DataLoader basics in PyTorch - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create a DataLoader for the dataset with batch size 4.

PyTorch

from torch.utils.data import DataLoader

dataloader = DataLoader(dataset, batch_size=[1])

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

A4

B16

C1

D8

Attempts:

3 left

💡 Hint

Common Mistakes

Using batch_size larger than dataset size causes errors.

Forgetting to set batch_size causes default of 1.

The batch size controls how many samples are loaded per batch. Here, batch_size=4 means each batch has 4 samples.

2fill in blank

medium

Complete the code to shuffle the dataset in the DataLoader.

PyTorch

dataloader = DataLoader(dataset, batch_size=4, shuffle=[1])

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

AFalse

BTrue

C0

DNone

Attempts:

3 left

💡 Hint

Common Mistakes

Setting shuffle to False causes data to be loaded in the same order every time.

Using None or 0 for shuffle causes errors.

Setting shuffle=True randomizes the order of samples each epoch, which helps training.

3fill in blank

hard

Fix the error in the code to correctly iterate over the DataLoader batches.

PyTorch

for [1] in dataloader:
    inputs, labels = batch
    print(inputs.shape, labels.shape)

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

Alabels

Bdataloader

Cinputs

Dbatch

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'inputs' or 'labels' as loop variable causes unpacking errors.

Using 'dataloader' as loop variable causes infinite loop.

Each iteration returns a batch, so the loop variable should be 'batch' to unpack inputs and labels.

4fill in blank

hard

Fill both blanks to create a DataLoader with batch size 8 and no shuffling.

PyTorch

dataloader = DataLoader(dataset, batch_size=[1], shuffle=[2])

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

A8

BTrue

CFalse

D16

Attempts:

3 left

💡 Hint

Common Mistakes

Using shuffle=True when no shuffling is wanted.

Setting batch size too large for dataset.

Batch size 8 loads 8 samples per batch. shuffle=False means data order stays the same.

5fill in blank

hard

Fill all three blanks to create a DataLoader with batch size 16, shuffling enabled, and 2 worker threads.

PyTorch

dataloader = DataLoader(dataset, batch_size=[1], shuffle=[2], num_workers=[3])

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

A4

BTrue

C2

D16

Attempts:

3 left

💡 Hint

Common Mistakes

Setting num_workers to 0 disables multi-threading.

Using shuffle=False when random order is needed.

Batch size 16 loads 16 samples per batch. shuffle=True randomizes data order. num_workers=2 uses two threads to load data faster.