NumPy: Replace all the nan of a given array with the mean of another array

NumPy: Array Object Exercise-178 with Solution

Write a NumPy program to replace all the nan (missing values) of a given array with the mean of another array.

Sample Solution:

Python Code:

# Importing the NumPy library
import numpy as np

# Creating NumPy arrays: array_nums1 from 0 to 19 reshaped into a 4x5 array and array_nums2 with NaN values
array_nums1 = np.arange(20).reshape(4, 5)
array_nums2 = np.array([[1, 2, np.nan], [4, 5, 6], [np.nan, 7, np.nan]])

# Printing the original arrays
print("Original arrays:")
print(array_nums1)
print(array_nums2)

# Replacing all the NaN values in array_nums2 with the mean of non-NaN values in array_nums1
array_nums2[np.isnan(array_nums2)] = np.nanmean(array_nums1)
print("\nAll the NaN of array_nums2 replaced by the mean of array_nums1:")
print(array_nums2)

Sample Output:

Original arrays:
[[ 0  1  2  3  4]
 [ 5  6  7  8  9]
 [10 11 12 13 14]
 [15 16 17 18 19]]
[[ 1.  2. nan]
 [ 4.  5.  6.]
 [nan  7. nan]]

All the nan of array_nums2 replaced by the mean of array_nums1:
[[1.  2.  9.5]
 [4.  5.  6. ]
 [9.5 7.  9.5]]

Explanation:

In the above example -

array_nums1 = np.arange(20).reshape(4,5) creates a 1-dimensional NumPy array containing numbers from 0 to 19 and then reshapes it into a 2-dimensional array with 4 rows and 5 columns.

array_nums2 = np.array([[1,2,np.nan],[4,5,6],[np.nan, 7, np.nan]]) creates a 2-dimensional NumPy array with NaN values.

array_nums2[np.isnan(array_nums2)]= np.nanmean(array_nums1)

In the above code -

• np.nanmean(array_nums1): This part computes the mean of the ‘array_nums1’ while ignoring any NaN values that might be present. In this case, since there are no NaN values in array_nums1, it is equivalent to computing the mean of all elements in array_nums1.
• array_nums2[np.isnan(array_nums2)]: This part selects all NaN values in array_nums2.
• array_nums2[np.isnan(array_nums2)] = np.nanmean(array_nums1) replaces the selected NaN values in array_nums2 with the computed mean from array_nums1.

Python-Numpy Code Editor:

Previous: Write a NumPy program to create an array of 4,5 shape and to reverse the rows of the said array. After reversing 1st row will be 4th and 4th will be 1st, 2nd row will be 3rd row and 3rd row will be 2nd row.
Next: Write a NumPy program to fetch all items from a given array of 4,5 shape which are either greater than 6 and a multiple of 3.