Adding rows or columns to a NumPy array means appending new data along a specific axis. For example, if you have a 2D array like [[1, 2], [3, 4]] and you add a new row [5, 6], the array becomes [[1, 2], [3, 4], [5, 6]]. Similarly, adding a column [7, 8, 9] to a 3x2 array transforms it into a 3x3 array.

Let’s explore different ways to do this efficiently using NumPy.

Add columns in the Numpy array

We can add columns to a NumPy array using append(), concatenate(), insert(), column_stack() or hstack() with axis=1. Just make sure the new column has the same number of rows as the original array.

Using np.append()

np.append() adds values to a NumPy array along a specified axis or flattens if axis is not set.

import numpy as np
a = np.array([[1, 2, 3], [45, 4, 7], [9, 6, 10]])
b = np.array([[1], [2], [3]])

c = np.append(a, b, axis=1)
print(c)

[[ 1  2  3  1]
 [45  4  7  2]
 [ 9  6 10  3]]

Explanation:

• a is base array and b is array to append.
• np.append(a, b, axis=1) adds b as a new column to a, resulting in a 3x4 matrix
• whereas, axis=1 append along columns (i.e., horizontally).

Using np.concatenate()

np.concatenate() joins two or more arrays along a specified axis.

import numpy as np
a = np.array([[1, 2, 3], [45, 4, 7], [9, 6, 10]])
b = np.array([[1], [2], [3]])

c = np.concatenate([a, b], axis=1)
print (c)

[[ 1  2  3  1]
 [45  4  7  2]
 [ 9  6 10  3]]

Explanation:

• np.concatenate([a, b], axis=1) is joining list of two arrays [a, b].
• axis=1 join along columns (horizontally).

Using np.insert()

np.insert() inserts values into a NumPy array at specified positions along a given axis.

import numpy as np
a = np.array([[1, 2, 3], [45, 4, 7], [9, 6, 10]])
b = np.array([[1], [2], [3]])

c = np.insert(a, 3, b.flatten(), axis=1)
print(c)

[[ 1  2  3  1]
 [45  4  7  2]
 [ 9  6 10  3]]

Explanation:

• b.flatten() converts b to a 1D array [1, 2, 3].
• np.insert(a, 3, b.flatten(), axis=1) inserts b as the 4th column (index 3), one value per row.

Using np.hstack()   

np.hstack() horizontally stacks arrays (adds columns side by side).

import numpy as np
a = np.array([[1, 2, 3], [45, 4, 7], [9, 6, 10]])
b = np.array([1, 2, 3])

c = np.hstack((a, np.atleast_2d(b).T))
print (c)

[[ 1  2  3  1]
 [45  4  7  2]
 [ 9  6 10  3]]

Explanation:

• np.atleast_2d(b).T converts b to a 3×1 column.
• np.hstack() adds the column to the right of a.

Using np.column_stack()  

np.column_stack() stacks 1D or 2D arrays as columns to form a 2D array.

import numpy as np
a = np.array([[1, 2, 3], [45, 4, 7], [9, 6, 10]])
b = np.array([1, 2, 3])

c = np.column_stack((a, b))
print (c)

[[ 1  2  3  1]
 [45  4  7  2]
 [ 9  6 10  3]]

Explanation: np.column_stack((a, b)) stacks b as a new column to the right of a.

Add row in Numpy array

To add rows to a NumPy array, use functions like np.r_, np.insert(), np.vstack() or np.append() with axis=0.

Using np.r_ 

np.r_ is a shortcut for stacking arrays vertically (row-wise).

import numpy as np
a = np.array([[1, 2, 3], [45, 4, 7], [9, 6, 10]])
b = np.array([1, 2, 3])

c = np.r_[a,[b]]
print (c)

[[ 1  2  3]
 [45  4  7]
 [ 9  6 10]
 [ 1  2  3]]

Explanation: np.r_[a, [b]] appends b as a new row at the bottom of a.

Using np.insert()

np.insert() can add rows to a NumPy array by specifying the row index and using axis=0.

import numpy as np
a = np.array([[1, 2, 3], [45, 4, 7], [9, 6, 10]])
b = np.array([1, 2, 3])

n = a.shape[0]  
c = np.insert(a, n, b, axis=0)
print(c)

[[ 1  2  3]
 [45  4  7]
 [ 9  6 10]
 [ 1  2  3]]

Explanation:

• n = a.shape[0] gets row count (3).
• np.insert(a, n, b, axis=0) inserts b as a new row at the end of a.

Using np.vstack()  

np.vstack() adds arrays vertically, stacking them as new rows.

import numpy as np
a = np.array([[1, 2, 3], [45, 4, 7], [9, 6, 10]])
b = np.array([1, 2, 3])

c = np.vstack ((a, b))
print (c)

[[ 1  2  3]
 [45  4  7]
 [ 9  6 10]
 [ 1  2  3]]

Explanation: np.vstack((a, b)) vertically stacks b as a new row under a.

Using numpy.append()

You can add rows to an empty NumPy array using the numpy.append() function.

Example 1: Adding new rows to an empty 2-D array

import numpy as np   

a = np.empty((0,2), int)
print("Empty array:\n", a)

a = np.append(a, np.array([[10,20]]), axis=0)
a = np.append(a, np.array([[40,50]]), axis=0)

print("\nNew array:\n",a )

Empty array:
 []

New array:
 [[10 20]
 [40 50]]

Explanation:

• np.empty((0,2), int) creates an empty array with 0 rows and 2 columns.
• np.append(..., axis=0) appends new rows to a (row-wise), first adds [10, 20] then [40, 50].

Example 2: Adding new rows to an existing 2-D array.

import numpy as np   
a = np.array([[1, 2, 3], [4, 5, 6]])
print("Original array:\n", a)

b = np.array([[7, 8, 9], [10, 11, 12]])
c = np.append(a, b, axis=0)
print("\nNew array:\n", c)

Original array:
 [[1 2 3]
 [4 5 6]]

New array:
 [[ 1  2  3]
 [ 4  5  6]
 [ 7  8  9]
 [10 11 12]]

Explanation: a is a 2D array, and b has the same number of columns. Using np.append(axis=0), the new rows are added to the bottom of the original array.

Related Articles:

• np.append()
• np.concatenate
• np.insert()
• np.hstack() 
• np.column_stack() 
• np.vstack()