How to select columns from NumPy array

Numpy

In this post, we are going to understand how to select columns from NumPy array, N-Dimensional Numpy array contains rows and columns, We can filter data by selecting columns or rows. To select data by column from NumPy array slicing or Ellipsis or np. r_[] method is used. To run this program NumPy library must be installed on the local system

1. How to select columns from NumPy array using slicing

The syntax to select the column using slicing is given below, where the colon(:) represents that we are selecting all rows and passing column index or position to select the correspondence column.

Syntax

numpyarry[  : ,column]

Parameters

  • column:The column represent column index or position for whatever column we have to select.
  • Colon(:) :We are selecting all rows.

How does it work

  • We have created a NumPy array of size 12 and distributed it into 3 rows and 4 columns.
  • Selecting the 3rd column by using [:, 2]
  • Selecting the 2nd column by using [:, 1]
  • Selecting first column by using [:, 0]

Python program to select columns from NumPy array

import numpy as np

origanlArr = np.arange(12).reshape(3,4)

print('Original array:\n',origanlArr)

Third_Col = origanlArr[:, 2]
Sec_Col = origanlArr[:, 1]
Frist_Col = origanlArr[:, 0]


print('\nThird col:',Third_Col)
print('Second col:',Sec_Col)
print('First col:',Frist_Col )

Output

Original array:
 [[ 0  1  2  3]
 [ 4  5  6  7]
 [ 8  9 10 11]]

Third col: [ 2  6 10]
Second col: [1 5 9]
First col: [0 4 8]

2. How to select columns from NumPy array Using Ellipsis

The Ellipsis works similar to slicing instead of colon(:) we are using Ellipsis three-dot(. . .) to select all rows and column position or index to select column. Let us understand with the below example.

Syntax

 numparray[…,column]

Parameters

  • column: The column represents the column index or position for whatever column we have to select.
  • Ellipsis (…) : We are selecting all rows.

How does it works

  • We have created an numpy array of size 12 and distributed it into 3 rows and 4 columns.
  • Selecting the 3rd column by using […, 2]
  • Selecting the 2nd column by using […, 1]
  • Selecting first column by using […, 0]
import numpy as np

origanlArr = np.arange(12).reshape(3,4)

print('Original array:\n',origanlArr)

Third_Col = origanlArr[..., 2]
Sec_Col = origanlArr[..., 1]
Frist_Col = origanlArr[..., 0]


print('\nThird col:',Third_Col)
print('Second col:',Sec_Col)
print('First col:',Frist_Col )

Output

Original array:
 [[ 0  1  2  3]
 [ 4  5  6  7]
 [ 8  9 10 11]]

Third col: [ 2  6 10]
Second col: [1 5 9]
First col: [0 4 8]

3. Using np.r_ [] to select range of columns from NumPy array

We can select the range of columns by using the np.r_ () function Translates slice objects to concatenation along the first axis. If we use slicing notation[start:stop:step] it works like the np.arange() function.In this example, we are using the np. r_[] to select the range of columns from the NumPy array.

  • We have created an array of size 12 and distributed it 3 rows and 4 columns.
  • Finally using the np.r_[] function to select the range of columns from the NumPy array.

import numpy as np

origanlArr = np.arange(12).reshape(3,4)

print('Original array:\n',origanlArr)

print(origanlArr[:, np.r_[0:1, 3]])

Output

Original array:
 [[ 0  1  2  3]
 [ 4  5  6  7]
 [ 8  9 10 11]]

 Selcted columns :
 [[ 0  3]
 [ 4  7]
 [ 8 11]]

4. np.r_[] to Select a range of columns from NumPy array

In this example, we are selecting all rows and range of columns from a NumPy array by using the np. r_[] function.We have followed these step

  • We have created an array of size 12 and distributed it in 3 rows and 4 columns.
  • Finally using the np.r_[] function to select the range of columns from the NumPy array.
import numpy as np

origanlArr = np.arange(12).reshape(3,4)

print('Original array:\n',origanlArr)



print('\n Selcted columns :\n',origanlArr[:, np.r_[:1, 3,1:2]])

Output

Original array:
 [[ 0  1  2  3]
 [ 4  5  6  7]
 [ 8  9 10 11]]

 Selcted columns :
 [[ 0  3  1]
 [ 4  7  5]
 [ 8 11  9]]

Summary

In this post, we have learned how to select columns from NumPy array by using the different built-in methods with examples

Related Posts