ML 4 - BACKPROPAGATION ALGORITHM

4. BUILD AN ARTIFICIAL NEURAL NETWORK BY IMPLEMENTING THE BACKPROPAGATION ALGORITHM AND TEST THE SAME USING APPROPRIATE DATASETS.

 SOLUTION 

NO DATASET

lab4.py

import numpy as np
X = np.array(([2, 9], [1, 5], [3, 6]), dtype=float)
y = np.array(([92], [86], [89]), dtype=float)
X = X/np.amax(X, axis=0)
y = y/100

class Neural_Network(object):
    def __init__(self):
        self.inputSize = 2
        self.outputSize = 1
        self.hiddenSize = 3
        self.W1 = np.random.randn(self.inputSize, self.hiddenSize)
        self.W2 = np.random.randn(self.hiddenSize, self.outputSize)

    def forward(self, X):
        self.z = np.dot(X, self.W1)
        self.z2 = self.sigmoid(self.z)
        self.z3 = np.dot(self.z2, self.W2)
        o = self.sigmoid(self.z3) 
        return o 

    def sigmoid(self, s):
        return 1/(1+np.exp(-s))

    def sigmoidPrime(self, s):
        return s * (1 - s)
    
    def backward(self, X, y, o):
        self.o_error = y - o
        self.o_delta = self.o_error*self.sigmoidPrime(o)
        self.z2_error = self.o_delta.dot(self.W2.T)
        self.z2_delta = self.z2_error*self.sigmoidPrime(self.z2)
        self.W1 += X.T.dot(self.z2_delta)
        self.W2 += self.z2.T.dot(self.o_delta)

    def train (self, X, y):
        o = self.forward(X)
        self.backward(X, y, o)

NN = Neural_Network()
for i in range(1000):
       print ("\nInput: \n" + str(X))
       print ("\nActual Output: \n" + str(y))
       print ("\nPredicted Output: \n" + str(NN.forward(X)))
       print ("\nLoss: \n" + str(np.mean(np.square(y - NN.forward(X)))))
       NN.train(X, y)

STEPS & OUTPUT:

to view steps & output click HERE