GPU tensors (to, cuda) in PyTorch - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to move the tensor to GPU using the recommended method.

PyTorch

import torch
x = torch.tensor([1, 2, 3])
x_gpu = x.[1]('cuda')

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

Acpu

Bcuda

Cdevice

Dto

Attempts:

3 left

💡 Hint

Common Mistakes

Using .cuda() directly instead of .to('cuda')

Using .cpu() which moves tensor to CPU

The to method is the recommended way to move tensors to a device like GPU by specifying 'cuda'.

2fill in blank

medium

Complete the code to check if CUDA is available before moving the tensor to GPU.

PyTorch

import torch
x = torch.tensor([4, 5, 6])
if torch.cuda.is_available():
    x_gpu = x.[1]('cuda')

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

Acuda

Bto

Cdevice

Dcopy

Attempts:

3 left

💡 Hint

Common Mistakes

Using .cuda() without checking availability

Using .copy() which is not a tensor method

Using to('cuda') moves the tensor to GPU if CUDA is available.

3fill in blank

hard

Fix the error in the code to move the tensor to GPU correctly.

PyTorch

import torch
x = torch.tensor([7, 8, 9])
x_gpu = x.[1]()

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

Ato

Bdevice

Ccuda

Dto('cuda')

Attempts:

3 left

💡 Hint

Common Mistakes

Using to without arguments

Using to('cuda') with parentheses inside blank

The cuda() method moves the tensor to the default CUDA device without arguments.

4fill in blank

hard

Fill both blanks to create a tensor on GPU and then move it back to CPU.

PyTorch

import torch
x_gpu = torch.tensor([10, 11, 12], device=[1])
x_cpu = x_gpu.[2]()

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

A'cuda'

Bcpu

Dcuda

Attempts:

3 left

💡 Hint

Common Mistakes

Using device='cpu' to create GPU tensor

Using to('cpu') instead of cpu() method

Specify device='cuda' to create tensor on GPU, then use cpu() to move it back to CPU.

5fill in blank

hard

Fill all three blanks to create a tensor on CPU, then move it to GPU.

PyTorch

import torch
x = torch.tensor([13, 14, 15], device=[1])
x_gpu = x.[2]([3])

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

A'cpu'

Bto

C'cuda'

Dcuda

Attempts:

3 left

💡 Hint

Common Mistakes

Using device='cuda' initially

Using cuda() method with argument

Using to without argument

Create tensor on CPU with device='cpu', move to GPU with to('cuda').