Source code for neural_network.functions.sigmoid

import math
from typing import List

from .abstract_function import AbstractFunction




class Sigmoid(AbstractFunction):
    """Class to represent the Sigmoid function
    """



    def __init__(self):
        """Constructor method
        """
        super().__init__()




    def __call__(self, x: float, w: List[float] = None) -> float:
        """Implementation of Sigmoid

        Parameters
        ----------
        x : float
            Input to function
        w : List[float]
            Weights (not used here)

        Returns
        -------
        float
            Output to function
        """
        # We do the below to avoid overflow errors
        if x < 0:
            return math.exp(x) / (1 + math.exp(x))
        else:
            return 1 / (1 + math.exp(-x))




    def gradient(self, x: float, w: List[float] = None) -> float:
        """Gradient of Sigmoid

        Parameters
        ----------
        x : float
            Input to function
        w : List[float]
            Not used for this class

        Returns
        -------
        float
            Gradient of Sigmoid
        """
        return self(x) * (1 - self(x))