Working with Numpy Arrays in Python

Numpy arrays are commonly encountered in Python, particularly in AI and other scientific work. This guide shows a number of examples of how numpy arrays can be manipulated.

Importing Numpy

import numpy as np

Convert a Regular Python List to a Numpy Array

>>> x = (0,1,2,3,4,5,6,7,8,9,10)
>>> a_numpy_array = np.array(x)

And convert it back again to a regular Python list:

regular_list = a_numpy_array.tolist()

Fill a Numpy Array with Sequential Digits

np.arange(12)

Result:

array([ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11])

Now create an array between 100 and 120, counting in steps of 2:

>>> np.arange(100,120,2)
array([100, 102, 104, 106, 108, 110, 112, 114, 116, 118])

Rearrange a Numpy Array from 1D to 2D and Back Again

Generate a 1D array of 12 integers and then re-arrange the 1D array into a 2D array with 4 rows and 3 columns

>>> a = np.arange(12)
>>> b = a.reshape(4,3)
>>> b
array([[ 0,  1,  2],
       [ 3,  4,  5],
       [ 6,  7,  8],
       [ 9, 10, 11]])

Now re-arrange the 4x3 2D array back into the original 1D array of 12 integers:

>>> c = b.reshape(1,12)
>>> c
array([[ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11]])

The 4 rows x 3 columns 2D array can be re-arranged into 3 rows x 4 columns:

>> d = b.reshape(3,4)
>>> d
array([[ 0,  1,  2,  3],
       [ 4,  5,  6,  7],
       [ 8,  9, 10, 11]])

Add an Extra Column to a 2D Numpy Array

>>> my_numpy_array
array([[ 0,  1,  2,  3],
       [ 4,  5,  6,  7],
       [ 8,  9, 10, 11]])

>>> extra_column = np.array( [100,200,300] ).reshape(3,1)
>>> extra_column
array([[100],
       [200],
       [300]])

>>> np.column_stack( ( my_numpy_array, extra_column ) )
array([[  0,   1,   2,   3, 100],
       [  4,   5,   6,   7, 200],
       [  8,   9,  10,  11, 300]])

Extract a Column from a 2D Numpy Array

Extract the 2nd column from the following two dimensional numpy array:

>> my_numpy_array,
(array([[ 0,  1,  2,  3],
       [ 4,  5,  6,  7],
       [ 8,  9, 10, 11]]),)
 
>>> my_numpy_array[:,1]
array([1, 5, 9])