def data_two(dirname):
data_path = os.path.join(dirname, "data/dataset2.txt")
data = open_and_sanitize_data(data_path)
ga = ARBGA(dataset=data,
mutation_probability=0.005,
population_size=100,
rule_count=30)
ga.evolve(epoch=5000)
def data_one(dirname):
data_path = os.path.join(dirname, "data/dataset1.txt")
data = open_and_sanitize_data(data_path)
ga = RuleBasedGeneticAlgorithm(dataset=data,
mutation_probability=0.0047,
crossover_probability=0.9,
population_size=100,
rule_count=30)
ga.evolve(epochs=5000)
def data_three(dirname):
data_path = os.path.join(dirname, "data/dataset3.txt")
data = open_and_sanitize_data(data_path)
training_set = data[0::2]
test_set = data[1::2]
ga = RuleBasedFloatingPointGA(dataset=training_set,
test_data=test_set,
mutation_probability=0.01,
crossover_probability=0.9,
population_size=100,
rule_count=5)
ga.evolve(epoch=2500)
def data_four(dirname):
data_path = os.path.join(dirname, "data/dataset4.txt")
data = open_and_sanitize_data(data_path)
training_set = data[0::2]
test_set = data[1::2]
training_set = training_set + test_set[0::2]
test_set = test_set[1::2]
# training_set = data[0::2]
# test_set = data[1::2]
ga = RuleBasedFloatingPointGA(dataset=training_set,
test_data=test_set,
mutation_probability=0.008,
crossover_probability=0.9,
population_size=50,
rule_count=5,
seed_edge_cases=True)
ga.evolve(epoch=10000)
def evaluate_rules(dirname):
data_two = os.path.join(dirname, "data/dataset2.txt")
rules = os.path.join(dirname, "observations/data2rules.txt")
data_two = open_and_sanitize_data(data_two)
data_two = initialise_data(data_two)[0]
with open(rules, mode="r") as f:
data = f.readlines()
for i in range(len(data)):
line = data[i]
data[i] = line.strip("\n")
data = [(d[:len(d) - 1], d[-1]) for d in data]
unmatched = []
for bd in data_two:
fm = False
for (x, y) in data:
if (match(bd.feature, x) and bd.label == y):
fm = True
break
if (fm == False):
unmatched.append(bd)
print(unmatched)