Correlation with np.correlate() in NumPy

Introduction

We use correlation to find how two sets of numbers relate to each other. np.correlate() helps us measure this relationship by sliding one set over the other and checking how similar they are.

To compare two time series data to see if they move together.

To find repeating patterns in signals like sound or sensor data.

To check similarity between two sequences in data analysis.

To detect delays or shifts between two related datasets.

Syntax

NumPy

np.correlate(a, v, mode='valid')

a and v are the input arrays you want to compare.

mode controls the size of the output: 'valid' returns only points where arrays fully overlap, 'full' returns all overlaps, and 'same' returns output the same size as a.

Examples

Computes correlation where the second array fully fits inside the first.

NumPy

np.correlate([1, 2, 3], [0, 1, 0.5], mode='valid')

Computes correlation at all possible overlaps, including partial ones.

NumPy

np.correlate([1, 2, 3], [0, 1, 0.5], mode='full')

Returns correlation output with the same length as the first array.

NumPy

np.correlate([1, 2, 3], [0, 1, 0.5], mode='same')

Sample Program

This code compares two arrays using np.correlate() with different modes to show how the output changes.

NumPy

import numpy as np

a = np.array([1, 2, 3, 4])
v = np.array([0, 1, 0.5])

result_valid = np.correlate(a, v, mode='valid')
result_full = np.correlate(a, v, mode='full')
result_same = np.correlate(a, v, mode='same')

print('Valid mode:', result_valid)
print('Full mode:', result_full)
print('Same mode:', result_same)

Important Notes

np.correlate() is similar to convolution but without flipping the second array.

Correlation values can help find how much one sequence matches another at different shifts.

Use mode='full' to see all possible overlaps, which is useful for pattern detection.

Summary

Correlation measures similarity between two sequences.

np.correlate() slides one array over another and calculates overlap sums.

Different mode options control the size and detail of the output.