Size mismatch for fc.bias and fc.weight in PyTorch

Question

I used the transfer learning approach to train a model and saved the best-detected weights. In another script, I tried to use the saved weights for prediction. But I am getting errors as follows. I have used ResNet for finetuning the network and have 4 classes.

RuntimeError: Error(s) in loading state_dict for ResNet:
size mismatch for fc.bias: copying a param of torch.Size([1000]) from 
checkpoint, where the shape is torch.Size([4]) in current model.
size mismatch for fc.weight: copying a param of torch.Size([1000, 
512]) from checkpoint, where the shape is torch.Size([4, 512]) in 
current model.

I am using the following code for prediction of output:

checkpoint = torch.load("./models/custom_model13.model")
model = resnet18(pretrained=True)

model.load_state_dict(checkpoint)
model.eval()

def predict_image(image_path):
    transformation = transforms.Compose([
        transforms.RandomResizedCrop(224),
        transforms.RandomHorizontalFlip(),
        transforms.ToTensor(),
        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
        ])
    image_tensor = transformation(image).float()
    image_tensor = image_tensor.unsqueeze_(0)

    if torch.cuda.is_available():
        image_tensor.cuda()

    input = Variable(image_tensor)
    output = model(input)

    index = output.data.numpy().argmax()
    return index

if __name__ == "main":
    imagefile = "image.png"
    imagepath = os.path.join(os.getcwd(),imagefile)
    prediction = predict_image(imagepath)
    print("Predicted Class: ",prediction)

And the following code to train and save the model:

Data_dir = 'Dataset'
image_datasets = {x: datasets.ImageFolder(os.path.join(data_dir, x),
                                          data_transforms[x])
                  for x in ['train', 'val']}
dataloaders = {x: torch.utils.data.DataLoader(image_datasets[x], batch_size=4,
                                             shuffle=True, num_workers=4)
              for x in ['train', 'val']}
dataset_sizes = {x: len(image_datasets[x]) for x in ['train', 'val']}
class_names = image_datasets['train'].classes

device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
print (device)

def save_models(epochs, model):
    torch.save(model.state_dict(), "custom_model{}.model".format(epochs))
    print("Checkpoint Saved")

def train_model(model, criterion, optimizer, scheduler, num_epochs=25):
    since = time.time()

    best_model_wts = copy.deepcopy(model.state_dict())
    best_acc = 0.0

    for epoch in range(num_epochs):
        print('Epoch {}/{}'.format(epoch, num_epochs - 1))
        print('-' * 10)

        # Each epoch has a training and validation phase
        for phase in ['train', 'val']:
            if phase == 'train':
                scheduler.step()
                model.train()  # Set model to training mode
            else:
                model.eval()   # Set model to evaluate mode

            running_loss = 0.0
            running_corrects = 0

            # Iterate over data.
            for inputs, labels in dataloaders[phase]:
                inputs = inputs.to(device)
                labels = labels.to(device)

                # zero the parameter gradients
                optimizer.zero_grad()

                # forward
                # track history if only in train
                with torch.set_grad_enabled(phase == 'train'):
                    outputs = model(inputs)
                    _, preds = torch.max(outputs, 1)
                    loss = criterion(outputs, labels)

                    # backward + optimize only if in training phase
                    if phase == 'train':
                        loss.backward()
                        optimizer.step()

                # statistics
                running_loss += loss.item() * inputs.size(0)
                running_corrects += torch.sum(preds == labels.data)

            epoch_loss = running_loss / dataset_sizes[phase]
            epoch_acc = running_corrects.double() / dataset_sizes[phase]

            print('{} Loss: {:.4f} Acc: {:.4f}'.format(
                phase, epoch_loss, epoch_acc))

            # deep copy the model
            if phase == 'train' and epoch_acc > best_acc:
                save_models(epoch,model)
                best_acc = epoch_acc
                best_model_wts = copy.deepcopy(model.state_dict())

        print()

    time_elapsed = time.time() - since
    print('Training complete in {:.0f}m {:.0f}s'.format(
        time_elapsed // 60, time_elapsed % 60))
    print('Best val Acc: {:4f}'.format(best_acc))

    # load best model weights
    model.load_state_dict(best_model_wts)
    return model

def visualize_model(model, num_images=6):
    was_training = model.training
    model.eval()
    images_so_far = 0
    fig = plt.figure()

    with torch.no_grad():
        for i, (inputs, labels) in enumerate(dataloaders['val']):
            inputs = inputs.to(device)
            labels = labels.to(device)

            outputs = model(inputs)
            _, preds = torch.max(outputs, 1)

            for j in range(inputs.size()[0]):
                images_so_far += 1
                ax = plt.subplot(num_images//2, 2, images_so_far)
                ax.axis('off')
                ax.set_title('predicted: {}'.format(class_names[preds[j]]))
                imshow(inputs.cpu().data[j])

                if images_so_far == num_images:
                    model.train(mode=was_training)
                    return
        model.train(mode=was_training)

model_ft = models.resnet18(pretrained=True)
num_ftrs = model_ft.fc.in_features
model_ft.fc = nn.Linear(num_ftrs, 4)

model_ft = model_ft.to(device)

criterion = nn.CrossEntropyLoss()


optimizer_ft = optim.SGD(model_ft.parameters(), lr=0.001, momentum=0.9)

exp_lr_scheduler = lr_scheduler.StepLR(optimizer_ft, step_size=7, gamma=0.1)

model_ft = train_model(model_ft, criterion, optimizer_ft, exp_lr_scheduler,
                       num_epochs=25)

Answer 1

Cause:

You trained a model derived from resnet18 in this way:

model_ft = models.resnet18(pretrained=True)
num_ftrs = model_ft.fc.in_features
model_ft.fc = nn.Linear(num_ftrs, 4)

That is, you changed the last nn.Linear layer to output 4 dim prediction instead of the default 1000.
When you try and load the model for prediction, your code is:

model = resnet18(pretrained=True)    
model.load_state_dict(checkpoint)

You did not apply the same change of the last nn.Linear layer to model therefore the checkpoint you are trying to load does not fit.

Fix:

(1) Apply the same change before loading the checkpoint:

model = resnet18(pretrained=True)    
num_ftrs = model_ft.fc.in_features
model_ft.fc = nn.Linear(num_ftrs, 4)  # make the change
model.load_state_dict(checkpoint)  # load

(2) Even better, use num_classes argument to construct resnet with the desired number of outputs to begin with:

model = resnet18(pretrained=True, num_classes=4)  
model.load_state_dict(checkpoint)  # load