fork download
copy 
  1. import numpy as np
  2.  
  3. # Parameters
  4. learning_rate = 0.1
  5. max_iterations = 1000
  6.  
  7. # Activation function with thresholding
  8. def activation_function(weighted_sum, threshold=0.5):
  9.     return 1 if weighted_sum > threshold else 0
  10.  
  11. # Inputs for XOR truth table
  12. inputs = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])
  13. targets = [0, 1, 1, 0]  # XOR Truth Table
  14.  
  15. # Initialize weights for 2 hidden neurons and 1 output neuron
  16. weights_input_to_hidden = np.random.rand(6)  # 2 inputs, 1 bias → 2 hidden neurons (3 weights for each hidden)
  17. weights_hidden_to_output = np.random.rand(3)  # 2 hidden neurons + bias → 1 output neuron
  18.  
  19. # Training loop
  20. epoch = 0
  21. network_trained = False
  22.  
  23. while epoch < max_iterations:
  24.     epoch += 1
  25.     all_correct = True  # Flag to track if all outputs are correct
  26.  
  27.     for input_vector, target in zip(inputs, targets):
  28.         N1, N2 = input_vector
  29.         bias = 1  # Bias input
  30.  
  31.         # Forward pass (for both hidden neurons)
  32.         N1_input = np.dot(np.append(input_vector, bias), weights_input_to_hidden[:3])
  33.         N1_hidden = activation_function(N1_input)  # Output of first hidden neuron
  34.  
  35.         N2_input = np.dot(np.append(input_vector, bias), weights_input_to_hidden[3:])
  36.         N2_hidden = activation_function(N2_input)  # Output of second hidden neuron
  37.  
  38.         # Combine hidden neuron outputs and pass through to output
  39.         N4_input = (N1_hidden * weights_hidden_to_output[0] +
  40.                     N2_hidden * weights_hidden_to_output[1] +
  41.                     bias * weights_hidden_to_output[2])  # Hidden to output
  42.         N4 = activation_function(N4_input)  # Final output
  43.  
  44.         # Error calculation
  45.         error = target - N4
  46.  
  47.         # Weight updates if error exists
  48.         if error != 0:
  49.             all_correct = False
  50.  
  51.             # Update weights for N1_hidden → N4
  52.             weights_hidden_to_output[0] += learning_rate * error * N1_hidden
  53.             weights_hidden_to_output[1] += learning_rate * error * N2_hidden
  54.             weights_hidden_to_output[2] += learning_rate * error * bias
  55.  
  56.             # Update weights for input → hidden neurons
  57.             weights_input_to_hidden[:3] += learning_rate * error * N1_hidden * np.append(input_vector, bias)
  58.             weights_input_to_hidden[3:] += learning_rate * error * N2_hidden * np.append(input_vector, bias)
  59.  
  60.     if all_correct:
  61.         network_trained = True
  62.         break
  63.  
  64. # Results
  65. if network_trained:
  66.     print(f"The network learned the XOR truth table correctly after {epoch} iterations.")
  67. else:
  68.     print(f"The network failed to learn the XOR truth table after {epoch} iterations.")
  69.  
  70. # Testing the trained network
  71. print("\nTesting the trained network:")
  72. for input_vector, target in zip(inputs, targets):
  73.     N1, N2 = input_vector
  74.     bias = 1
  75.  
  76.     # Forward pass (for both hidden neurons)
  77.     N1_input = np.dot(np.append(input_vector, bias), weights_input_to_hidden[:3])
  78.     N1_hidden = activation_function(N1_input)
  79.  
  80.     N2_input = np.dot(np.append(input_vector, bias), weights_input_to_hidden[3:])
  81.     N2_hidden = activation_function(N2_input)
  82.  
  83.     # Combine hidden neuron outputs and pass through to output
  84.     N4_input = (N1_hidden * weights_hidden_to_output[0] +
  85.                 N2_hidden * weights_hidden_to_output[1] +
  86.                 bias * weights_hidden_to_output[2])
  87.     N4 = activation_function(N4_input)
  88.  
  89.     print(f"Input: {input_vector}, Target: {target}, Output: {N4}")
  90.  
Success #stdin #stdout 0.39s 28888KB
comments (?)
stdin
copy 
Standard input is empty
stdout
copy 
The network failed to learn the XOR truth table after 1000 iterations.

Testing the trained network:
Input: [0 0], Target: 0, Output: 0
Input: [0 1], Target: 1, Output: 1
Input: [1 0], Target: 1, Output: 0
Input: [1 1], Target: 0, Output: 0
https://ideone.com/gtA20d
language:
Python 3 (python  3.9.5)
created:
7 hours ago
visibility:
public

Share or Embed source code

Discover > Sphere Engine API

The brand new service which powers Ideone!

Discover > IDE Widget

Widget for compiling and running the source code in a web browser!