Merge pull request #226 from ahmedfgad/github-actions

None index with adaptive mutation

Author: ahmedfgad

docs/source/gann.rst

@@ -649,6 +649,10 @@ its complete code is listed below.
    def fitness_func(ga_instance, solution, sol_idx):
        global GANN_instance, data_inputs, data_outputs
 
+       # If adaptive mutation is used, sometimes sol_idx is None.
+       if sol_idx == None:
+           sol_idx = 1
+
        predictions = pygad.nn.predict(last_layer=GANN_instance.population_networks[sol_idx],
                                       data_inputs=data_inputs)
        correct_predictions = numpy.where(predictions == data_outputs)[0].size
@@ -659,7 +663,7 @@ its complete code is listed below.
    def callback_generation(ga_instance):
        global GANN_instance, last_fitness
 
-       population_matrices = pygad.gann.population_as_matrices(population_networks=GANN_instance.population_networks, 
+       population_matrices = pygad.gann.population_as_matrices(population_networks=GANN_instance.population_networks,
                                                                population_vectors=ga_instance.population)
 
        GANN_instance.update_population_trained_weights(population_trained_weights=population_matrices)
@@ -680,9 +684,9 @@ its complete code is listed below.
                               [0, 0]])
 
    # Preparing the NumPy array of the outputs.
-   data_outputs = numpy.array([0, 
-                               1, 
-                               1, 
+   data_outputs = numpy.array([0,
+                               1,
+                               1,
                                0])
 
    # The length of the input vector for each sample (i.e. number of neurons in the input layer).
@@ -712,21 +716,21 @@ its complete code is listed below.
 
    num_generations = 500 # Number of generations.
 
-   mutation_percent_genes = 5 # Percentage of genes to mutate. This parameter has no action if the parameter mutation_num_genes exists.
+   mutation_percent_genes = [5, 10] # Percentage of genes to mutate. This parameter has no action if the parameter mutation_num_genes exists.
 
    parent_selection_type = "sss" # Type of parent selection.
 
    crossover_type = "single_point" # Type of the crossover operator.
 
-   mutation_type = "random" # Type of the mutation operator.
+   mutation_type = "adaptive" # Type of the mutation operator.
 
    keep_parents = 1 # Number of parents to keep in the next population. -1 means keep all parents and 0 means keep nothing.
 
    init_range_low = -2
    init_range_high = 5
 
-   ga_instance = pygad.GA(num_generations=num_generations, 
-                          num_parents_mating=num_parents_mating, 
+   ga_instance = pygad.GA(num_generations=num_generations,
+                          num_parents_mating=num_parents_mating,
                           initial_population=initial_population,
                           fitness_func=fitness_func,
                           mutation_percent_genes=mutation_percent_genes,
@@ -736,6 +740,7 @@ its complete code is listed below.
                           crossover_type=crossover_type,
                           mutation_type=mutation_type,
                           keep_parents=keep_parents,
+                          suppress_warnings=True,
                           on_generation=callback_generation)
 
    ga_instance.run()

examples/gann/example_XOR_classification.py

@@ -6,6 +6,10 @@
 def fitness_func(ga_instance, solution, sol_idx):
     global GANN_instance, data_inputs, data_outputs
 
+    # If adaptive mutation is used, sometimes sol_idx is None.
+    if sol_idx == None:
+        sol_idx = 1
+
     predictions = pygad.nn.predict(last_layer=GANN_instance.population_networks[sol_idx],
                                    data_inputs=data_inputs)
     correct_predictions = numpy.where(predictions == data_outputs)[0].size
@@ -16,16 +20,16 @@ def fitness_func(ga_instance, solution, sol_idx):
 def callback_generation(ga_instance):
     global GANN_instance, last_fitness
 
-    population_matrices = pygad.gann.population_as_matrices(population_networks=GANN_instance.population_networks, 
+    population_matrices = pygad.gann.population_as_matrices(population_networks=GANN_instance.population_networks,
                                                             population_vectors=ga_instance.population)
 
     GANN_instance.update_population_trained_weights(population_trained_weights=population_matrices)
 
     print("Generation = {generation}".format(generation=ga_instance.generations_completed))
-    print("Fitness    = {fitness}".format(fitness=ga_instance.best_solution(pop_fitness=ga_instance.last_generation_fitness)[1]))
-    print("Change     = {change}".format(change=ga_instance.best_solution(pop_fitness=ga_instance.last_generation_fitness)[1] - last_fitness))
+    print("Fitness    = {fitness}".format(fitness=ga_instance.best_solution()[1]))
+    print("Change     = {change}".format(change=ga_instance.best_solution()[1] - last_fitness))
 
-    last_fitness = ga_instance.best_solution(pop_fitness=ga_instance.last_generation_fitness)[1].copy()
+    last_fitness = ga_instance.best_solution()[1].copy()
 
 # Holds the fitness value of the previous generation.
 last_fitness = 0
@@ -37,9 +41,9 @@ def callback_generation(ga_instance):
                            [0, 0]])
 
 # Preparing the NumPy array of the outputs.
-data_outputs = numpy.array([0, 
-                            1, 
-                            1, 
+data_outputs = numpy.array([0,
+                            1,
+                            1,
                             0])
 
 # The length of the input vector for each sample (i.e. number of neurons in the input layer).
@@ -69,21 +73,21 @@ def callback_generation(ga_instance):
 
 num_generations = 500 # Number of generations.
 
-mutation_percent_genes = 5 # Percentage of genes to mutate. This parameter has no action if the parameter mutation_num_genes exists.
+mutation_percent_genes = [5, 10] # Percentage of genes to mutate. This parameter has no action if the parameter mutation_num_genes exists.
 
 parent_selection_type = "sss" # Type of parent selection.
 
 crossover_type = "single_point" # Type of the crossover operator.
 
-mutation_type = "random" # Type of the mutation operator.
+mutation_type = "adaptive" # Type of the mutation operator.
 
 keep_parents = 1 # Number of parents to keep in the next population. -1 means keep all parents and 0 means keep nothing.
 
 init_range_low = -2
 init_range_high = 5
 
-ga_instance = pygad.GA(num_generations=num_generations, 
-                       num_parents_mating=num_parents_mating, 
+ga_instance = pygad.GA(num_generations=num_generations,
+                       num_parents_mating=num_parents_mating,
                        initial_population=initial_population,
                        fitness_func=fitness_func,
                        mutation_percent_genes=mutation_percent_genes,
@@ -93,6 +97,7 @@ def callback_generation(ga_instance):
                        crossover_type=crossover_type,
                        mutation_type=mutation_type,
                        keep_parents=keep_parents,
+                       suppress_warnings=True,
                        on_generation=callback_generation)
 
 ga_instance.run()
@@ -101,7 +106,7 @@ def callback_generation(ga_instance):
 ga_instance.plot_fitness()
 
 # Returning the details of the best solution.
-solution, solution_fitness, solution_idx = ga_instance.best_solution(pop_fitness=ga_instance.last_generation_fitness)
+solution, solution_fitness, solution_idx = ga_instance.best_solution()
 print("Parameters of the best solution : {solution}".format(solution=solution))
 print("Fitness value of the best solution = {solution_fitness}".format(solution_fitness=solution_fitness))
 print("Index of the best solution : {solution_idx}".format(solution_idx=solution_idx))

pygad/utils/mutation.py

@@ -476,6 +476,9 @@ def adaptive_mutation_population_fitness(self, offspring):
         if self.fitness_batch_size in [1, None]:
             # Calculate the fitness for each individual solution.
             for idx in range(first_idx, last_idx):
+                # We cannot return the index of the solution within the population.
+                # Because the new solution (offspring) does not yet exist in the population.
+                # The user should handle this situation if the solution index is used anywhere.
                 fitness[idx] = self.fitness_func(self, 
                                                  temp_population[idx], 
                                                  None)
@@ -495,6 +498,9 @@ def adaptive_mutation_population_fitness(self, offspring):
                     batch_last_index = first_idx + (batch_idx + 1) * self.fitness_batch_size
 
                 # Calculate the fitness values for the batch.
+                # We cannot return the index/indices of the solution(s) within the population.
+                # Because the new solution(s) (offspring) do(es) not yet exist in the population.
+                # The user should handle this situation if the solution index is used anywhere.
                 fitness_temp = self.fitness_func(self, 
                                                  temp_population[batch_first_index:batch_last_index], 
                                                  None)