fork download
copy 
  1. import numpy as np
  2.  
  3. # Parameters
  4. learning_rate = 0.1
  5. max_iterations = 1000
  6. lower_threshold = 0.8  # Lower threshold for activation
  7. upper_threshold = 1.2  # Upper threshold for activation
  8.  
  9. # Threshold activation function with two thresholds
  10. def activation_function(weighted_sum):
  11.     if weighted_sum >= upper_threshold:
  12.         return 1
  13.     else:
  14.         return 0
  15.  
  16. # Generate all 16 possible truth tables for 2 inputs
  17. inputs = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])
  18. all_targets = [
  19.     [0, 0, 0, 0],  # 0
  20.     [0, 0, 0, 1],  # 1
  21.     [0, 0, 1, 0],  # 2
  22.     [0, 0, 1, 1],  # 3
  23.     [0, 1, 0, 0],  # 4
  24.     [0, 1, 0, 1],  # 5
  25.     [0, 1, 1, 0],  # 6
  26.     [0, 1, 1, 1],  # 7
  27.     [1, 0, 0, 0],  # 8
  28.     [1, 0, 0, 1],  # 9
  29.     [1, 0, 1, 0],  # 10
  30.     [1, 0, 1, 1],  # 11
  31.     [1, 1, 0, 0],  # 12
  32.     [1, 1, 0, 1],  # 13
  33.     [1, 1, 1, 0],  # 14
  34.     [1, 1, 1, 1],  # 15
  35. ]
  36.  
  37. # Initialize weights for the neurons in a smaller range
  38. weights_input_to_hidden = np.random.rand(3) * 0.1  # 2 inputs + 1 bias → 1 hidden neuron
  39. weights_hidden_to_output = np.random.rand(2) * 0.1  # 1 hidden neuron + 1 bias → 1 output neuron
  40.  
  41. # Training loop
  42. network_trained = False
  43. for target in all_targets:  # Train for all possible tables
  44.     epoch = 0
  45.     while epoch < max_iterations:
  46.         epoch += 1
  47.         all_correct = True  # Flag to track if all outputs are correct
  48.  
  49.         for input_vector, expected_target in zip(inputs, target):
  50.             N1, N2 = input_vector
  51.             bias = 1  # Bias input
  52.  
  53.             # Forward pass (hidden layer)
  54.             N3_input = np.dot(np.append(input_vector, bias), weights_input_to_hidden)  # Hidden layer sum
  55.             N3 = activation_function(N3_input)  # Output of the hidden neuron
  56.  
  57.             # Output layer
  58.             N4_input = (N3 * weights_hidden_to_output[0] + bias * weights_hidden_to_output[1])  # Weighted sum to output
  59.             N4 = activation_function(N4_input)  # Output neuron decision
  60.  
  61.             # Error calculation
  62.             error = expected_target - N4
  63.  
  64.             # If there is an error, update the weights
  65.  
  66.  
Success #stdin #stdout 1.18s 28732KB
comments (?)
stdin
copy 
Standard input is empty
stdout
copy 
Standard output is empty
https://ideone.com/hBrmme
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!