numpy.mean() in Python

numpy.mean() is a NumPy function used to calculate the average (arithmetic mean) of numeric values. It can compute the mean of a 1D list/array or compute mean row-wise and column-wise for multi-dimensional arrays.

Example:

Input: [1, 2, 3]
Output: 2.0

Syntax

We use the following syntax to calculate the mean in NumPy:

numpy.mean(arr, axis=None, dtype=None, out=None)

Parameters:

• arr: Input array of numbers
• axis: None - mean of all elements, 0 - column-wise mean and 1 - row-wise mean
• dtype(Optional): type used while computing mean
• out(Optional): array to store the result

Examples

Example 1: This example finds the average value of a 1D list using np.mean().

import numpy as np
arr = [20, 2, 7, 1, 34]
res = np.mean(arr)
print(res)

12.8

Explanation: (20 + 2 + 7 + 1 + 34)/5 = 12.8

Example 2: This example shows how to compute the mean of all elements, each column, and each row using axis.

import numpy as np

arr = [[14, 17, 12],
       [15,  6, 27],
       [23,  2, 54]]

print(np.mean(arr))           # entire array
print(np.mean(arr, axis=0))   # column-wise mean
print(np.mean(arr, axis=1))   # row-wise mean

18.88888888888889
[17.33333333  8.33333333 31.        ]
[14.33333333 16.         26.33333333]

Example 3: This example stores the result of row-wise mean into another array using out.

import numpy as np

arr = [[5, 10, 15],
       [3,  6,  9],
       [8, 16, 24]]

res = np.zeros(3)
np.mean(arr, axis=1, out=res)
print(res)

[10.  6. 16.]

Explanation: out=res stores the row-wise mean values into res.