This Model is Fraud Email and Email and Text Classifier using Artificial Neural Networks. Whole Model is Built from Scratch in such a way that it can be used for any number of Hidden Layers and any number of Hidden units and Output Classes Without Changing any of the code.
Main Project File is in ProjectMain Directory.
ProjectMain Directory Contains Three Files.
0.Main File - This file Contains Main code which read data and initialize the algorithm.
1.EmailProcessing.py - This File Contains code for Processing the Text or Email and return a Numeric Vector which can be used for the calculation.
2.VocabArray.py - This File Contains Code for reading and returning Vocab list also known as bag of words.
DeepNetsOptimization Contains Files for Neural Networks as follows.
-
ActivationDerivative.py Ths file Calculate the derivatives of Activation Function like Sigmoid and ReLU.
-
HypothesisFunc.py This file Contains Code for Calculating Hypothesis or Output of the Neurons or Units in the Layers.
-
LossFunc.py This file contains code for calculating Loss or cost which we calculate after Output layer.