FashionMNIST

Import torch

import torch

Exercises require PyTorch > 1.10.0

print(torch.version)

TODO: Setup device agnostic code

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

Import Libraries

import torchvision from torchvision import datasets from torch import nn from torchvision.transforms import ToTensor from torchmetrics import Accuracy import matplotlib.pyplot as plt from torch.utils.data import DataLoader

Create Datasets & Loaders

train_data = datasets.MNIST( root="data", train=True, download=True, transform=ToTensor(), target_transform=None )

test_data = datasets.MNIST( root="data", train=False, download=True, transform=ToTensor(), target_transform=None

BATCH_SIZE = 32 train_dataloader = DataLoader(train_data, batch_size=BATCH_SIZE, shuffle=True )

test_dataloader = DataLoader(test_data, batch_size=BATCH_SIZE, shuffle=False )

)

Model

class MNISTModelV0(nn.Module): def init(self, input_shape, hidden_units, output_shape): super().init() self.block_1 = nn.Sequential( nn.Conv2d(in_channels=input_shape, out_channels=hidden_units, kernel_size=3, stride=1, padding=1), nn.ReLU(), nn.Conv2d(in_channels=hidden_units, out_channels=hidden_units, kernel_size=3, stride=1, padding=1), nn.ReLU(), nn.MaxPool2d(kernel_size=2, stride=2) ) self.block_2 = nn.Sequential( nn.Conv2d(hidden_units, hidden_units, 3, padding=1), nn.ReLU(), nn.Conv2d(hidden_units, hidden_units, 3, padding=1), nn.ReLU(), nn.MaxPool2d(2) ) self.classifier = nn.Sequential( nn.Flatten(), nn.Linear(in_features=hidden_units77, out_features=output_shape) )

def forward(self, x: torch.Tensor):
    x = self.block_1(x)
    x = self.block_2(x)
    x = self.classifier(x)
    return x

torch.manual_seed(42) mnist_model = MNISTModelV0(input_shape=1, hidden_units=10, output_shape=len(classes)).to(device) mnist_model

Training and Testing

from tqdm.auto import tqdm loss_fn = nn.CrossEntropyLoss() accuracy_fn = Accuracy(task='multiclass', num_classes=len(classes)) optimizer= torch.optim.SGD(mnist_model.parameters(), lr=0.1)

def train_step(model, data_loader, loss_fn, optimizer, accuracy_fn, device): train_loss, train_acc = 0, 0 model.to(device) for batch, (X,y) in enumerate(data_loader): X, y = X.to(device), y.to(device) y_pred = model(X) loss = loss_fn(y_pred, y) train_loss += loss acc = accuracy_fn(y_pred.argmax(dim=1), y) train_acc += acc optimizer.zero_grad() loss.backward() optimizer.step() train_loss /= len(data_loader) train_acc /= len(data_loader) print(f"Train loss: {train_loss:.5f} | Train accuracy: {train_acc:.2f}%")

def test_step(model, data_loader, loss_fn, accuracy_fn, device): test_loss, test_acc = 0, 0 model.to(device) model.eval() with torch.inference_mode(): for (X,y) in data_loader: X, y = X.to(device), y.to(device) y_pred = model(X) test_loss += loss_fn(y_pred, y) test_acc += accuracy_fn(y_pred.argmax(dim=1), y) test_loss /= len(data_loader) test_acc /= len(data_loader) print(f"Test loss: {test_loss:.5f} | Test accuracy: {test_acc:.2f}%")

epochs = 5

for epoch in tqdm(range(epochs)): print(f"Epoch: {epoch}\n---------") train_step(data_loader=train_dataloader, model=mnist_model, loss_fn=loss_fn, optimizer=optimizer, accuracy_fn=accuracy_fn, device=device ) test_step(data_loader=test_dataloader, model=mnist_model, loss_fn=loss_fn, accuracy_fn=accuracy_fn, device=device )