In this article we will discuss different ways to delete elements from a Numpy Array by matching value or based on multiple conditions.

## Remove all occurrences of an element with given value from numpy array

Suppose we have a numpy array of numbers i.e.

# Create a numpy array from a list arr = np.array([4,5,6,7,8,9,10,11,4,5,6,33,6,7])

Now suppose we want to delete all occurrences of 6 from the above numpy array. Let’s see how to do that,

# Remove all occurrences of elements with value 6 from numpy array arr = arr[arr != 6] print('Modified Numpy Array by deleting all occurrences of 6') print(arr)

Output:

Modified Numpy Array by deleting all occurrences of 6 [ 4 5 7 8 9 10 11 4 5 33 7]

How does this worked ?

Basically arr != 6 returned a bool array of same size as arr with True at places where value is not 6 and False at other places i.e.

### Frequently Asked:

- Remove items in Array that are not in Another Array in Python
- Remove duplicates from NumPy Array in Python
- Remove All Occurrences of a Value from NumPy Array
- Delete elements from a Numpy Array by value or conditions in Python

[ True True False True True True True True True True False True False True]

Now if we pass this bool array to [] operator of numpy array arr, then it will select the elements from arr foe which bool array has **True** at corresponding index. Basically it returns the elements from arr which are not 6. Another point to be noted is that it returns a copy of existing array with elements with value 6. We can assign this new array back to arr to have the deletion effect of all occurrences of 6 from the numpy array.

## Delete elements in Numpy Array based on multiple conditions

Suppose we have a numpy array of numbers i.e.

# Create a numpy array from a list arr = np.array([4, 5, 6, 7, 8, 9, 10, 11, 4, 5, 6, 33, 6, 7])

Now we want to delete all occurrences of elements below 6 & greater than 10 i.e. keep elements between range 6 to 10 only. Let’s see how to do that,

# Remove all occurrences of elements below 6 & greater than 10 i.e. keep elements between range 6 to 10 only arr = arr[ (arr >= 6) & (arr <= 10) ] print('Modified Numpy Array by deleting all occurrences of elements not in range 6 to 10 : ') print(arr)

Output:

Modified Numpy Array by deleting all occurrences of elements not in range 6 to 10 : [ 6 7 8 9 10 6 6 7]

We basically created a bool array using multiple conditions on numpy array and then passed that bool array to [] operator of numpy array to select the elements only which satisfies the given conditions. So, it returned a copy of numpy array by selecting values below 6 & greater than 10 only and we assigned this new array back to arr to have the deletion effect.

## Delete elements by value or condition using np.argwhere() & np.delete()

Suppose we have a numpy array of numbers i.e.

arr = np.array([4, 5, 6, 7, 8, 9, 10, 11, 4, 5, 6, 33, 6, 7])

Now let’s delete all occurrences of 6 from the above numpy array using np.argwhere() & np.delete() i.e.

# Single line solution to delete all occurrences of element with value 6 arr = np.delete(arr, np.argwhere(arr == 6)) print('Modified Numpy Array :') print(arr)

Output:

Modified Numpy Array : [ 4 5 7 8 9 10 11 4 5 33 7]

**How did that worked ?**

boolArr = (arr == 6)

**arr == 6** Returned a Numpy array of bool type with **True** at places where arr has 6 and False at other places. Size of this bool array will be equal to size of arr. Therefore contents of **boolArr** are,

[False False True False False False False False False False True False True False]

Now pass this bool array to **np.argwhere()** which accepts a bool array and return the index positions where bool array has True value i.e.

indexArr = np.argwhere(arr == 6)

Contents of **indexArr** are,

[[ 2] [10] [12]]

These are index positions from array arr where element value is 6. Now pass this index positions to np.delete() to delete elements from arra at given index positions i.e.

# Delete elements at given index position i.e. elements with value 6 arr = np.delete(arr, indexArr) print('Modified Numpy Array :') print(arr)

Output:

Modified Numpy Array : [ 4 5 7 8 9 10 11 4 5 33 7]

It deleted all occurrences of element with value 6.

## Delete elements by multiple conditions using np.argwhere() & np.delete()

Contents of original Numpy array **arr** is,

[4, 5, 6, 7, 8, 9, 10, 11, 4, 5, 6, 33, 6, 7]

Let’s delete all occurrences of elements between 6 to 10 in a single line i.e.

# Single line solution to delete all occurrences of element between 6 to 10 arr = np.delete(arr, np.argwhere( (arr >= 6) & (arr <= 10) )) print('Modified Numpy Array :') print(arr)

Output:

Modified Numpy Array : [ 4 5 11 4 5 33]

**Complete example is as follows:**

import numpy as np def main(): # Create a numpy array from a list arr = np.array([4,5,6,7,8,9,10,11,4,5,6,33,6,7]) print('Original Array : ', arr) print('*** Delete all occurrences of an element in Numpy Array ***') print(arr != 6) # Remove all occurrences of elements with value 6 from numpy array arr = arr[arr != 6] print('Modified Numpy Array by deleting all occurrences of 6') print(arr) print('*** Delete elements in Numpy Array based on multiple conditions ***') # Create a numpy array from a list arr = np.array([4, 5, 6, 7, 8, 9, 10, 11, 4, 5, 6, 33, 6, 7]) print('Original Array : ', arr) # Remove all occurrences of elements below 6 & greater than 10 i.e. keep elements between range 6 to 10 only arr = arr[ (arr >= 6) & (arr <= 10) ] print('Modified Numpy Array by deleting all occurrences of elements not in range 6 to 10 : ') print(arr) print('*** Delete elements by value using np.argwhere() & np.delete() ***') arr = np.array([4, 5, 6, 7, 8, 9, 10, 11, 4, 5, 6, 33, 6, 7]) print('Original Array : ') print(arr) boolArr = (arr == 6) print('Bool Array with True for elements with value 6 : ') print(boolArr) indexArr = np.argwhere(boolArr) print('Index positions from array arr where element value is 6 :') print(indexArr) # Delete elements at given index position i.e. elements with value 6 arr = np.delete(arr, indexArr) print('Modified Numpy Array :') print(arr) arr = np.array([4, 5, 6, 7, 8, 9, 10, 11, 4, 5, 6, 33, 6, 7]) # Single line solution to delete all occurrences of element with value 6 arr = np.delete(arr, np.argwhere(arr == 6)) print('Modified Numpy Array :') print(arr) arr = np.array([4, 5, 6, 7, 8, 9, 10, 11, 4, 5, 6, 33, 6, 7]) # Single line solution to delete all occurrences of element between 6 to 10 arr = np.delete(arr, np.argwhere( (arr >= 6) & (arr <= 10) )) print('Modified Numpy Array :') print(arr) if __name__ == '__main__': main()

**Output:**

Original Array : [ 4 5 6 7 8 9 10 11 4 5 6 33 6 7] *** Delete all occurrences of an element in Numpy Array *** [ True True False True True True True True True True False True False True] Modified Numpy Array by deleting all occurrences of 6 [ 4 5 7 8 9 10 11 4 5 33 7] *** Delete elements in Numpy Array based on multiple conditions *** Original Array : [ 4 5 6 7 8 9 10 11 4 5 6 33 6 7] Modified Numpy Array by deleting all occurrences of elements not in range 6 to 10 : [ 6 7 8 9 10 6 6 7] *** Delete elements by value using np.argwhere() & np.delete() *** Original Array : [ 4 5 6 7 8 9 10 11 4 5 6 33 6 7] Bool Array with True for elements with value 6 : [False False True False False False False False False False True False True False] Index positions from array arr where element value is 6 : [[ 2] [10] [12]] Modified Numpy Array : [ 4 5 7 8 9 10 11 4 5 33 7] Modified Numpy Array : [ 4 5 7 8 9 10 11 4 5 33 7] Modified Numpy Array : [ 4 5 11 4 5 33]