GA implementation

- DONE best results
- DONE parents selection

with Michał Malinowski

AI/GeneticAlgorithm.py

@@ -15,7 +15,7 @@ def genetic_algorithm_setup(field):
     population_text = []
     population_text_single = []
 
-    population_size = 3
+    population_size = 10
 
     # Populate the population_text array
     for k in range(population_size):
@@ -26,9 +26,9 @@ def genetic_algorithm_setup(field):
                 population_text_single[row].append(random.choice(population_units))
         population_text.append(population_text_single)
 
-    # printer
+    # # printer
-    for _ in population_text:
+    # for _ in population_text:
-        print(population_text)
+    #     print(population_text)
 
     """
     Genetic algorithm parameters:
@@ -41,7 +41,9 @@ def genetic_algorithm_setup(field):
 
     best_outputs = []
     num_generations = 10
+    num_parents = 2
 
+    # iterative var
     generation = 0
 
     while generation < num_generations:
@@ -52,18 +54,24 @@ def genetic_algorithm_setup(field):
             # Measuring the fitness of each chromosome in the population.
 
             # population Fitness
-            fitness = population_fitness(population_text, field, population_size)
+            fitness = []
+            for i in range(0, population_size):
+                fitness.append((i, population_fitness(population_text[i], field, population_size)))
 
             print("Fitness")
             print(fitness)
 
-            best_outputs.append(best_Output(new_population))
+            best = sorted(fitness, key=lambda tup: tup[1])[0:num_parents]
-            # The best result in the current iteration.
-            print("Best result : ", best_Output(new_population))
 
+            # Leaderboard only
+            best_outputs.append(best[0][1])
+
+            # The best result in the current iteration.
+            print("Best result : ", best[0])
+
+            # TODO METODA WYBORU OSOBNIKA - RANKING
             # Selecting the best parents in the population for mating.
-            parents = select_mating_pool(new_population, fitness,
+            parents = [population_text[i[0]] for i in best]
-                                         num_parents_mating)
             print("Parents")
             print(parents)
 

AI/ga_methods.py

@@ -24,7 +24,6 @@ def local_fitness(field, x, y, plants_case):
         plant_value = 1
 
     neighbour_bonus = 1
-    print(D.GSIZE, x, y)
     if x - 1 >= 0:
         if plants_case[x][y] == plants_case[x - 1][y]:
             neighbour_bonus += 1
@@ -60,21 +59,10 @@ def population_fitness(population_text, field, population_size):
 
         for i in range(D.GSIZE):
             fitness_row.append(sum(population_values_single[i]))
-        fitness.append(sum(fitness_row))
+        fitness = sum(fitness_row)
     return fitness
 
 
-def select_mating_pool(pop, fitness, num_parents):
-    # Selecting the best individuals in the current generation as parents for producing the offspring of the next generation.
-    parents = numpy.empty((num_parents, pop.shape[1]))
-    for parent_num in range(num_parents):
-        max_fitness_idx = numpy.where(fitness == numpy.max(fitness))
-        max_fitness_idx = max_fitness_idx[0][0]
-        parents[parent_num, :] = pop[max_fitness_idx, :]
-        fitness[max_fitness_idx] = -99999999999
-    return parents
-
-
 def crossover(parents, offspring_size):
     offspring = numpy.empty(offspring_size)
     # The point at which crossover takes place between two parents. Usually, it is at the center.
@@ -105,5 +93,3 @@ def mutation(offspring_crossover, num_mutations=1):
     return offspring_crossover
 
 
-def best_Output(new_population):
-    return numpy.max(numpy.sum(new_population * equation_inputs, axis=1))