numpy.append() in Python

The function syntax is:

numpy.append(arr, values, axis=None)

• The arr can be an array-like object or a NumPy array. The values are appended to a copy of this array.
• The values are array-like objects and it’s appended to the end of the “arr” elements.
• The axis specifies the axis along which values are appended. If the axis is not provided, both the arrays are flattened.

Let’s look at some examples of NumPy append() function.

1. Flattening Two Arrays

import numpy as np
arr1 = np.array([[1, 2], [3, 4]])
arr2 = np.array([[10, 20], [30, 40]])
# no axis provided, array elements will be flattened
arr_flat = np.append(arr1, arr2)
print(arr_flat)  # [ 1  2  3  4 10 20 30 40]

2. Merging Along Axis

import numpy as np
arr_merged = np.append([[1, 2], [3, 4]], [[10, 20], [30, 40]], axis=0)
print(f'Merged 2x2 Arrays along Axis-0:\n{arr_merged}')
arr_merged = np.append([[1, 2], [3, 4]], [[10, 20], [30, 40]], axis=1)
print(f'Merged 2x2 Arrays along Axis-1:\n{arr_merged}')

Output:

Merged 2x2 Arrays along Axis-0:
[[ 1  2]
 [ 3  4]
 [10 20]
 [30 40]]
Merged 2x2 Arrays along Axis-1:
[[ 1  2 10 20]
 [ 3  4 30 40]]

• When the 2x2 arrays are merged along the x-axis, the output array size is 4x2.
• When the 2x2 arrays are merged along the y-axis, the output array size is 2x4.

3. Merging Arrays of different shapes

The append() function throws ValueError if both the arrays are of different shape, excluding the axis. Let’s understand this scenario with a simple example.

arr3 = np.append([[1, 2]], [[1, 2, 3], [1, 2, 3]])
print(arr3)
arr3 = np.append([[1, 2]], [[1, 2], [3, 4]], axis=0)
print(arr3)

• In the first example, the array elements are flattened. So even if they have completely different size - 1x2 and 2x3, the append() works fine.
• In the second example, the array shapes are 1x2 and 2x2. Since we are appending along the 0-axis, the 0-axis shape can be different. The other shapes should be the same, so this append() will also work fine.

Output:

[1 2 1 2 3 1 2 3]
[[1 2]
 [1 2]
 [3 4]]

Let’s look at another example where ValueError will be raised.

>>> import numpy as np
>>> 
>>> arr3 = np.append([[1, 2]], [[1, 2, 3]], axis=0)
Traceback (most recent call last):
  File "", line 1, in 
  File "/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/numpy/lib/function_base.py", line 4528, in append
    return concatenate((arr, values), axis=axis)
ValueError: all the input array dimensions except for the concatenation axis must match exactly
>>> 

The array shapes are 1x2 and 2x3. Since the axis-1 shapes are different, ValueError is raised. Reference: API Doc