Transfer Learning for Computer Vision Tutorial

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 initializaion, 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

from __future__ import print_function, division

import torch
import torch.nn as nn
import torch.optim as optim
from torch.optim import lr_scheduler
import numpy as np
import torchvision
from torchvision import datasets, models, transforms
import matplotlib.pyplot as plt
import time
import os
import copy

plt.ion()   # interactive mode

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

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

Visualize a few images

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

def imshow(inp, title=None):
    """Imshow 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()

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

            # deep copy the model
            if phase == 'val' and epoch_acc > best_acc:
                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

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('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)

Finetuning the convnet

Load a pretrained model and reset final fully connected layer.

model_ft = models.resnet18(pretrained=True)
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)

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)

Out:

Epoch 0/24
----------
train Loss: 0.6567 Acc: 0.6967
val Loss: 0.3229 Acc: 0.8889

Epoch 1/24
----------
train Loss: 0.5948 Acc: 0.7336
val Loss: 0.1969 Acc: 0.9412

Epoch 2/24
----------
train Loss: 0.4150 Acc: 0.8402
val Loss: 0.2588 Acc: 0.9150

Epoch 3/24
----------
train Loss: 0.4527 Acc: 0.8238
val Loss: 0.2312 Acc: 0.9346

Epoch 4/24
----------
train Loss: 0.5157 Acc: 0.8156
val Loss: 0.3275 Acc: 0.8693

Epoch 5/24
----------
train Loss: 0.5087 Acc: 0.7623
val Loss: 0.2670 Acc: 0.9020

Epoch 6/24
----------
train Loss: 0.4585 Acc: 0.8156
val Loss: 0.1689 Acc: 0.9346

Epoch 7/24
----------
train Loss: 0.3337 Acc: 0.8402
val Loss: 0.1575 Acc: 0.9281

Epoch 8/24
----------
train Loss: 0.2701 Acc: 0.8730
val Loss: 0.1475 Acc: 0.9412

Epoch 9/24
----------
train Loss: 0.3255 Acc: 0.8443
val Loss: 0.1745 Acc: 0.9281

Epoch 10/24
----------
train Loss: 0.2576 Acc: 0.8893
val Loss: 0.1607 Acc: 0.9346

Epoch 11/24
----------
train Loss: 0.3904 Acc: 0.8443
val Loss: 0.1533 Acc: 0.9608

Epoch 12/24
----------
train Loss: 0.2795 Acc: 0.8934
val Loss: 0.1522 Acc: 0.9477

Epoch 13/24
----------
train Loss: 0.2701 Acc: 0.8811
val Loss: 0.1553 Acc: 0.9412

Epoch 14/24
----------
train Loss: 0.3131 Acc: 0.8689
val Loss: 0.1761 Acc: 0.9477

Epoch 15/24
----------
train Loss: 0.3001 Acc: 0.8525
val Loss: 0.1406 Acc: 0.9412

Epoch 16/24
----------
train Loss: 0.3006 Acc: 0.8566
val Loss: 0.1473 Acc: 0.9281

Epoch 17/24
----------
train Loss: 0.3371 Acc: 0.8402
val Loss: 0.1472 Acc: 0.9542

Epoch 18/24
----------
train Loss: 0.3147 Acc: 0.8811
val Loss: 0.1381 Acc: 0.9542

Epoch 19/24
----------
train Loss: 0.3374 Acc: 0.8402
val Loss: 0.1452 Acc: 0.9346

Epoch 20/24
----------
train Loss: 0.2309 Acc: 0.9098
val Loss: 0.1732 Acc: 0.9346

Epoch 21/24
----------
train Loss: 0.2467 Acc: 0.8811
val Loss: 0.1485 Acc: 0.9346

Epoch 22/24
----------
train Loss: 0.2280 Acc: 0.9016
val Loss: 0.1526 Acc: 0.9346

Epoch 23/24
----------
train Loss: 0.3162 Acc: 0.8648
val Loss: 0.1422 Acc: 0.9346

Epoch 24/24
----------
train Loss: 0.2524 Acc: 0.9016
val Loss: 0.1450 Acc: 0.9477

Training complete in 0m 48s
Best val Acc: 0.960784

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(pretrained=True)
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)

Out:

Epoch 0/24
----------
train Loss: 0.6548 Acc: 0.6434
val Loss: 0.3279 Acc: 0.8627

Epoch 1/24
----------
train Loss: 0.4907 Acc: 0.7787
val Loss: 0.3273 Acc: 0.8366

Epoch 2/24
----------
train Loss: 0.4351 Acc: 0.8033
val Loss: 0.2670 Acc: 0.8758

Epoch 3/24
----------
train Loss: 0.5002 Acc: 0.8033
val Loss: 0.1670 Acc: 0.9542

Epoch 4/24
----------
train Loss: 0.5262 Acc: 0.7828
val Loss: 0.1780 Acc: 0.9542

Epoch 5/24
----------
train Loss: 0.4103 Acc: 0.7992
val Loss: 0.2593 Acc: 0.8889

Epoch 6/24
----------
train Loss: 0.6280 Acc: 0.7377
val Loss: 0.1740 Acc: 0.9542

Epoch 7/24
----------
train Loss: 0.3855 Acc: 0.8238
val Loss: 0.1810 Acc: 0.9542

Epoch 8/24
----------
train Loss: 0.4041 Acc: 0.8033
val Loss: 0.1905 Acc: 0.9346

Epoch 9/24
----------
train Loss: 0.4469 Acc: 0.8156
val Loss: 0.1858 Acc: 0.9346

Epoch 10/24
----------
train Loss: 0.3172 Acc: 0.8689
val Loss: 0.1610 Acc: 0.9608

Epoch 11/24
----------
train Loss: 0.3104 Acc: 0.8402
val Loss: 0.1924 Acc: 0.9346

Epoch 12/24
----------
train Loss: 0.4036 Acc: 0.8361
val Loss: 0.1846 Acc: 0.9477

Epoch 13/24
----------
train Loss: 0.3462 Acc: 0.8361
val Loss: 0.1869 Acc: 0.9281

Epoch 14/24
----------
train Loss: 0.4521 Acc: 0.8033
val Loss: 0.1832 Acc: 0.9346

Epoch 15/24
----------
train Loss: 0.3577 Acc: 0.8238
val Loss: 0.1862 Acc: 0.9608

Epoch 16/24
----------
train Loss: 0.3645 Acc: 0.8730
val Loss: 0.1906 Acc: 0.9542

Epoch 17/24
----------
train Loss: 0.3366 Acc: 0.8607
val Loss: 0.1729 Acc: 0.9477

Epoch 18/24
----------
train Loss: 0.3793 Acc: 0.8279
val Loss: 0.2263 Acc: 0.8954

Epoch 19/24
----------
train Loss: 0.3282 Acc: 0.8607
val Loss: 0.1634 Acc: 0.9608

Epoch 20/24
----------
train Loss: 0.2573 Acc: 0.8975
val Loss: 0.2018 Acc: 0.9346

Epoch 21/24
----------
train Loss: 0.3195 Acc: 0.8893
val Loss: 0.1804 Acc: 0.9542

Epoch 22/24
----------
train Loss: 0.3004 Acc: 0.8770
val Loss: 0.1810 Acc: 0.9608

Epoch 23/24
----------
train Loss: 0.2837 Acc: 0.8893
val Loss: 0.1755 Acc: 0.9608

Epoch 24/24
----------
train Loss: 0.4070 Acc: 0.8156
val Loss: 0.2219 Acc: 0.9085

Training complete in 0m 37s
Best val Acc: 0.960784

visualize_model(model_conv)

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.