JIYIK CN >

This tutorial will discuss the Relu function and how to implement it in Python.

ReLUfunction

ReluFunctions are the foundation of machine learning and are essential when using deep learning.

ReLUThe term is Rectified Linear Unitan acronym for and is used as an activation function in the case of artificial neural networks, which are the basis of deep learning. Python, as one of the programming languages ​​suitable for implementing machine learning and deep learning algorithms, has ReLUa scope for the use of function.

In simple mathematical terms, ReLUthe function can be defined as,

f(x) = max(0,x)

This function is linear in x and sets all negative values ​​to zero.

ReLUImplementing the function in Python

To implement the function in Python ReLU, we can define a new function and use the NumPy library.

The NumPy library makes it possible to work with matrices and arrays in Python, as they cannot be directly implemented in this programming language. maximum()The NumPy function from the NumPy library can be used in our newly created function definition to create a NumPy ReLUfunction.

The following code implements the function in Python ReLU.

import numpy as np


def relu1(a):
    return np.maximum(0, a)


print(relu1(-3))

The above code gives the following output:

0

In the code above, we deal with a single integer. However, ReLUthe functions we create can easily handle anything from single integers to NumPy arrays and similar objects.

When this function gets a number as input, the output will always be a number. The same rule is followed when we pass an array to this function.

The object type of the input is always the object type returned as output.

Interestingly, when working with arrays, we can use plotlythe library and even create graphs that describe ReLUwhat the functions do on arrays and similar objects.

To better explain ReLUhow the function works, we will now take an example of a simple array and complete the task at hand. In addition, we will also draw a graph and see ReLUthe actual action of the function on this array.

The following code uses the function on an array in Python ReLU.

import numpy as np
import plotly.express as px


def relu1(a):
    return np.maximum(0, a)


x1 = np.linspace(start=-5, stop=5, num=26)
print(x1)
x2 = relu1(x1)
print(x2)
px.line(x=x1, y=x2)

The above code gives the following output: