Built-in datasets (torchvision.datasets) in PyTorch - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the CIFAR10 dataset from torchvision.datasets.

PyTorch

from torchvision.datasets import [1]

Drag options to blanks, or click blank then click option'

AFashionMNIST

BMNIST

CCIFAR10

DImageNet

Attempts:

3 left

💡 Hint

Common Mistakes

Importing MNIST instead of CIFAR10

Using ImageNet which is not in torchvision.datasets by default

Confusing FashionMNIST with CIFAR10

The CIFAR10 dataset is imported from torchvision.datasets using from torchvision.datasets import CIFAR10.

2fill in blank

medium

Complete the code to download the CIFAR10 training dataset with transforms.

PyTorch

train_dataset = CIFAR10(root='./data', train=[1], download=True, transform=None)

Drag options to blanks, or click blank then click option'

ATrue

BFalse

C'train'

D1

Attempts:

3 left

💡 Hint

Common Mistakes

Setting train to False loads the test set

Using string 'train' instead of boolean True

Passing 1 instead of True

Setting train=True loads the training split of CIFAR10.

3fill in blank

hard

Fix the error in the code to apply a ToTensor transform to the CIFAR10 dataset.

PyTorch

from torchvision.transforms import [1]
train_dataset = CIFAR10(root='./data', train=True, download=True, transform=ToTensor())

Drag options to blanks, or click blank then click option'

ACompose

BNormalize

CResize

DToTensor

Attempts:

3 left

💡 Hint

Common Mistakes

Importing Compose instead of ToTensor

Using Normalize without ToTensor first

Trying to resize images instead of converting to tensor

The ToTensor transform converts images to PyTorch tensors and must be imported from torchvision.transforms.

4fill in blank

hard

Fill both blanks to create a DataLoader for the training dataset with batch size 32 and shuffling enabled.

PyTorch

from torch.utils.data import DataLoader
train_loader = DataLoader(train_dataset, batch_size=[1], shuffle=[2])

Drag options to blanks, or click blank then click option'

A32

BTrue

CFalse

D64

Attempts:

3 left

💡 Hint

Common Mistakes

Setting shuffle to False which disables random order

Using batch size 64 which is valid but not the intended answer

Confusing batch_size and shuffle parameters

Batch size 32 and shuffle=True are common settings for training data loaders.

5fill in blank

hard

Fill all three blanks to create a dictionary comprehension that maps class indices to class names from CIFAR10.

PyTorch

class_to_name = [1]: [2] for [3] in range(len(train_dataset.classes))}

Drag options to blanks, or click blank then click option'

Ai

Btrain_dataset.classes[i]

Didx

Attempts:

3 left

💡 Hint

Common Mistakes

Using idx instead of i inconsistently

Swapping keys and values

Using range without len()

This dictionary comprehension uses the index i to map to the class name at train_dataset.classes[i].