Eddie-Carrizales/Neural-Network-Activation-Functions

This project consists of implementing combinations of hyperparameters on a Neural Network to determine what activation function (Logistic, ReLu, or Tanh) and what hyperparameters with each of those activation functions performs the best.

Neural Network Hyperparameter Implementation and Activation Function Analysis

This project consists of implementing combinations of hyperparameters on a Neural Network to determine what activation function (Logistic, ReLu, or Tanh) and what hyperparameters with each of those activation functions performs the best.

Below are three Loss/Error/Cost curve graphs for each of the activation functions.

Note: There are more graphs shown in the code file.

Logistic (Sigmoid) Loss Curves

ReLu Loss Curves

Tanh Loss Curves