Write a Simple Image Classifier (Using package)

Writing a simple image classifier using PyTorch during an interview demonstrates your understanding of deep learning concepts and your proficiency with PyTorch. Let's create a basic classifier that can categorize images from the CIFAR-10 dataset, which is a common benchmark in machine learning for image recognition. This dataset consists of 60,000 32x32 color images in 10 classes, with 6,000 images per class.

Here's a step-by-step guide to creating a simple image classifier using PyTorch:

Step 1: Import Necessary Libraries

First, you'll need to import PyTorch and other necessary libraries.

import torch
import torchvision
import torchvision.transforms as transforms
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim

Step 2: Load and Normalize CIFAR-10 Data

Use the torchvision package to load the CIFAR-10 dataset and normalize it.

transform = transforms.Compose(
    [transforms.ToTensor(),
     transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])

trainset = torchvision.datasets.CIFAR10(root='./data', train=True,
                                        download=True, transform=transform)
trainloader = torch.utils.data.DataLoader(trainset, batch_size=4,
                                          shuffle=True, num_workers=2)

testset = torchvision.datasets.CIFAR10(root='./data', train=False,
                                       download=True, transform=transform)
testloader = torch.utils.data.DataLoader(testset, batch_size=4,
                                         shuffle=False, num_workers=2)

classes = ('plane', 'car', 'bird', 'cat',
           'deer', 'dog', 'frog', 'horse', 'ship', 'truck')

Step 3: Define a Convolutional Neural Network

Define a simple CNN architecture.

class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.conv1 = nn.Conv2d(3, 6, 5)
        self.pool = nn.MaxPool2d(2, 2)
        self.conv2 = nn.Conv2d(6, 16, 5)
        self.fc1 = nn.Linear(16 * 5 * 5, 120)
        self.fc2 = nn.Linear(120, 84)
        self.fc3 = nn.Linear(84, 10)

    def forward(self, x):
        x = self.pool(F.relu(self.conv1(x)))
        x = self.pool(F.relu(self.conv2(x)))
        x = x.view(-1, 16 * 5 * 5)
        x = F.relu(self.fc1(x))
        x = F.relu(self.fc2(x))
        x = self.fc3(x)
        return x

net = Net()

To calculate the spatial size of the feature map after applying a convolutional layer followed by max pooling, you need to consider the formula for calculating the output size of a convolutional layer and the pooling operation.

For the convolutional layer:

• Output height \( H_{out} \) and width \( W_{out} \) can be calculated using the formula:
  
  
  
  where \( H
  {in} \) and \( W_{in} \) are the input height and width, respectively, and padding is the amount of zero-padding added to the input.

For the max pooling layer:

• If kernel_size and stride are the same, the output size is reduced by half.

Given that the input to the second convolutional layer (self.conv2) has a spatial size of \( 7 \times 7 \) after the max pooling operation, we can infer the input size of the second convolutional layer and then calculate the spatial size of the feature map after applying self.conv2.

Step 4: Define a Loss Function and Optimizer

Specify a loss function and an optimizer for training the network.

criterion = nn.CrossEntropyLoss()
optimizer = optim.SGD(net.parameters(), lr=0.001, momentum=0.9)

Step 5: Train the Network

Train the network on the training data.

for epoch in range(2):  # loop over the dataset multiple times

    running_loss = 0.0
    for i, data in enumerate(trainloader, 0):
        # get the inputs; data is a list of [inputs, labels]
        inputs, labels = data

        # zero the parameter gradients
        optimizer.zero_grad()

        # forward + backward + optimize
        outputs = net(inputs)
        loss = criterion(outputs, labels)
        loss.backward()
        optimizer.step()

        # print statistics
        running_loss += loss.item()
        if i % 2000 == 1999:    # print every 2000 mini-batches
            print('[%d, %5d] loss: %.3f' %
                  (epoch + 1, i + 1, running_loss / 2000))
            running_loss = 0.0

print('Finished Training')

Step 6: Test the Network on the Test Data

After training the network, test it on the test data to evaluate its performance.

correct = 0
total = 0
with torch.no_grad():
    for data in testloader:
        images, labels = data
        outputs = net(images)
        _, predicted = torch.max(outputs.data, 1)
        total += labels.size(0)
        correct += (predicted == labels).sum().item()

print('Accuracy of the network on the 10000 test images: %d %%' % (
    100 * correct / total))

This code provides a basic framework for an image classifier using PyTorch. During an interview, you can explain each step to demonstrate