finished??

03_euler_gen_alg/main.py

@@ -13,7 +13,6 @@ import random
 import struct
 import time
 import utils
-# import matplotlib.pyplot as plt
 
 POPULATION_SIZE = 10
 SELECTION_SIZE = (POPULATION_SIZE * 7) // 10  # 70% of population, rounded down for selection
@@ -28,8 +27,6 @@ bin_pop = [] # 32 Bit-Binary as String
 bin_pop_params = [] # Arrays with 4 Binary values of 8s
 new_pop = [] # 32 Bit Gray-Code as String
 
-e_func = lambda x: np.e**x
-
 def generate_random_population(num=POPULATION_SIZE):
     """ Puts random 32 Bit binary strings into 4 * 8 Bit long params. """
         
@@ -63,12 +60,12 @@ def eval_fitness(bin_pop_values):
         
         # Create polynomial function with current parameters
         approx = lambda x: a*x**3 + b*x**2 + c*x + d
-        quad_error = quadratic_error(e_func, approx, 6)
+        e_func = lambda x: np.e**x
+        quad_error = quadratic_error(e_func, approx, 3)
         
         inverse_fitness = 1 / quad_error # the bigger the error, the worse the fitness
-        print("Fitness: " + str(inverse_fitness)) # debugging
+        # print("Fitness: " + str(inverse_fitness)) # debugging
         fitness_arr.append(inverse_fitness) # save fitness
-        # save params # already saved in gray_pop
         
     return fitness_arr
 
@@ -96,14 +93,13 @@ def select(fitness_arr):
                     break # break for loop
     return selected_pop
      
-# TODO: xover the old population not the new one      
 def xover(population):
     """Performs crossover on pairs of individuals from population."""
     offspring = []
     
     # Process pairs while we have enough individuals and haven't reached xover_rate
     i = 0
-    while i < len(population) - 1 and len(population) + len(offspring) < 10:
+    while i < XOVER_PAIR_SIZE:
         parent_a = population[i]
         parent_b = population[i + 1]
         
@@ -139,8 +135,9 @@ def main():
     bin_pop_values = generate_random_population(POPULATION_SIZE)
     
     iteration = 0
-    while not np.all((1 / np.array(fitness_arr)) <= 1):  # Continue while any fitness value is > 1
-        print("Iteration: " + str(iteration)) # debugging
+    print("Working...")
+    while not np.any((np.array(fitness_arr)) > 200):  # Continue while any fitness value is > 1
+        # print("Iteration: " + str(iteration)) # debugging
         
         # Evaluate fitness
         fitness_arr = eval_fitness(bin_pop_values)
@@ -149,7 +146,7 @@ def main():
         new_pop = select(fitness_arr) # assigns
                 
         # Crossover        
-        offspring = xover(new_pop)
+        offspring = xover(gray_pop)
         new_pop.extend(offspring)  # Add offspring to population
         
         # Mutation
@@ -163,10 +160,19 @@ def main():
             params = [bin_str[i:i+7] for i in range(0, 31, 8)]
             bin_pop_values.append(params)
         
-        # print(new_pop)
         # time.sleep(0.5)
         
         iteration += 1
+    
+    max_fitness_index = np.argmax(np.array(fitness_arr))
+    a, b, c, d = [utils.bin_to_param(param) for param in bin_pop_values[max_fitness_index]]
+    
+    
+    print("index: " + str(max_fitness_index))
+    print("a: " + str(a) + "; b: " + str(b) + "; c: " + str(c) + "; d: " + str(d) )
+    
+    utils.plot_graph(a, b, c, d)
+    
     return 0
 
 if __name__ == "__main__":

03_euler_gen_alg/utils.py

@@ -1,3 +1,6 @@
+import matplotlib.pyplot as plt
+import numpy as np
+
 def gray_to_bin(gray):
     """
     Convert Gray code to binary, operating on the integer value directly.
@@ -40,3 +43,16 @@ def bin_to_param(binary, q_min = 0.0, q_max = 10.0):
     except ValueError as e:
         print(f"Error in bin_to_param with input: '{binary}'")
         raise e
+
+def plot_graph(a, b, c, d): 
+    x = np.arange(-5., 5., 0.1)
+    
+    fig, ax = plt.subplots()
+    y_approx = a*x**3 + b*x**2 + c*x + d
+    y_exact = np.e**x
+    ax.plot(x, y_approx, label='approx. func.')
+    ax.plot(x, y_exact, label='e^x')
+    ax.set_xlim(-5, 5)
+    ax.set_ylim(-1, 5)
+    plt.legend()
+    plt.show()