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 0x7fd03a669090>

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]
 92%|#########1| 41.0M/44.7M [00:00<00:00, 430MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 430MB/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: 0.5894 Acc: 0.6844
val Loss: 0.2282 Acc: 0.9281

Epoch 1/24
----------
train Loss: 0.4490 Acc: 0.8115
val Loss: 0.2017 Acc: 0.9216

Epoch 2/24
----------
train Loss: 0.3718 Acc: 0.8361
val Loss: 0.1896 Acc: 0.9216

Epoch 3/24
----------
train Loss: 0.6726 Acc: 0.7582
val Loss: 0.2200 Acc: 0.9085

Epoch 4/24
----------
train Loss: 0.4802 Acc: 0.7951
val Loss: 0.2091 Acc: 0.9216

Epoch 5/24
----------
train Loss: 0.5718 Acc: 0.7828
val Loss: 0.4132 Acc: 0.8693

Epoch 6/24
----------
train Loss: 0.5300 Acc: 0.7828
val Loss: 0.2516 Acc: 0.9346

Epoch 7/24
----------
train Loss: 0.2955 Acc: 0.8770
val Loss: 0.2195 Acc: 0.9216

Epoch 8/24
----------
train Loss: 0.2848 Acc: 0.8607
val Loss: 0.2060 Acc: 0.9281

Epoch 9/24
----------
train Loss: 0.3362 Acc: 0.8607
val Loss: 0.2023 Acc: 0.9281

Epoch 10/24
----------
train Loss: 0.3156 Acc: 0.8893
val Loss: 0.1879 Acc: 0.9477

Epoch 11/24
----------
train Loss: 0.3230 Acc: 0.8484
val Loss: 0.2053 Acc: 0.9412

Epoch 12/24
----------
train Loss: 0.3497 Acc: 0.8525
val Loss: 0.1911 Acc: 0.9412

Epoch 13/24
----------
train Loss: 0.3187 Acc: 0.8730
val Loss: 0.1944 Acc: 0.9412

Epoch 14/24
----------
train Loss: 0.3172 Acc: 0.8648
val Loss: 0.2119 Acc: 0.9216

Epoch 15/24
----------
train Loss: 0.2848 Acc: 0.9057
val Loss: 0.2125 Acc: 0.9216

Epoch 16/24
----------
train Loss: 0.3151 Acc: 0.8648
val Loss: 0.1981 Acc: 0.9281

Epoch 17/24
----------
train Loss: 0.2951 Acc: 0.9057
val Loss: 0.2114 Acc: 0.9412

Epoch 18/24
----------
train Loss: 0.2970 Acc: 0.8566
val Loss: 0.1984 Acc: 0.9346

Epoch 19/24
----------
train Loss: 0.3276 Acc: 0.8566
val Loss: 0.2128 Acc: 0.9412

Epoch 20/24
----------
train Loss: 0.2619 Acc: 0.8934
val Loss: 0.2143 Acc: 0.9281

Epoch 21/24
----------
train Loss: 0.2827 Acc: 0.8770
val Loss: 0.2026 Acc: 0.9346

Epoch 22/24
----------
train Loss: 0.3180 Acc: 0.8730
val Loss: 0.1976 Acc: 0.9412

Epoch 23/24
----------
train Loss: 0.2576 Acc: 0.8975
val Loss: 0.2068 Acc: 0.9412

Epoch 24/24
----------
train Loss: 0.2752 Acc: 0.8689
val Loss: 0.2122 Acc: 0.9346

Training complete in 0m 35s
Best val Acc: 0.947712

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: 0.4999 Acc: 0.7418
val Loss: 0.2889 Acc: 0.8693

Epoch 1/24
----------
train Loss: 0.4880 Acc: 0.7623
val Loss: 0.1834 Acc: 0.9412

Epoch 2/24
----------
train Loss: 0.5824 Acc: 0.7336
val Loss: 0.2352 Acc: 0.9216

Epoch 3/24
----------
train Loss: 0.5771 Acc: 0.7787
val Loss: 0.2371 Acc: 0.9216

Epoch 4/24
----------
train Loss: 0.4688 Acc: 0.7951
val Loss: 0.2097 Acc: 0.9412

Epoch 5/24
----------
train Loss: 0.4198 Acc: 0.8197
val Loss: 0.2107 Acc: 0.9412

Epoch 6/24
----------
train Loss: 0.3475 Acc: 0.8279
val Loss: 0.2168 Acc: 0.9281

Epoch 7/24
----------
train Loss: 0.3873 Acc: 0.7992
val Loss: 0.1752 Acc: 0.9542

Epoch 8/24
----------
train Loss: 0.3670 Acc: 0.8115
val Loss: 0.1901 Acc: 0.9412

Epoch 9/24
----------
train Loss: 0.3436 Acc: 0.8361
val Loss: 0.1704 Acc: 0.9477

Epoch 10/24
----------
train Loss: 0.3640 Acc: 0.8607
val Loss: 0.1867 Acc: 0.9542

Epoch 11/24
----------
train Loss: 0.3165 Acc: 0.8770
val Loss: 0.1869 Acc: 0.9477

Epoch 12/24
----------
train Loss: 0.2984 Acc: 0.8893
val Loss: 0.2084 Acc: 0.9477

Epoch 13/24
----------
train Loss: 0.3373 Acc: 0.8443
val Loss: 0.1769 Acc: 0.9412

Epoch 14/24
----------
train Loss: 0.3864 Acc: 0.8115
val Loss: 0.1736 Acc: 0.9412

Epoch 15/24
----------
train Loss: 0.3072 Acc: 0.8443
val Loss: 0.1812 Acc: 0.9412

Epoch 16/24
----------
train Loss: 0.3223 Acc: 0.8607
val Loss: 0.1700 Acc: 0.9608

Epoch 17/24
----------
train Loss: 0.3325 Acc: 0.8279
val Loss: 0.1828 Acc: 0.9477

Epoch 18/24
----------
train Loss: 0.3788 Acc: 0.8361
val Loss: 0.2016 Acc: 0.9477

Epoch 19/24
----------
train Loss: 0.3254 Acc: 0.8402
val Loss: 0.1730 Acc: 0.9477

Epoch 20/24
----------
train Loss: 0.3041 Acc: 0.8689
val Loss: 0.1750 Acc: 0.9412

Epoch 21/24
----------
train Loss: 0.3409 Acc: 0.8361
val Loss: 0.2081 Acc: 0.9346

Epoch 22/24
----------
train Loss: 0.2774 Acc: 0.8770
val Loss: 0.1765 Acc: 0.9412

Epoch 23/24
----------
train Loss: 0.4064 Acc: 0.7951
val Loss: 0.1806 Acc: 0.9477

Epoch 24/24
----------
train Loss: 0.2889 Acc: 0.8320
val Loss: 0.1855 Acc: 0.9477

Training complete in 0m 27s
Best val Acc: 0.960784

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.