In this chapter, we will learn about the XOR implementation using TensorFlow. Before starting with XOR implementation in TensorFlow, let us see the XOR table values. This will help us understand encryption and decryption process.
A | B | A XOR B |
0 | 0 | 0 |
0 | 1 | 1 |
1 | 0 | 1 |
1 | 1 | 0 |
XOR Cipher encryption method is basically used to encrypt data which is hard to crack with brute force method, i.e., by generating random encryption keys which match the appropriate key.
The concept of implementation with XOR Cipher is to define a XOR encryption key and then perform XOR operation of the characters in the specified string with this key, which a user tries to encrypt. Now we will focus on XOR implementation using TensorFlow, which is mentioned below −
#Declaring necessary modules import tensorflow as tf import numpy as np """ A simple numpy implementation of a XOR gate to understand the backpropagation algorithm """ x = tf.placeholder(tf.float64,shape = [4,2],name = "x") #declaring a place holder for input x y = tf.placeholder(tf.float64,shape = [4,1],name = "y") #declaring a place holder for desired output y m = np.shape(x)[0]#number of training examples n = np.shape(x)[1]#number of features hidden_s = 2 #number of nodes in the hidden layer l_r = 1#learning rate initialization theta1 = tf.cast(tf.Variable(tf.random_normal([3,hidden_s]),name = "theta1"),tf.float64) theta2 = tf.cast(tf.Variable(tf.random_normal([hidden_s+1,1]),name = "theta2"),tf.float64) #conducting forward propagation a1 = tf.concat([np.c_[np.ones(x.shape[0])],x],1) #the weights of the first layer are multiplied by the input of the first layer z1 = tf.matmul(a1,theta1) #the input of the second layer is the output of the first layer, passed through the activation function and column of biases is added a2 = tf.concat([np.c_[np.ones(x.shape[0])],tf.sigmoid(z1)],1) #the input of the second layer is multiplied by the weights z3 = tf.matmul(a2,theta2) #the output is passed through the activation function to obtain the final probability h3 = tf.sigmoid(z3) cost_func = -tf.reduce_sum(y*tf.log(h3)+(1-y)*tf.log(1-h3),axis = 1) #built in tensorflow optimizer that conducts gradient descent using specified learning rate to obtain theta values optimiser = tf.train.GradientDescentOptimizer(learning_rate = l_r).minimize(cost_func) #setting required X and Y values to perform XOR operation X = [[0,0],[0,1],[1,0],[1,1]] Y = [[0],[1],[1],[0]] #initializing all variables, creating a session and running a tensorflow session init = tf.global_variables_initializer() sess = tf.Session() sess.run(init) #running gradient descent for each iteration and printing the hypothesis obtained using the updated theta values for i in range(100000): sess.run(optimiser, feed_dict = {x:X,y:Y})#setting place holder values using feed_dict if i%100==0: print("Epoch:",i) print("Hyp:",sess.run(h3,feed_dict = {x:X,y:Y}))
The above line of code generates an output as shown in the screenshot below −