NumPy bridge (from_numpy, numpy) in PyTorch - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to convert a NumPy array to a PyTorch tensor using the correct function.

PyTorch

import numpy as np
import torch
np_array = np.array([1, 2, 3])
tensor = torch.[1](np_array)

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

Atensor

Bfrom_numpy

Cas_tensor

Dfrom_array

Attempts:

3 left

💡 Hint

Common Mistakes

Using torch.tensor() which copies data instead of sharing memory.

Using a non-existent function like torch.from_array.

The torch.from_numpy() function converts a NumPy array into a PyTorch tensor sharing the same memory.

2fill in blank

medium

Complete the code to convert a PyTorch tensor back to a NumPy array.

PyTorch

import torch
import numpy as np
tensor = torch.tensor([4, 5, 6])
np_array = tensor.[1]()

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

Atoarray

Bto_numpy

Casarray

Dnumpy

Attempts:

3 left

💡 Hint

Common Mistakes

Using .to_numpy() which does not exist in PyTorch.

Using .asarray() which is a NumPy function, not a tensor method.

The .numpy() method converts a PyTorch tensor to a NumPy array sharing the same memory.

3fill in blank

hard

Fix the error in the code to correctly convert a NumPy array to a PyTorch tensor and modify the tensor without affecting the original array.

PyTorch

import numpy as np
import torch
np_array = np.array([7, 8, 9])
tensor = torch.[1](np_array).clone()
tensor[0] = 100

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

Aas_tensor

Btensor

Cfrom_numpy

Dfrom_array

Attempts:

3 left

💡 Hint

Common Mistakes

Using torch.tensor() which copies data but does not share memory.

Not cloning the tensor, so changes affect the original NumPy array.

torch.from_numpy() creates a tensor sharing memory with the NumPy array. Using .clone() makes a copy so changes to the tensor don't affect the original array.

4fill in blank

hard

Fill both blanks to create a NumPy array from a PyTorch tensor and then modify the tensor without changing the NumPy array.

PyTorch

import torch
import numpy as np
tensor = torch.tensor([10, 20, 30])
np_array = tensor.[1]()
tensor = tensor.[2]()
tensor[1] = 200

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

Anumpy

Bclone

Cdetach

Dcopy

Attempts:

3 left

💡 Hint

Common Mistakes

Using .detach() instead of .clone(), which does not copy data.

Using .copy() which is not a tensor method.

The .numpy() method converts the tensor to a NumPy array sharing memory. Cloning the tensor creates a new tensor so changes don't affect the original NumPy array.

5fill in blank

hard

Fill all three blanks to create a PyTorch tensor from a NumPy array, convert it back to NumPy, and then modify the tensor without affecting the NumPy array.

PyTorch

import numpy as np
import torch
np_array = np.array([1, 3, 5])
tensor = torch.[1](np_array)
np_array2 = tensor.[2]()
tensor = tensor.[3]()
tensor[2] = 999

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

Afrom_numpy

Bnumpy

Cclone

Dtensor

Attempts:

3 left

💡 Hint

Common Mistakes

Using torch.tensor() instead of from_numpy.

Not cloning the tensor before modifying it.

Using .detach() instead of .clone().

torch.from_numpy() creates a tensor sharing memory with the NumPy array. The .numpy() method converts the tensor back to a NumPy array sharing memory. Cloning the tensor creates a new tensor so changes don't affect the NumPy array.