Simple Neural Network For XOR Gate

Neural network of one hidden layer with five neurons import numpy as np print("\n--------------------Simple neural network for XOR gat...

Neural network of one hidden layer with five neurons

import numpy as np

print("\n--------------------Simple neural network for XOR gate--------------------\n\n")

learning_rate = float(input("Enter leraning rate for model. Example 0.01\n"))

epochs = int(input("\nEnter no of epoches. Example 5000\n"))

x=np.array([[0,0,1,1],[0,1,0,1]])

y=np.array([[0,1,1,0]])

m = x.shape[1]

no_of_input = 2

no_of_output = 1

no_of_neurons = 5

np.random.seed(2)

w1 = np.random.rand(no_of_neurons,no_of_input)   # Weight matrix for hidden layer 

w2 = np.random.rand(no_of_output,no_of_neurons)   # Weight matrix for output layer

b1 = np.zeros((no_of_neurons, 1))

b2 = np.zeros((no_of_output, 1))

# sigmoid activation function 

def sigmoid(z):

    z= 1/(1+np.exp(-z))

    return z

# Forward propagation

def forward_propagation(w1, w2, x, b1, b2):

    z1 = np.dot(w1,x) + b1

    a1 = sigmoid(z1)    

    z2 = np.dot(w2,a1) + b2

    a2 = sigmoid(z2)

    return z1,a1,z2,a2

# Backward propagation

def backward_propagation(m, w1, w2, z1, a1, z2, a2, y):

    dz2 = a2-y

    dw2 = np.dot(dz2,a1.T)/m

    db2 = np.sum(dz2, axis = 1, keepdims = True)

    dz1 = np.dot(w2.T,dz2) * a1*(1-a1)

    dw1 = np.dot(dz1,x.T)/m

    dw1 = np.reshape(dw1,w1.shape)

    db1 = np.sum(dz1, axis = 1, keepdims = True) / m

    dw2 = np.reshape(dw2,w2.shape)    

    return dz2,dw2,dz1,dw1,db1,db2

for i in range(epochs):

    z1,a1,z2,a2 = forward_propagation(w1,w2,x,b1,b2)

    loss = -(1/m)*np.sum(y*np.log(a2)+(1-y)*np.log(1-a2))

    da2,dw2,dz1,dw1,db1,db2 = backward_propagation(m,w1,w2,z1,a1,z2,a2,y)

    w2 = w2-learning_rate*dw2

    w1 = w1-learning_rate*dw1

    b1 = b1-learning_rate*db1

    b2 = b2-learning_rate*db2

    output = np.where(a2 > 0.5, 1, 0) # setting threshold value 0.5 for prediction 

    print("Epoch--({}/{})------------------Loss--{}".format(i+1,epochs,loss))

    print("\nHidden layer \nweights ----- {}\nbiases ----- {}".format(w1,b1))

    print("\nOutput layer \nweights ----- {}\nbiases ----- {}".format(w2,b2))

    print("\nPredicted ----- {}".format(output))

    print("\n_______________________________________________________________________________________\n")

COMPUTER PROGRAMMING