np.cos() behaves exactly like np.sin() in terms of expecting radians and being bounded between -1 and 1, but computes the cosine instead of the sine — the two are phase-shifted versions of the same underlying wave, with cos(x) equal to sin(x + pi/2). Together, sin and cos are the building blocks for representing rotations, oscillations, and periodic signals, and they satisfy the identity sin(x) squared plus cos(x) squared equals 1 for any real x.

1Understanding np.cos()

💡

Just like np.sin(), np.cos() expects radians, not degrees — use np.radians() to convert degree values first if that's the form your input data is in.

editor.html

import numpy as np

angles = np.array([0, np.pi / 2, np.pi])
print(np.cos(angles))

localhost:3000

2Practical Example

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

editor.html

import numpy as np

theta = np.linspace(0, 2 * np.pi, 4, endpoint=False)
x = np.cos(theta)
y = np.sin(theta)
print(np.round(x, 2))
print(np.round(y, 2))

localhost:3000

3Best Practices

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

1. Convert degree values to radians with np.radians() before calling np.cos(), the same as for np.sin()

2. Use np.cos() and np.sin() together for rotation calculations or generating circular/periodic coordinate data

3. Use np.isclose() rather than == when validating trigonometric identities or expected results numerically, due to floating-point precision

⚠️

Tip: Just like np.sin(), np.cos() expects radians, not degrees — use np.radians() to convert degree values first if that's the form your input data is in.

editor.html

import numpy as np

angles = np.array([0, np.pi / 2, np.pi])
print(np.cos(angles))

localhost:3000

Examples

Example 01Basic Usage

import numpy as np

angles = np.array([0, np.pi / 2, np.pi])
print(np.cos(angles))

Example 02Advanced Example

import numpy as np

theta = np.linspace(0, 2 * np.pi, 4, endpoint=False)
x = np.cos(theta)
y = np.sin(theta)
print(np.round(x, 2))
print(np.round(y, 2))

Best Practices

Convert degree values to radians with np.radians() before calling np.cos(), the same as for np.sin()
Use np.cos() and np.sin() together for rotation calculations or generating circular/periodic coordinate data
Use np.isclose() rather than == when validating trigonometric identities or expected results numerically, due to floating-point precision

Interview Question

How are np.sin() and np.cos() related mathematically, and how does that show up in the code2 example generating points around a circle?▼

Hint: Think about the standard parametric equations for a circle.

sin and cos are the same wave shifted by a quarter cycle, or pi/2 radians, relative to each other, and together they satisfy the identity that sin squared plus cos squared always equals 1 for any input. This identity is exactly what makes the pair (cos(theta), sin(theta)) trace out points on a unit circle as theta varies — which is why generating x-coordinates from cos() and y-coordinates from sin() over an evenly spaced set of angles, as in the example, produces points evenly distributed around a circle.

Exercises

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

View Solution

import numpy as np

angles = np.array([0, np.pi / 2, np.pi])
print(np.cos(angles))

Frequently Asked Questions

How are np.sin() and np.cos() related mathematically, and how does that show up in the code2 example generating points around a circle?▼

Related Functions

np-sinnp-tannp-linspace