Note
Click here to download the full example code
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 0x7f98821a45e0>
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])
![['ants', 'bees', 'bees', 'ants']](../_images/sphx_glr_transfer_learning_tutorial_001.png)
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]
88%|########8 | 39.5M/44.7M [00:00<00:00, 414MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 415MB/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: 0.5415 Acc: 0.7418
val Loss: 0.2128 Acc: 0.9281
Epoch 1/24
----------
train Loss: 0.5028 Acc: 0.7787
val Loss: 0.4516 Acc: 0.8235
Epoch 2/24
----------
train Loss: 0.7671 Acc: 0.7254
val Loss: 0.1876 Acc: 0.9477
Epoch 3/24
----------
train Loss: 0.4618 Acc: 0.8156
val Loss: 0.5527 Acc: 0.7778
Epoch 4/24
----------
train Loss: 0.3962 Acc: 0.8156
val Loss: 0.1940 Acc: 0.9346
Epoch 5/24
----------
train Loss: 0.4974 Acc: 0.7951
val Loss: 0.2170 Acc: 0.9281
Epoch 6/24
----------
train Loss: 0.4857 Acc: 0.7746
val Loss: 0.2192 Acc: 0.9216
Epoch 7/24
----------
train Loss: 0.2680 Acc: 0.8934
val Loss: 0.1946 Acc: 0.9150
Epoch 8/24
----------
train Loss: 0.4014 Acc: 0.8402
val Loss: 0.1807 Acc: 0.9281
Epoch 9/24
----------
train Loss: 0.2858 Acc: 0.8811
val Loss: 0.1940 Acc: 0.9150
Epoch 10/24
----------
train Loss: 0.3770 Acc: 0.8197
val Loss: 0.2452 Acc: 0.9150
Epoch 11/24
----------
train Loss: 0.3201 Acc: 0.8566
val Loss: 0.1855 Acc: 0.9346
Epoch 12/24
----------
train Loss: 0.3132 Acc: 0.8730
val Loss: 0.1871 Acc: 0.9216
Epoch 13/24
----------
train Loss: 0.2619 Acc: 0.8852
val Loss: 0.1720 Acc: 0.9281
Epoch 14/24
----------
train Loss: 0.2290 Acc: 0.9016
val Loss: 0.1781 Acc: 0.9281
Epoch 15/24
----------
train Loss: 0.3000 Acc: 0.8770
val Loss: 0.1786 Acc: 0.9542
Epoch 16/24
----------
train Loss: 0.1665 Acc: 0.9549
val Loss: 0.1725 Acc: 0.9346
Epoch 17/24
----------
train Loss: 0.2258 Acc: 0.9057
val Loss: 0.1726 Acc: 0.9346
Epoch 18/24
----------
train Loss: 0.3041 Acc: 0.8648
val Loss: 0.1778 Acc: 0.9216
Epoch 19/24
----------
train Loss: 0.2456 Acc: 0.9139
val Loss: 0.1840 Acc: 0.9216
Epoch 20/24
----------
train Loss: 0.2714 Acc: 0.8852
val Loss: 0.1854 Acc: 0.9281
Epoch 21/24
----------
train Loss: 0.3034 Acc: 0.8689
val Loss: 0.1767 Acc: 0.9281
Epoch 22/24
----------
train Loss: 0.3413 Acc: 0.8648
val Loss: 0.1760 Acc: 0.9281
Epoch 23/24
----------
train Loss: 0.3145 Acc: 0.8893
val Loss: 0.1849 Acc: 0.9412
Epoch 24/24
----------
train Loss: 0.2270 Acc: 0.9180
val Loss: 0.1748 Acc: 0.9346
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: 0.6382 Acc: 0.6967
val Loss: 0.2202 Acc: 0.9085
Epoch 1/24
----------
train Loss: 0.4366 Acc: 0.8033
val Loss: 0.3129 Acc: 0.8562
Epoch 2/24
----------
train Loss: 0.4774 Acc: 0.7951
val Loss: 0.3238 Acc: 0.8693
Epoch 3/24
----------
train Loss: 0.5183 Acc: 0.7664
val Loss: 0.2770 Acc: 0.8954
Epoch 4/24
----------
train Loss: 0.4923 Acc: 0.7705
val Loss: 0.1775 Acc: 0.9346
Epoch 5/24
----------
train Loss: 0.3740 Acc: 0.8525
val Loss: 0.1809 Acc: 0.9477
Epoch 6/24
----------
train Loss: 0.4162 Acc: 0.8443
val Loss: 0.2364 Acc: 0.9281
Epoch 7/24
----------
train Loss: 0.3546 Acc: 0.8607
val Loss: 0.1915 Acc: 0.9412
Epoch 8/24
----------
train Loss: 0.3522 Acc: 0.8402
val Loss: 0.1860 Acc: 0.9477
Epoch 9/24
----------
train Loss: 0.4244 Acc: 0.8279
val Loss: 0.2172 Acc: 0.9281
Epoch 10/24
----------
train Loss: 0.3660 Acc: 0.8525
val Loss: 0.2034 Acc: 0.9477
Epoch 11/24
----------
train Loss: 0.3069 Acc: 0.8525
val Loss: 0.2174 Acc: 0.9216
Epoch 12/24
----------
train Loss: 0.2625 Acc: 0.8934
val Loss: 0.1836 Acc: 0.9412
Epoch 13/24
----------
train Loss: 0.3361 Acc: 0.8566
val Loss: 0.1939 Acc: 0.9542
Epoch 14/24
----------
train Loss: 0.3465 Acc: 0.8484
val Loss: 0.1809 Acc: 0.9542
Epoch 15/24
----------
train Loss: 0.2952 Acc: 0.8689
val Loss: 0.2085 Acc: 0.9281
Epoch 16/24
----------
train Loss: 0.3063 Acc: 0.8811
val Loss: 0.1769 Acc: 0.9542
Epoch 17/24
----------
train Loss: 0.3667 Acc: 0.8484
val Loss: 0.1860 Acc: 0.9477
Epoch 18/24
----------
train Loss: 0.3548 Acc: 0.8320
val Loss: 0.1815 Acc: 0.9412
Epoch 19/24
----------
train Loss: 0.2830 Acc: 0.8566
val Loss: 0.1972 Acc: 0.9477
Epoch 20/24
----------
train Loss: 0.2709 Acc: 0.8770
val Loss: 0.1955 Acc: 0.9346
Epoch 21/24
----------
train Loss: 0.3727 Acc: 0.8484
val Loss: 0.1840 Acc: 0.9542
Epoch 22/24
----------
train Loss: 0.3630 Acc: 0.8279
val Loss: 0.1791 Acc: 0.9542
Epoch 23/24
----------
train Loss: 0.3171 Acc: 0.8689
val Loss: 0.2002 Acc: 0.9412
Epoch 24/24
----------
train Loss: 0.3524 Acc: 0.8402
val Loss: 0.1895 Acc: 0.9477
Training complete in 0m 28s
Best val Acc: 0.954248
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.
Total running time of the script: ( 1 minutes 4.527 seconds)