2023-04-25 16:06:14 +00:00
|
|
|
"""
|
|
|
|
This script demonstrates the implementation of the tangent hyperbolic
|
|
|
|
or tanh function.
|
|
|
|
|
|
|
|
The function takes a vector of K real numbers as input and
|
|
|
|
then (e^x - e^(-x))/(e^x + e^(-x)). After through tanh, the
|
|
|
|
element of the vector mostly -1 between 1.
|
|
|
|
|
|
|
|
Script inspired from its corresponding Wikipedia article
|
|
|
|
https://en.wikipedia.org/wiki/Activation_function
|
|
|
|
"""
|
|
|
|
import numpy as np
|
|
|
|
|
|
|
|
|
2023-08-15 21:27:41 +00:00
|
|
|
def tangent_hyperbolic(vector: np.ndarray) -> np.ndarray:
|
2023-04-25 16:06:14 +00:00
|
|
|
"""
|
|
|
|
Implements the tanh function
|
|
|
|
|
|
|
|
Parameters:
|
2023-08-15 21:27:41 +00:00
|
|
|
vector: np.ndarray
|
2023-04-25 16:06:14 +00:00
|
|
|
|
|
|
|
Returns:
|
|
|
|
tanh (np.array): The input numpy array after applying tanh.
|
|
|
|
|
|
|
|
mathematically (e^x - e^(-x))/(e^x + e^(-x)) can be written as (2/(1+e^(-2x))-1
|
|
|
|
|
|
|
|
Examples:
|
|
|
|
>>> tangent_hyperbolic(np.array([1,5,6,-0.67]))
|
|
|
|
array([ 0.76159416, 0.9999092 , 0.99998771, -0.58497988])
|
|
|
|
|
|
|
|
>>> tangent_hyperbolic(np.array([8,10,2,-0.98,13]))
|
|
|
|
array([ 0.99999977, 1. , 0.96402758, -0.7530659 , 1. ])
|
|
|
|
|
|
|
|
"""
|
|
|
|
|
|
|
|
return (2 / (1 + np.exp(-2 * vector))) - 1
|
|
|
|
|
|
|
|
|
|
|
|
if __name__ == "__main__":
|
|
|
|
import doctest
|
|
|
|
|
|
|
|
doctest.testmod()
|