Transfer Learning for Computer Vision Tutorial

Created On: Mar 24, 2017 | Last Updated: Jan 27, 2025 | Last Verified: Nov 05, 2024

Author: Sasank Chilamkurthy

In this tutorial, you will learn how to train a convolutional neural network for image classification using transfer learning. You can read more about the transfer learning at cs231n notes

Quoting these notes,

In practice, very few people train an entire Convolutional Network from scratch (with random initialization), because it is relatively rare to have a dataset of sufficient size. Instead, it is common to pretrain a ConvNet on a very large dataset (e.g. ImageNet, which contains 1.2 million images with 1000 categories), and then use the ConvNet either as an initialization or a fixed feature extractor for the task of interest.

These two major transfer learning scenarios look as follows:

• Finetuning the ConvNet: Instead of random initialization, we initialize the network with a pretrained network, like the one that is trained on imagenet 1000 dataset. Rest of the training looks as usual.

• ConvNet as fixed feature extractor: Here, we will freeze the weights for all of the network except that of the final fully connected layer. This last fully connected layer is replaced with a new one with random weights and only this layer is trained.

# License: BSD
# Author: Sasank Chilamkurthy

import torch
import torch.nn as nn
import torch.optim as optim
from torch.optim import lr_scheduler
import torch.backends.cudnn as cudnn
import numpy as np
import torchvision
from torchvision import datasets, models, transforms
import matplotlib.pyplot as plt
import time
import os
from PIL import Image
from tempfile import TemporaryDirectory

cudnn.benchmark = True
plt.ion()   # interactive mode

<contextlib.ExitStack object at 0x7effe1ea8670>

Load Data

We will use torchvision and torch.utils.data packages for loading the data.

The problem we’re going to solve today is to train a model to classify ants and bees. We have about 120 training images each for ants and bees. There are 75 validation images for each class. Usually, this is a very small dataset to generalize upon, if trained from scratch. Since we are using transfer learning, we should be able to generalize reasonably well.

This dataset is a very small subset of imagenet.

Note

Download the data from here and extract it to the current directory.

# Data augmentation and normalization for training
# Just normalization for validation
data_transforms = {
    'train': transforms.Compose([
        transforms.RandomResizedCrop(224),
        transforms.RandomHorizontalFlip(),
        transforms.ToTensor(),
        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
    ]),
    'val': transforms.Compose([
        transforms.Resize(256),
        transforms.CenterCrop(224),
        transforms.ToTensor(),
        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
    ]),
}

data_dir = 'data/hymenoptera_data'
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

# We want to be able to train our model on an `accelerator <https://pytorch.org/docs/stable/torch.html#accelerators>`__
# such as CUDA, MPS, MTIA, or XPU. If the current accelerator is available, we will use it. Otherwise, we use the CPU.

device = torch.accelerator.current_accelerator().type if torch.accelerator.is_available() else "cpu"
print(f"Using {device} device")

Using cuda device

Visualize a few images

Let’s visualize a few training images so as to understand the data augmentations.

def imshow(inp, title=None):
    """Display image for Tensor."""
    inp = inp.numpy().transpose((1, 2, 0))
    mean = np.array([0.485, 0.456, 0.406])
    std = np.array([0.229, 0.224, 0.225])
    inp = std * inp + mean
    inp = np.clip(inp, 0, 1)
    plt.imshow(inp)
    if title is not None:
        plt.title(title)
    plt.pause(0.001)  # pause a bit so that plots are updated


# Get a batch of training data
inputs, classes = next(iter(dataloaders['train']))

# Make a grid from batch
out = torchvision.utils.make_grid(inputs)

imshow(out, title=[class_names[x] for x in classes])

Training the model

Now, let’s write a general function to train a model. Here, we will illustrate:

• Scheduling the learning rate

• Saving the best model

In the following, parameter scheduler is an LR scheduler object from torch.optim.lr_scheduler.

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

    # Create a temporary directory to save training checkpoints
    with TemporaryDirectory() as tempdir:
        best_model_params_path = os.path.join(tempdir, 'best_model_params.pt')

        torch.save(model.state_dict(), best_model_params_path)
        best_acc = 0.0

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

            # Each epoch has a training and validation phase
            for phase in ['train', 'val']:
                if phase == 'train':
                    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)
                if phase == 'train':
                    scheduler.step()

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

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

                # deep copy the model
                if phase == 'val' and epoch_acc > best_acc:
                    best_acc = epoch_acc
                    torch.save(model.state_dict(), best_model_params_path)

            print()

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

        # load best model weights
        model.load_state_dict(torch.load(best_model_params_path, weights_only=True))
    return model

Visualizing the model predictions

Generic function to display predictions for a few images

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(f'predicted: {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)

Finetuning the ConvNet

Load a pretrained model and reset final fully connected layer.

model_ft = models.resnet18(weights='IMAGENET1K_V1')
num_ftrs = model_ft.fc.in_features
# Here the size of each output sample is set to 2.
# Alternatively, it can be generalized to ``nn.Linear(num_ftrs, len(class_names))``.
model_ft.fc = nn.Linear(num_ftrs, 2)

model_ft = model_ft.to(device)

criterion = nn.CrossEntropyLoss()

# Observe that all parameters are being optimized
optimizer_ft = optim.SGD(model_ft.parameters(), lr=0.001, momentum=0.9)

# Decay LR by a factor of 0.1 every 7 epochs
exp_lr_scheduler = lr_scheduler.StepLR(optimizer_ft, step_size=7, gamma=0.1)

Downloading: "https://download.pytorch.org/models/resnet18-f37072fd.pth" to /var/lib/ci-user/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth

  0%|          | 0.00/44.7M [00:00<?, ?B/s]
 80%|########  | 35.8M/44.7M [00:00<00:00, 375MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 375MB/s]

Train and evaluate

It should take around 15-25 min on CPU. On GPU though, it takes less than a minute.

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

Epoch 0/24
----------
train Loss: 0.6023 Acc: 0.6557
val Loss: 0.2522 Acc: 0.8889

Epoch 1/24
----------
train Loss: 0.4422 Acc: 0.8115
val Loss: 0.3640 Acc: 0.8431

Epoch 2/24
----------
train Loss: 0.4504 Acc: 0.8320
val Loss: 0.2131 Acc: 0.9281

Epoch 3/24
----------
train Loss: 0.5622 Acc: 0.7623
val Loss: 0.2396 Acc: 0.9216

Epoch 4/24
----------
train Loss: 0.5623 Acc: 0.8197
val Loss: 0.7665 Acc: 0.7386

Epoch 5/24
----------
train Loss: 0.4030 Acc: 0.8115
val Loss: 0.2360 Acc: 0.9216

Epoch 6/24
----------
train Loss: 0.3391 Acc: 0.8730
val Loss: 0.2995 Acc: 0.9150

Epoch 7/24
----------
train Loss: 0.3992 Acc: 0.8402
val Loss: 0.2339 Acc: 0.9412

Epoch 8/24
----------
train Loss: 0.2601 Acc: 0.8770
val Loss: 0.2320 Acc: 0.9412

Epoch 9/24
----------
train Loss: 0.3812 Acc: 0.8197
val Loss: 0.2153 Acc: 0.9477

Epoch 10/24
----------
train Loss: 0.2992 Acc: 0.8811
val Loss: 0.2129 Acc: 0.9477

Epoch 11/24
----------
train Loss: 0.2756 Acc: 0.8934
val Loss: 0.2023 Acc: 0.9477

Epoch 12/24
----------
train Loss: 0.3573 Acc: 0.8525
val Loss: 0.2067 Acc: 0.9216

Epoch 13/24
----------
train Loss: 0.2309 Acc: 0.9057
val Loss: 0.2210 Acc: 0.9216

Epoch 14/24
----------
train Loss: 0.3068 Acc: 0.8648
val Loss: 0.2048 Acc: 0.9412

Epoch 15/24
----------
train Loss: 0.3095 Acc: 0.8484
val Loss: 0.2156 Acc: 0.9346

Epoch 16/24
----------
train Loss: 0.3204 Acc: 0.8607
val Loss: 0.2023 Acc: 0.9346

Epoch 17/24
----------
train Loss: 0.3182 Acc: 0.8607
val Loss: 0.1989 Acc: 0.9542

Epoch 18/24
----------
train Loss: 0.3350 Acc: 0.8320
val Loss: 0.2555 Acc: 0.9020

Epoch 19/24
----------
train Loss: 0.2090 Acc: 0.9139
val Loss: 0.2034 Acc: 0.9477

Epoch 20/24
----------
train Loss: 0.2970 Acc: 0.8730
val Loss: 0.2047 Acc: 0.9542

Epoch 21/24
----------
train Loss: 0.1968 Acc: 0.9180
val Loss: 0.2033 Acc: 0.9412

Epoch 22/24
----------
train Loss: 0.2891 Acc: 0.8893
val Loss: 0.2081 Acc: 0.9542

Epoch 23/24
----------
train Loss: 0.3157 Acc: 0.8402
val Loss: 0.1919 Acc: 0.9542

Epoch 24/24
----------
train Loss: 0.1884 Acc: 0.9221
val Loss: 0.2073 Acc: 0.9542

Training complete in 0m 35s
Best val Acc: 0.954248

visualize_model(model_ft)

ConvNet as fixed feature extractor

Here, we need to freeze all the network except the final layer. We need to set requires_grad = False to freeze the parameters so that the gradients are not computed in backward().

You can read more about this in the documentation here.

model_conv = torchvision.models.resnet18(weights='IMAGENET1K_V1')
for param in model_conv.parameters():
    param.requires_grad = False

# Parameters of newly constructed modules have requires_grad=True by default
num_ftrs = model_conv.fc.in_features
model_conv.fc = nn.Linear(num_ftrs, 2)

model_conv = model_conv.to(device)

criterion = nn.CrossEntropyLoss()

# Observe that only parameters of final layer are being optimized as
# opposed to before.
optimizer_conv = optim.SGD(model_conv.fc.parameters(), lr=0.001, momentum=0.9)

# Decay LR by a factor of 0.1 every 7 epochs
exp_lr_scheduler = lr_scheduler.StepLR(optimizer_conv, step_size=7, gamma=0.1)

Train and evaluate

On CPU this will take about half the time compared to previous scenario. This is expected as gradients don’t need to be computed for most of the network. However, forward does need to be computed.

model_conv = train_model(model_conv, criterion, optimizer_conv,
                         exp_lr_scheduler, num_epochs=25)

Epoch 0/24
----------
train Loss: 0.6047 Acc: 0.6803
val Loss: 0.2463 Acc: 0.9346

Epoch 1/24
----------
train Loss: 0.3858 Acc: 0.8320
val Loss: 0.1976 Acc: 0.9281

Epoch 2/24
----------
train Loss: 0.5564 Acc: 0.7992
val Loss: 0.1575 Acc: 0.9673

Epoch 3/24
----------
train Loss: 0.4867 Acc: 0.7910
val Loss: 0.2104 Acc: 0.9346

Epoch 4/24
----------
train Loss: 0.3925 Acc: 0.8197
val Loss: 0.2177 Acc: 0.9020

Epoch 5/24
----------
train Loss: 0.4411 Acc: 0.7951
val Loss: 0.3881 Acc: 0.8366

Epoch 6/24
----------
train Loss: 0.6255 Acc: 0.7541
val Loss: 0.2421 Acc: 0.9020

Epoch 7/24
----------
train Loss: 0.4114 Acc: 0.8074
val Loss: 0.1970 Acc: 0.9346

Epoch 8/24
----------
train Loss: 0.3822 Acc: 0.8402
val Loss: 0.1834 Acc: 0.9608

Epoch 9/24
----------
train Loss: 0.4043 Acc: 0.8320
val Loss: 0.1887 Acc: 0.9477

Epoch 10/24
----------
train Loss: 0.3351 Acc: 0.8648
val Loss: 0.1667 Acc: 0.9608

Epoch 11/24
----------
train Loss: 0.3524 Acc: 0.8443
val Loss: 0.1779 Acc: 0.9477

Epoch 12/24
----------
train Loss: 0.3554 Acc: 0.8484
val Loss: 0.1783 Acc: 0.9542

Epoch 13/24
----------
train Loss: 0.5044 Acc: 0.7787
val Loss: 0.1901 Acc: 0.9281

Epoch 14/24
----------
train Loss: 0.4052 Acc: 0.8197
val Loss: 0.2193 Acc: 0.9020

Epoch 15/24
----------
train Loss: 0.3635 Acc: 0.8361
val Loss: 0.1919 Acc: 0.9346

Epoch 16/24
----------
train Loss: 0.3314 Acc: 0.8238
val Loss: 0.2211 Acc: 0.9281

Epoch 17/24
----------
train Loss: 0.3425 Acc: 0.8402
val Loss: 0.1970 Acc: 0.9412

Epoch 18/24
----------
train Loss: 0.3330 Acc: 0.8443
val Loss: 0.2088 Acc: 0.9281

Epoch 19/24
----------
train Loss: 0.3282 Acc: 0.8484
val Loss: 0.1741 Acc: 0.9542

Epoch 20/24
----------
train Loss: 0.3353 Acc: 0.8443
val Loss: 0.1731 Acc: 0.9477

Epoch 21/24
----------
train Loss: 0.2944 Acc: 0.8648
val Loss: 0.2171 Acc: 0.9216

Epoch 22/24
----------
train Loss: 0.3734 Acc: 0.8320
val Loss: 0.1805 Acc: 0.9477

Epoch 23/24
----------
train Loss: 0.3179 Acc: 0.8770
val Loss: 0.1770 Acc: 0.9412

Epoch 24/24
----------
train Loss: 0.3497 Acc: 0.8443
val Loss: 0.1840 Acc: 0.9542

Training complete in 0m 27s
Best val Acc: 0.967320

visualize_model(model_conv)

plt.ioff()
plt.show()

Inference on custom images

Use the trained model to make predictions on custom images and visualize the predicted class labels along with the images.

def visualize_model_predictions(model,img_path):
    was_training = model.training
    model.eval()

    img = Image.open(img_path)
    img = data_transforms['val'](img)
    img = img.unsqueeze(0)
    img = img.to(device)

    with torch.no_grad():
        outputs = model(img)
        _, preds = torch.max(outputs, 1)

        ax = plt.subplot(2,2,1)
        ax.axis('off')
        ax.set_title(f'Predicted: {class_names[preds[0]]}')
        imshow(img.cpu().data[0])

        model.train(mode=was_training)

visualize_model_predictions(
    model_conv,
    img_path='data/hymenoptera_data/val/bees/72100438_73de9f17af.jpg'
)

plt.ioff()
plt.show()

Further Learning

If you would like to learn more about the applications of transfer learning, checkout our Quantized Transfer Learning for Computer Vision Tutorial.