Data loading with torchvision in Computer Vision - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the torchvision datasets module.

Computer Vision

from torchvision import [1]

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

Amodels

Bdatasets

Ctransforms

Dutils

Attempts:

3 left

💡 Hint

Common Mistakes

Importing 'models' instead of 'datasets'.

Importing 'transforms' when datasets are needed.

The datasets module in torchvision provides easy access to popular vision datasets.

2fill in blank

medium

Complete the code to load the MNIST training dataset with torchvision.

Computer Vision

train_data = datasets.MNIST(root='./data', train=[1], download=True)

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

A1

BFalse

C'yes'

DTrue

Attempts:

3 left

💡 Hint

Common Mistakes

Using string 'yes' instead of boolean True.

Using 1 instead of True.

Setting train=True loads the training split of MNIST.

3fill in blank

hard

Fix the error in the code to apply a transform that converts images to tensors.

Computer Vision

from torchvision import transforms

transform = transforms.[1]()
train_data = datasets.MNIST(root='./data', train=True, download=True, transform=transform)

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

AToTensor

Bto_tensor

CtoTensor

DTensor

Attempts:

3 left

💡 Hint

Common Mistakes

Using lowercase 'to_tensor' which does not exist.

Using 'Tensor' which is not a transform class.

The correct transform class is ToTensor with capital T and capital T in Tensor.

4fill in blank

hard

Fill both blanks to create a DataLoader for the training data with batch size 64 and shuffling enabled.

Computer Vision

from torch.utils.data import DataLoader

train_loader = DataLoader(train_data, batch_size=[1], shuffle=[2])

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

A64

BTrue

C32

DFalse

Attempts:

3 left

💡 Hint

Common Mistakes

Using batch size 32 instead of 64.

Setting shuffle to False which disables randomization.

Batch size 64 is common for training, and shuffle=True randomizes data order each epoch.

5fill in blank

hard

Fill all three blanks to create a transform pipeline that resizes images to 28x28, converts to tensor, and normalizes with mean 0.5 and std 0.5.

Computer Vision

transform = transforms.Compose([
    transforms.Resize(([1], [2])),
    transforms.ToTensor(),
    transforms.Normalize(mean=[[3]], std=[0.5])
])

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

A28

B0.5

C32

D0.0

Attempts:

3 left

💡 Hint

Common Mistakes

Using 32 instead of 28 for resize dimensions.

Using 0.0 for mean which is not balanced normalization.

Images are resized to 28x28 pixels, normalized with mean 0.5 and std 0.5 for balanced input.