PyTorch Dataset Loader and Transformer

This function is used to load and transform data from a specified PyTorch dataset. It supports the CIFAR10 and MNIST datasets. The function returns a DataLoader object for batch data loading.

Technology Stack : PyTorch, torchvision

Code Type : Data Loading and Conversion Functions

Code Difficulty : Intermediate

                
                    
import torch
import torchvision.transforms as transforms
import torchvision.datasets as datasets

def load_and_transform_data(batch_size, dataset_name='CIFAR10'):
    """
    Load and transform data from a specified dataset.
    """
    # Define the dataset
    if dataset_name == 'CIFAR10':
        dataset = datasets.CIFAR10(root='./data', download=True, transform=transforms.ToTensor())
    elif dataset_name == 'MNIST':
        dataset = datasets.MNIST(root='./data', download=True, transform=transforms.ToTensor())
    else:
        raise ValueError("Unsupported dataset. Choose 'CIFAR10' or 'MNIST'.")

    # Create DataLoader
    data_loader = torch.utils.data.DataLoader(dataset, batch_size=batch_size, shuffle=True)

    return data_loader