How to Calculate a Running Mean (Moving Average) in Python Using NumPy?

Running Mean in Python with NumPy

Calculating the running mean, also known as the moving average, of a 1D array is a common task in data analysis. NumPy provides a powerful tool called np.convolve for performing convolution operations, including the running mean.

Definition and Implementation:

The running mean involves sliding a window along the input array and computing the mean of the values within the window at each step. In NumPy, this is achieved as follows:

import numpy as np

array = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
window_size = 3

result = np.convolve(array, np.ones(window_size) / window_size, mode='valid')

Explanation:

• np.ones(window_size) creates an array of ones with size equal to the window size.
• np.ones(window_size) / window_size normalizes the array by dividing each element by the window size, resulting in a kernel for computing the arithmetic mean.
• np.convolve takes the kernel and convolves it with the input array, performing a sliding mean calculation.
• mode='valid' specifies that only the portion of the array that can be completely covered by the window should be returned, resulting in a result of size len(array) - window_size 1.

Edge Handling:

The mode argument in np.convolve controls how the edges of the array are handled during convolution. The available modes are 'full', 'same', and 'valid':

• 'full' includes both the original length and appended zeros.
• 'same' appends zeros until the output shape is the same as the input shape.
• 'valid' only includes the portion of the array that can be completely covered by the window.

The 'valid' mode is typically used for the running mean, as it provides a result that does not include the windowed sections of the beginning and end of the array.

Example:

In the example above, the result will be:

[4.  5.  6.  7.  8.  9.]

This represents the running mean of the input array with a window size of 3.

The above is the detailed content of How to Calculate a Running Mean (Moving Average) in Python Using NumPy?. For more information, please follow other related articles on the PHP Chinese website!