Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly
Sign up
Appearance settings

Commit 01cb71c

Browse files
None solution index
1 parent b84d062 commit 01cb71c

File tree

2 files changed

+30
-20
lines changed

2 files changed

+30
-20
lines changed

‎docs/source/gann.rst

Lines changed: 13 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -649,6 +649,10 @@ its complete code is listed below.
649649
def fitness_func(ga_instance, solution, sol_idx):
650650
global GANN_instance, data_inputs, data_outputs
651651
652+
# If adaptive mutation is used, sometimes sol_idx is None.
653+
if sol_idx == None:
654+
sol_idx = 1
655+
652656
predictions = pygad.nn.predict(last_layer=GANN_instance.population_networks[sol_idx],
653657
data_inputs=data_inputs)
654658
correct_predictions = numpy.where(predictions == data_outputs)[0].size
@@ -659,7 +663,7 @@ its complete code is listed below.
659663
def callback_generation(ga_instance):
660664
global GANN_instance, last_fitness
661665
662-
population_matrices = pygad.gann.population_as_matrices(population_networks=GANN_instance.population_networks,
666+
population_matrices = pygad.gann.population_as_matrices(population_networks=GANN_instance.population_networks,
663667
population_vectors=ga_instance.population)
664668
665669
GANN_instance.update_population_trained_weights(population_trained_weights=population_matrices)
@@ -680,9 +684,9 @@ its complete code is listed below.
680684
[0, 0]])
681685
682686
# Preparing the NumPy array of the outputs.
683-
data_outputs = numpy.array([0,
684-
1,
685-
1,
687+
data_outputs = numpy.array([0,
688+
1,
689+
1,
686690
0])
687691
688692
# 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.
712716
713717
num_generations = 500 # Number of generations.
714718
715-
mutation_percent_genes = 5 # Percentage of genes to mutate. This parameter has no action if the parameter mutation_num_genes exists.
719+
mutation_percent_genes = [5, 10] # Percentage of genes to mutate. This parameter has no action if the parameter mutation_num_genes exists.
716720
717721
parent_selection_type = "sss" # Type of parent selection.
718722
719723
crossover_type = "single_point" # Type of the crossover operator.
720724
721-
mutation_type = "random" # Type of the mutation operator.
725+
mutation_type = "adaptive" # Type of the mutation operator.
722726
723727
keep_parents = 1 # Number of parents to keep in the next population. -1 means keep all parents and 0 means keep nothing.
724728
725729
init_range_low = -2
726730
init_range_high = 5
727731
728-
ga_instance = pygad.GA(num_generations=num_generations,
729-
num_parents_mating=num_parents_mating,
732+
ga_instance = pygad.GA(num_generations=num_generations,
733+
num_parents_mating=num_parents_mating,
730734
initial_population=initial_population,
731735
fitness_func=fitness_func,
732736
mutation_percent_genes=mutation_percent_genes,
@@ -736,6 +740,7 @@ its complete code is listed below.
736740
crossover_type=crossover_type,
737741
mutation_type=mutation_type,
738742
keep_parents=keep_parents,
743+
suppress_warnings=True,
739744
on_generation=callback_generation)
740745
741746
ga_instance.run()

‎examples/gann/example_XOR_classification.py

Lines changed: 17 additions & 12 deletions
Original file line numberDiff line numberDiff line change
@@ -6,6 +6,10 @@
66
def fitness_func(ga_instance, solution, sol_idx):
77
global GANN_instance, data_inputs, data_outputs
88

9+
# If adaptive mutation is used, sometimes sol_idx is None.
10+
if sol_idx == None:
11+
sol_idx = 1
12+
913
predictions = pygad.nn.predict(last_layer=GANN_instance.population_networks[sol_idx],
1014
data_inputs=data_inputs)
1115
correct_predictions = numpy.where(predictions == data_outputs)[0].size
@@ -16,16 +20,16 @@ def fitness_func(ga_instance, solution, sol_idx):
1620
def callback_generation(ga_instance):
1721
global GANN_instance, last_fitness
1822

19-
population_matrices = pygad.gann.population_as_matrices(population_networks=GANN_instance.population_networks,
23+
population_matrices = pygad.gann.population_as_matrices(population_networks=GANN_instance.population_networks,
2024
population_vectors=ga_instance.population)
2125

2226
GANN_instance.update_population_trained_weights(population_trained_weights=population_matrices)
2327

2428
print("Generation = {generation}".format(generation=ga_instance.generations_completed))
25-
print("Fitness = {fitness}".format(fitness=ga_instance.best_solution(pop_fitness=ga_instance.last_generation_fitness)[1]))
26-
print("Change = {change}".format(change=ga_instance.best_solution(pop_fitness=ga_instance.last_generation_fitness)[1] - last_fitness))
29+
print("Fitness = {fitness}".format(fitness=ga_instance.best_solution()[1]))
30+
print("Change = {change}".format(change=ga_instance.best_solution()[1] - last_fitness))
2731

28-
last_fitness = ga_instance.best_solution(pop_fitness=ga_instance.last_generation_fitness)[1].copy()
32+
last_fitness = ga_instance.best_solution()[1].copy()
2933

3034
# Holds the fitness value of the previous generation.
3135
last_fitness = 0
@@ -37,9 +41,9 @@ def callback_generation(ga_instance):
3741
[0, 0]])
3842

3943
# Preparing the NumPy array of the outputs.
40-
data_outputs = numpy.array([0,
41-
1,
42-
1,
44+
data_outputs = numpy.array([0,
45+
1,
46+
1,
4347
0])
4448

4549
# 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):
6973

7074
num_generations = 500 # Number of generations.
7175

72-
mutation_percent_genes = 5 # Percentage of genes to mutate. This parameter has no action if the parameter mutation_num_genes exists.
76+
mutation_percent_genes = [5, 10] # Percentage of genes to mutate. This parameter has no action if the parameter mutation_num_genes exists.
7377

7478
parent_selection_type = "sss" # Type of parent selection.
7579

7680
crossover_type = "single_point" # Type of the crossover operator.
7781

78-
mutation_type = "random" # Type of the mutation operator.
82+
mutation_type = "adaptive" # Type of the mutation operator.
7983

8084
keep_parents = 1 # Number of parents to keep in the next population. -1 means keep all parents and 0 means keep nothing.
8185

8286
init_range_low = -2
8387
init_range_high = 5
8488

85-
ga_instance = pygad.GA(num_generations=num_generations,
86-
num_parents_mating=num_parents_mating,
89+
ga_instance = pygad.GA(num_generations=num_generations,
90+
num_parents_mating=num_parents_mating,
8791
initial_population=initial_population,
8892
fitness_func=fitness_func,
8993
mutation_percent_genes=mutation_percent_genes,
@@ -93,6 +97,7 @@ def callback_generation(ga_instance):
9397
crossover_type=crossover_type,
9498
mutation_type=mutation_type,
9599
keep_parents=keep_parents,
100+
suppress_warnings=True,
96101
on_generation=callback_generation)
97102

98103
ga_instance.run()
@@ -101,7 +106,7 @@ def callback_generation(ga_instance):
101106
ga_instance.plot_fitness()
102107

103108
# Returning the details of the best solution.
104-
solution, solution_fitness, solution_idx = ga_instance.best_solution(pop_fitness=ga_instance.last_generation_fitness)
109+
solution, solution_fitness, solution_idx = ga_instance.best_solution()
105110
print("Parameters of the best solution : {solution}".format(solution=solution))
106111
print("Fitness value of the best solution = {solution_fitness}".format(solution_fitness=solution_fitness))
107112
print("Index of the best solution : {solution_idx}".format(solution_idx=solution_idx))

0 commit comments

Comments
(0)

AltStyle によって変換されたページ (->オリジナル) /