Why tensors are PyTorch's core data structure - Test Your Understanding

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create a tensor from a list.

PyTorch

import torch

my_tensor = torch.[1]([1, 2, 3, 4])
print(my_tensor)

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

Alist

Barray

Cmatrix

Dtensor

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'array' instead of 'tensor' causes an error.

Trying to use 'list' is invalid in PyTorch.

The torch.tensor function creates a tensor from a Python list.

2fill in blank

medium

Complete the code to get the shape of a tensor.

PyTorch

import torch

x = torch.tensor([[1, 2], [3, 4], [5, 6]])
shape = x.[1]
print(shape)

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

Alength

Bshape

Csize

Ddim

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'size' as a method requires parentheses.

Using 'length' is not valid for tensors.

The shape attribute returns the dimensions of the tensor as a tuple.

3fill in blank

hard

Fix the error in the code to perform element-wise addition of two tensors.

PyTorch

import torch

a = torch.tensor([1, 2, 3])
b = torch.tensor([4, 5, 6])
result = a [1] b
print(result)

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

A-

B*

C+

D/

Attempts:

3 left

💡 Hint

Common Mistakes

Using '-' subtracts instead of adds.

Using '*' multiplies instead of adds.

The + operator performs element-wise addition between tensors.

4fill in blank

hard

Fill both blanks to create a tensor of zeros with shape (3, 4) and type float.

PyTorch

import torch

zeros_tensor = torch.[1]((3, 4), dtype=torch.[2])
print(zeros_tensor)

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

Azeros

Bones

Cint32

Dfloat32

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'ones' creates a tensor of ones.

Using 'int32' sets integer type instead of float.

torch.zeros creates a tensor filled with zeros. The dtype=torch.float32 sets the data type to float.

5fill in blank

hard

Fill all three blanks to create a tensor from a list, move it to GPU if available, and print its device.

PyTorch

import torch

x = torch.[1]([1, 2, 3])
if torch.cuda.is_available():
    x = x.[2]('cuda')
print(x.[3])

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

Atensor

Bto

Cdevice

Dcuda

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'cuda' as a method instead of a string argument.

Trying to print 'to' instead of 'device'.

torch.tensor creates the tensor. to('cuda') moves it to GPU. device shows where the tensor is stored.