loss = 0.0
    val_correct = 0
    val_total = 0

    with torch.no_grad():
        for inputs, labels in val_loader:
            inputs, labels = inputs.to(device), labels.to(device)
            outputs = model(inputs)
            loss = criterion(outputs, labels)

            val_loss += loss.item()
            _, preds = torch.max(outputs, 1)
            val_correct += (preds == labels).sum().item()
            val_total += labels.size(0)

    val_loss /= len(val_loader)
    val_accuracy = val_correct / val_total

    wb.log(
        {
          "Train Loss" : train_loss ,
          "Train Accuracy" : train_accuracy ,
          "Val Loss" : val_loss ,
          "Val Accuracy" : val_accuracy ,
        },
        step = epoch + 1 ,
        commit = True ,
        sync = True ,
    )

    print(f"Epoch {epoch+1}/{num_epochs}, "
          f"Train Loss: {train_loss:.4f}, Train Acc: {train_accuracy:.4f}, "
          f"Val Loss: {val_loss:.4f}, Val Acc: {val_accuracy:.4f}")