﻿ NumPy - Array with random values and normalize it column-wise

Advanced NumPy Exercises - Normalize a 5x5 array column-wise with random values

Write a NumPy program to create a 5x5 array with random values and normalize it column-wise.

To normalize a 5x5 array column-wise using NumPy, you can generate the array with numpy.random.rand. For column-wise normalization, compute the mean and standard deviation of each column using numpy.mean and numpy.std with the axis parameter set to 0. Subtract the mean and divide by the standard deviation for each element in the column, resulting in normalized columns with a mean of 0 and a standard deviation of 1.

Sample Solution:

Python Code:

``````import numpy as np
# create a 5x5 array with random values
nums = np.random.rand(5, 5)
print("Original array elements:")
print(nums)
# compute the mean of each column
col_means = np.mean(nums, axis=0)

# normalize each column by subtracting its mean and dividing by its standard deviation
arr_normalized = (nums - col_means) / np.std(nums, axis=0)

print("Original Array:")
print(nums)
print("\nNormalized Array (column-wise):")
print(arr_normalized)
``````

Output:

```Original array elements:
[[0.83858454 0.53779431 0.04162028 0.26871443 0.1940342 ]
[0.9761744  0.71604666 0.27396056 0.31518587 0.54501194]
[0.54896879 0.15611648 0.11122861 0.05373429 0.73362389]
[0.09963178 0.24274453 0.85315162 0.72934953 0.0869224 ]
[0.81961978 0.16084128 0.18108736 0.87256786 0.5897874 ]]
Original Array:
[[0.83858454 0.53779431 0.04162028 0.26871443 0.1940342 ]
[0.9761744  0.71604666 0.27396056 0.31518587 0.54501194]
[0.54896879 0.15611648 0.11122861 0.05373429 0.73362389]
[0.09963178 0.24274453 0.85315162 0.72934953 0.0869224 ]
[0.81961978 0.16084128 0.18108736 0.87256786 0.5897874 ]]

Normalized Array (column-wise):
[[ 0.58516372  0.77785141 -0.86165997 -0.58783195 -0.95594713]
[ 1.02756813  1.56977146 -0.06275022 -0.43538767  0.46668539]
[-0.34606249 -0.91782512 -0.62230944 -1.2930498   1.23119395]
[-1.79085407 -0.53296347  1.92881744  0.92322901 -1.3901078 ]
[ 0.52418471 -0.89683428 -0.38209781  1.39304041  0.64817558]]
```

Explanation:

In the above exercise -

nums = np.random.rand(5, 5): Create a 5x5 array with random values between 0 and 1.

col_means = np.mean(nums, axis=0): Calculate the mean of each column in nums using np.mean and specifying axis=0. This gives a 1D array with length 5 containing the column means.

arr_normalized = (nums - col_means) / np.std(nums, axis=0): Subtract the column means from nums to center the data around zero. Then, divide each element by the standard deviation of its column to normalize the data. This is known as standardization or z-score normalization. The np.std function is used to calculate the standard deviation along the columns (axis=0) and the resulting array is broadcasted to the same shape as nums so that each element can be divided by the standard deviation of its column. The normalized array is stored in arr_normalized.

Python-Numpy Code Editor:

Have another way to solve this solution? Contribute your code (and comments) through Disqus.

What is the difficulty level of this exercise?

Test your Programming skills with w3resource's quiz.

﻿