fuzzy_logic_classifier.prepare_data(
k_fold_number=K_FOLD_NUMBER)
rough_set_classifier.prepare_data(k_fold_number=K_FOLD_NUMBER)
fuzzy_logic_classifier.k_fold_cross_validation(k=k)
fuzzy_logic_classifier.generate_membership_functions(
divisions=DIVISIONS)
fuzzy_logic_classifier.initialize_genetic(
generations=200,
mutation=MUTATION_PROP,
crossover=CROSS_OVER_PROP)
fuzzy_logic_classifier.create_population(population_size=20)
fuzzy_logic_classifier.run()
rough_set_classifier.k_fold_cross_validation(k=k)
rough_set_classifier.initialize_genetic(
generations=200,
mutation_prop=MUTATION_PROP,
crossover_prop=CROSS_OVER_PROP)
rough_set_classifier.create_population(
population_size=POPULATION_SIZE, division=MAX_GRANULATION)
rough_set_classifier.run()
# classify
size = len(fuzzy_logic_classifier.testing_data)
classified = 0
for i in range(size):
ret = rough_set_classifier.classify(
fuzzy_logic_classifier.testing_data[i])
if ret == PATTERN_REJECTED:
#!/usr/bin/python
# -*- coding: utf-8 -*-
from genetic_rough_sets_classifier import GeneticRoughSetsClassifier
if __name__ == '__main__':
K_FOLD_NUMBER = 4
K = 1
GENERATIONS = 500
MUTATION = 0.3
CROSS_OVER = 0.9
G = 7
POPULATION = 10
grs = GeneticRoughSetsClassifier(debug=DEBUG)
filename = 'datasets/iris.data.txt'
grs.read_data(filepath=filename, label_is_last=False)
grs.prepare_data(k_fold_number=K_FOLD_NUMBER)
grs.k_fold_cross_validation(k=K)
grs.initialize_genetic(generations=GENERATIONS, mutation_prop=MUTATION, crossover_prop=CROSS_OVER)
grs.create_population(population_size=POPULATION, division=G)
# run genetic algorithm
grs.run()
# genetic algorithm found granulation
# now classify patterns
classified = 0
for i in range(len(patterns)):
if labels[i] == grs.classify(patterns[i]):
classified += 1
grs.print_summary()
result = fuzzy_logic_classifier.read_data(filepath=filename, label_is_last=(bool)(datasets[d][1]))
result = rough_set_classifier.read_data(filepath=filename, label_is_last=(bool)(datasets[d][1]))
if not result:
logging.error("Wrong dataset name")
sys.exit(1)
fuzzy_logic_classifier.prepare_data(k_fold_number=K_FOLD_NUMBER)
rough_set_classifier.prepare_data(k_fold_number=K_FOLD_NUMBER)
fuzzy_logic_classifier.k_fold_cross_validation(k=k)
fuzzy_logic_classifier.generate_membership_functions(divisions=DIVISIONS)
fuzzy_logic_classifier.initialize_genetic(generations=200, mutation=MUTATION_PROP, crossover=CROSS_OVER_PROP)
fuzzy_logic_classifier.create_population(population_size=20)
fuzzy_logic_classifier.run()
rough_set_classifier.k_fold_cross_validation(k=k)
rough_set_classifier.initialize_genetic(generations=200, mutation_prop=MUTATION_PROP, crossover_prop=CROSS_OVER_PROP)
rough_set_classifier.create_population(population_size=POPULATION_SIZE, division=MAX_GRANULATION)
rough_set_classifier.run()
# classify
size = len(fuzzy_logic_classifier.testing_data)
classified = 0
for i in range(size):
ret = rough_set_classifier.classify(fuzzy_logic_classifier.testing_data[i])
if ret == PATTERN_REJECTED:
if fuzzy_logic_classifier.testing_label[i] == fuzzy_logic_classifier.classify(fuzzy_logic_classifier.testing_data[i]):
classified += 1
elif ret == fuzzy_logic_classifier.testing_label[i]:
classified += 1
print "FINISHED iteration %d for %d-fold %s dataset" % (r, k, datasets[d][0])
