Table of Contents
In NumPy, you can use the numpy.where() function to get the indices of elements in an array where the value is True. This function takes in a boolean array and returns a tuple of arrays, one for each dimension of the input array, containing the indices of the elements that are True. This is a useful way to quickly find the locations of certain elements in an array without having to manually iterate over it.
You can use the following methods to get the indices where some condition is true in NumPy:
Method 1: Get Indices Where Condition is True in NumPy Array
#get indices of values greater than 10
np.asarray(my_array>10).nonzero()
Method 2: Get Indices Where Condition is True in NumPy Matrix
#get indices of values greater than 10
np.transpose((my_matrix>10).nonzero())
Method 3: Get Indices Where Condition is True in Any Row of NumPy Matrix
#get indices of rows where any value is greater than 10 np.asarray(np.any(my_matrix>10, axis=1)).nonzero()
The following examples show how to use each method in practice.
Example 1: Get Indices Where Condition is True in NumPy Array
The following code shows how to get all indices in a NumPy array where the value is greater than 10:
import numpy as np
#create NumPy array
my_array = np.array([2, 2, 4, 5, 7, 9, 11, 12, 3, 19])
#get index of values greater than 10
np.asarray(my_array>10).nonzero()
(array([6, 7, 9], dtype=int32),)
From the output we can see that the values in index positions 6, 7, and 9 of the original NumPy array have values greater than 10.
Example 2: Get Indices Where Condition is True in NumPy Matrix
The following code shows how to get all indices in a NumPy matrix where the value is greater than 10:
import numpy as np
#create NumPy matrix
my_matrix = np.array([[2, 5, 9, 12],
[6, 7, 8, 8],
[2, 5, 7, 8],
[4, 1, 15, 11]])
#get index of values greater than 10
np.transpose((my_matrix>10).nonzero())
array([[0, 3],
[3, 2],
[3, 3]], dtype=int32) From the output we can see that the values in the following index positions of the matrix have values greater than 10:
- [0, 3]
- [3, 2]
- [3, 3]
Example 3: Get Indices Where Condition is True in Any Row of NumPy Matrix
The following code shows how to get all row indices in a NumPy matrix where the value is greater than 10 in any element of the row:
import numpy as np
#create NumPy matrix
my_matrix = np.array([[2, 5, 9, 12],
[6, 7, 8, 8],
[2, 5, 7, 8],
[4, 1, 15, 11]])
#get index of rows where any value is greater than 10
np.asarray(np.any(my_matrix>10, axis=1)).nonzero()
(array([0, 3], dtype=int32),)
From the output we can see that rows 0 and 3 have at least one value greater than 10.
Note: To get indices where a condition is true in a column, use axis=0 instead.
The following tutorials explain how to perform other common tasks in NumPy: