JIYIK CN >

Python does not allow working with arrays directly. This is NumPywhere the library comes in, which makes it possible to handle and manipulate arrays in Python.

Arrays can be of any specified size and dimensions. Sometimes, there is a need to compensate for the dimensions of any particular array, and this is where padding comes in handy.

In simple terms, padding refers to adding meaningless values, usually zeros, to any row or column when talking about an array. It is widely used to compensate for missing large number of rows or columns in an array or matrix.

This tutorial demonstrates how to fill NumPyan array in Python. For example, we will fill the given NumPyarray with zeros in this tutorial.

Filling an array using NumPy.pad()functions in PythonNumPy

As the name suggests, the function is used to perform fill operations NumPy.pad()on an array.NumPy

NumPy.pad()The syntax of the function is as follows.

numpy.pad(array, pad_width, mode="constant", **kwargs)

NumPy.pad()All the parameters of the function have been defined below to make it easier for the reader to understand.

1. arrayThe - parameter specifies the array that must be filled.

2. pad_width− It specifies the number of values ​​that will be added to the edges of all axes. The tuple is used to specify the width of a multidimensional array.

3. mode- An optional parameter specifies the mode of the array.

4. **kwargs- It is possible to pass arguments of variable keyword length within a function. Mentioning is optional, you can read more about it online.

   The examples in this article do not use this parameter.

Here, we will take an example of multidimensional arrays, but the same can be done to handle one-dimensional arrays by slightly adjusting the code.

The following code uses NumPy.pad()the function to fill an array in Python NumPy.

import numpy as np

x = np.array([[1.0, 1.0, 1.0, 1.0], [1.0, 1.0, 1.0, 1.0], [1.0, 1.0, 1.0, 1.0]])
y = np.pad(x, [(0, 1), (0, 1)], mode="constant")
print(y)

The above code gives the following output.

[[1. 1. 1. 1. 0.]
 [1. 1. 1. 1. 0.]
 [1. 1. 1. 1. 0.]
 [0. 0. 0. 0. 0.]]

Here, we perform a padding operation on a multidimensional array to expand its dimensions to the size we specify.

Filling an array using shape()functions in PythonNumPy

This is an indirect method that achieves NumPy.pad()the same result as the function. shape()The function is NumPyanother function contained in the library that can be accessed after importing the library into your code.

shapeThe function is used to determine the dimensions of a given array or matrix.

This method is a bit contrived because it creates a new empty matrix with the desired dimensions based on the user's needs and then inserts the original matrix into the newly created empty matrix. It achieves the same result but takes a different indirect path in getting there.

In addition to shape()the function, this method also uses NumPy.zeros()the method to create an empty matrix.

For this approach, we need a reference matrix whose dimensions satisfy the user's needs for the target dimensions after padding, since we first reference the final dimensions required to create the empty matrix.

The following code uses shape()the function to fill an array in Python NumPy.

import numpy as np

x = np.array([[1.0, 1.0, 1.0, 1.0], [1.0, 1.0, 1.0, 1.0], [1.0, 1.0, 1.0, 1.0]])
y = np.array(
    [
        [1.0, 1.0, 1.0, 1.0, 1.0],
        [1.0, 1.0, 1.0, 1.0, 1.0],
        [1.0, 1.0, 1.0, 1.0, 1.0],
        [1.0, 1.0, 1.0, 1.0, 1.0],
    ]
)
z = np.zeros(np.shape(y))
z[: x.shape[0], : x.shape[1]] = x
print(z)

The above code gives the following output.

[[1. 1. 1. 1. 0.]
 [1. 1. 1. 1. 0.]
 [1. 1. 1. 1. 0.]
 [0. 0. 0. 0. 0.]]