Tensor creation (torch.tensor, zeros, ones, rand) in PyTorch - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

What is the shape of the tensor created?

Consider the following PyTorch code that creates a tensor. What will be the shape of the resulting tensor?

PyTorch

import torch
x = torch.ones((3, 4))
shape = x.shape
print(shape)

A (4, 3)

B (4,)

C (3,)

D (3, 4)

💡 Hint

Remember that torch.ones takes a tuple for the shape.

The torch.ones function creates a tensor filled with ones with the shape given by the tuple (3, 4). So the shape is (3, 4).

❓ Predict Output

intermediate

What is the output of this tensor creation?

What will be the output of the following PyTorch code?

PyTorch

import torch
x = torch.tensor([[1, 2], [3, 4]], dtype=torch.float32)
print(x)

A tensor([[1., 2.], [3., 4.]])

B tensor([[1, 2], [3, 4]])

C tensor([1., 2., 3., 4.])

D tensor([[1, 2], [3, 4]], dtype=torch.int32)

💡 Hint

Check the dtype argument and how it affects the tensor values.

The tensor is created with dtype=torch.float32, so all values are floats and printed with a decimal point.

❓ Model Choice

advanced

Which tensor creation method produces a tensor with random values between 0 and 1?

You want to create a tensor of shape (2, 3) filled with random values between 0 and 1. Which PyTorch function should you use?

A torch.zeros((2, 3))

B torch.ones((2, 3))

C torch.rand((2, 3))

D torch.tensor([[0, 1], [0, 1], [0, 1]])

💡 Hint

Think about which function generates random numbers.

torch.rand creates a tensor with random values sampled uniformly from [0, 1).

❓ Metrics

advanced

How many elements are in the tensor created by torch.zeros((4, 5, 2))?

If you create a tensor with torch.zeros((4, 5, 2)), how many total elements does this tensor contain?

A 40

B 20

C 10

D 50

💡 Hint

Multiply the dimensions to find the total number of elements.

The tensor shape is (4, 5, 2). Multiply 4 * 5 * 2 = 40 elements total.

🔧 Debug

expert

What error does this code raise?

What error will the following PyTorch code raise when executed?

PyTorch

import torch
x = torch.tensor([1, 2, 3], dtype='float64')
y = torch.ones(3, 3)
z = x + y

A TypeError: dtype must be a torch.dtype, not str

B RuntimeError: The size of tensor a (3) must match the size of tensor b (3) at non-singleton dimension 1

C No error, z is a tensor of shape (3, 3)

D ValueError: operands could not be broadcast together with shapes (3,) (3,3)

💡 Hint

Check the shapes of x and y before adding.

x has shape (3,) and y has shape (3, 3). They cannot be added because their shapes are incompatible for broadcasting.