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 0x7f94a3234bb0>

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]
 81%|########  | 36.0M/44.7M [00:00<00:00, 376MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 383MB/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: 0.6661 Acc: 0.6926
val Loss: 0.2670 Acc: 0.9085

Epoch 1/24
----------
train Loss: 0.4811 Acc: 0.7992
val Loss: 0.2006 Acc: 0.9281

Epoch 2/24
----------
train Loss: 0.5171 Acc: 0.7664
val Loss: 0.3335 Acc: 0.8693

Epoch 3/24
----------
train Loss: 0.5182 Acc: 0.7459
val Loss: 0.3361 Acc: 0.8889

Epoch 4/24
----------
train Loss: 0.4826 Acc: 0.8115
val Loss: 0.4395 Acc: 0.8170

Epoch 5/24
----------
train Loss: 0.3460 Acc: 0.8607
val Loss: 0.2264 Acc: 0.9281

Epoch 6/24
----------
train Loss: 0.3779 Acc: 0.8525
val Loss: 0.2984 Acc: 0.8954

Epoch 7/24
----------
train Loss: 0.3792 Acc: 0.8361
val Loss: 0.2395 Acc: 0.9216

Epoch 8/24
----------
train Loss: 0.3578 Acc: 0.8525
val Loss: 0.2127 Acc: 0.9216

Epoch 9/24
----------
train Loss: 0.2452 Acc: 0.8811
val Loss: 0.2258 Acc: 0.9085

Epoch 10/24
----------
train Loss: 0.3843 Acc: 0.8402
val Loss: 0.2053 Acc: 0.9281

Epoch 11/24
----------
train Loss: 0.3594 Acc: 0.8525
val Loss: 0.2049 Acc: 0.9346

Epoch 12/24
----------
train Loss: 0.2574 Acc: 0.8934
val Loss: 0.1991 Acc: 0.9216

Epoch 13/24
----------
train Loss: 0.2844 Acc: 0.8648
val Loss: 0.2218 Acc: 0.9346

Epoch 14/24
----------
train Loss: 0.3261 Acc: 0.8730
val Loss: 0.1973 Acc: 0.9281

Epoch 15/24
----------
train Loss: 0.2203 Acc: 0.9098
val Loss: 0.2108 Acc: 0.9281

Epoch 16/24
----------
train Loss: 0.2952 Acc: 0.8730
val Loss: 0.2277 Acc: 0.8954

Epoch 17/24
----------
train Loss: 0.2609 Acc: 0.9098
val Loss: 0.2099 Acc: 0.9020

Epoch 18/24
----------
train Loss: 0.3285 Acc: 0.8525
val Loss: 0.2418 Acc: 0.8889

Epoch 19/24
----------
train Loss: 0.2846 Acc: 0.8852
val Loss: 0.2028 Acc: 0.9346

Epoch 20/24
----------
train Loss: 0.2061 Acc: 0.8893
val Loss: 0.2001 Acc: 0.9346

Epoch 21/24
----------
train Loss: 0.3256 Acc: 0.8525
val Loss: 0.2328 Acc: 0.9020

Epoch 22/24
----------
train Loss: 0.3051 Acc: 0.8689
val Loss: 0.2088 Acc: 0.9020

Epoch 23/24
----------
train Loss: 0.2973 Acc: 0.8811
val Loss: 0.1996 Acc: 0.9216

Epoch 24/24
----------
train Loss: 0.2526 Acc: 0.8852
val Loss: 0.1919 Acc: 0.9281

Training complete in 0m 35s
Best val Acc: 0.934641

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: 0.6799 Acc: 0.6189
val Loss: 0.6541 Acc: 0.7255

Epoch 1/24
----------
train Loss: 0.4532 Acc: 0.7705
val Loss: 0.2069 Acc: 0.9346

Epoch 2/24
----------
train Loss: 0.5362 Acc: 0.7541
val Loss: 0.1709 Acc: 0.9412

Epoch 3/24
----------
train Loss: 0.6783 Acc: 0.7131
val Loss: 0.3411 Acc: 0.8627

Epoch 4/24
----------
train Loss: 0.4486 Acc: 0.8033
val Loss: 0.2169 Acc: 0.9346

Epoch 5/24
----------
train Loss: 0.3330 Acc: 0.8484
val Loss: 0.2074 Acc: 0.9477

Epoch 6/24
----------
train Loss: 0.3896 Acc: 0.8443
val Loss: 0.2060 Acc: 0.9281

Epoch 7/24
----------
train Loss: 0.3333 Acc: 0.8566
val Loss: 0.2341 Acc: 0.9346

Epoch 8/24
----------
train Loss: 0.3304 Acc: 0.8811
val Loss: 0.1980 Acc: 0.9412

Epoch 9/24
----------
train Loss: 0.3043 Acc: 0.8811
val Loss: 0.2133 Acc: 0.9346

Epoch 10/24
----------
train Loss: 0.3349 Acc: 0.8566
val Loss: 0.2229 Acc: 0.9346

Epoch 11/24
----------
train Loss: 0.3538 Acc: 0.8566
val Loss: 0.2400 Acc: 0.9346

Epoch 12/24
----------
train Loss: 0.3040 Acc: 0.8525
val Loss: 0.1977 Acc: 0.9346

Epoch 13/24
----------
train Loss: 0.3172 Acc: 0.8402
val Loss: 0.2120 Acc: 0.9346

Epoch 14/24
----------
train Loss: 0.4178 Acc: 0.8279
val Loss: 0.2033 Acc: 0.9346

Epoch 15/24
----------
train Loss: 0.2817 Acc: 0.8689
val Loss: 0.1956 Acc: 0.9412

Epoch 16/24
----------
train Loss: 0.2974 Acc: 0.8484
val Loss: 0.2159 Acc: 0.9216

Epoch 17/24
----------
train Loss: 0.3350 Acc: 0.8566
val Loss: 0.2102 Acc: 0.9477

Epoch 18/24
----------
train Loss: 0.2745 Acc: 0.8975
val Loss: 0.2054 Acc: 0.9477

Epoch 19/24
----------
train Loss: 0.3022 Acc: 0.8811
val Loss: 0.2194 Acc: 0.9216

Epoch 20/24
----------
train Loss: 0.3221 Acc: 0.8402
val Loss: 0.2097 Acc: 0.9412

Epoch 21/24
----------
train Loss: 0.3578 Acc: 0.8443
val Loss: 0.1994 Acc: 0.9477

Epoch 22/24
----------
train Loss: 0.2608 Acc: 0.8893
val Loss: 0.1999 Acc: 0.9477

Epoch 23/24
----------
train Loss: 0.3085 Acc: 0.8730
val Loss: 0.2015 Acc: 0.9346

Epoch 24/24
----------
train Loss: 0.4056 Acc: 0.8238
val Loss: 0.2066 Acc: 0.9412

Training complete in 0m 27s
Best val Acc: 0.947712

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.