np.resize() returns a new array with a specified shape, repeating or truncating the original data as needed to exactly fill that shape, regardless of whether the total element count matches.

Syntax

np.resize(arr, new_shape)

Deep Dive Course

Unlike np.reshape(), which requires the new shape to have exactly the same total element count as the original array, np.resize() will happily produce a shape with more or fewer elements: if the new shape needs more elements than the original array has, resize() repeats the original data from the beginning to fill the extra space; if it needs fewer, resize() simply truncates the original data. This makes np.resize() a fundamentally different operation from reshape(), not just a more flexible version of it, and it's easy to reach for it by mistake when reshape() was actually intended.

1Understanding np.resize()

💡

Don't confuse np.resize() with reshape() — resize() can silently repeat or discard data to force-fit a new shape, which is rarely what you want if you actually just meant to reorganize an array's existing elements without changing them.

editor.html

import numpy as np

arr = np.array([1, 2, 3])
resized = np.resize(arr, (2, 4))
print(resized)

localhost:3000

2Practical Example

Here is a real-world application of np.resize() showing how it is used in production NumPy code.

editor.html

import numpy as np

arr = np.array([1, 2, 3, 4, 5])
smaller = np.resize(arr, (3,))
print(smaller)

localhost:3000

3Best Practices

Follow these guidelines when working with np.resize():

1. Use np.reshape() (not np.resize()) whenever you want to preserve every original element exactly, just rearranged into a new shape

2. Reach for np.resize() specifically when you deliberately want repeated or truncated data to force-fill a target shape

3. Double-check the resulting array's contents after calling np.resize(), since silent repetition or truncation is easy to overlook

⚠️

Tip: Don't confuse np.resize() with reshape() — resize() can silently repeat or discard data to force-fit a new shape, which is rarely what you want if you actually just meant to reorganize an array's existing elements without changing them.

editor.html

import numpy as np

arr = np.array([1, 2, 3])
resized = np.resize(arr, (2, 4))
print(resized)

localhost:3000

Examples

Example 01Basic Usage

import numpy as np

arr = np.array([1, 2, 3])
resized = np.resize(arr, (2, 4))
print(resized)

Example 02Advanced Example

import numpy as np

arr = np.array([1, 2, 3, 4, 5])
smaller = np.resize(arr, (3,))
print(smaller)

Best Practices

Use np.reshape() (not np.resize()) whenever you want to preserve every original element exactly, just rearranged into a new shape
Reach for np.resize() specifically when you deliberately want repeated or truncated data to force-fill a target shape
Double-check the resulting array's contents after calling np.resize(), since silent repetition or truncation is easy to overlook

Interview Question

What's the fundamental difference between np.resize() and np.reshape(), beyond resize() just being 'more flexible'?▼

Hint: Think about what happens when the requested shape has a different total number of elements than the original array.

reshape() strictly preserves every original element, requiring the new shape's total element count to exactly match the original — it only ever reorganizes existing data. resize() instead actively changes the data itself to force-fit a target shape: if the target needs more elements, it repeats the original data from the start; if it needs fewer, it truncates. This makes resize() a genuinely different, data-altering operation, not simply a permissive version of reshape(), which is why using it by mistake when reshape() was intended can silently corrupt what should have been a pure rearrangement.

Exercises

MediumPractice using np.resize() in a real scenario.

View Solution

import numpy as np

arr = np.array([1, 2, 3])
resized = np.resize(arr, (2, 4))
print(resized)

Frequently Asked Questions

What's the fundamental difference between np.resize() and np.reshape(), beyond resize() just being 'more flexible'?▼

Related Functions

np-reshapenp-appendndarray-shape