help='no transfer blocks from input file to output')
parser.add_argument('-c',
'--covering',
dest='covering',
type=int,
default=1,
help='amount of column which should be covered')
parser.add_argument('-l',
'--result-limit',
dest='results_limit',
type=int,
default=10,
help='maximal amount of result')
args = parser.parse_args()
input_file = commonlib.DataFile()
input_file.load(args.input)
feature_count_penalty_koef = 1.0 / (input_file.features_count - args.covering +
1)
covering_penalty_koef = 1.0 - feature_count_penalty_koef
def fitness(individual):
res = 1 - (sum(individual)-1)/(input_file.features_count-1) * feature_count_penalty_koef\
- (input_file.uim_count - calc_cover(individual, args.covering)) / input_file.uim_count * covering_penalty_koef
return res,
creator.create("FitnessMax", base.Fitness, weights=(1.0, ))
creator.create("Individual", list, fitness=creator.FitnessMax)
#!/usr/bin/env python
# -*- coding: utf-8 -*-
import argparse
import sys
import commonlib
parser = argparse.ArgumentParser()
parser.add_argument('input', help='file for check')
parser.add_argument('reference', help='file with reference uim')
args = parser.parse_args()
input_file = commonlib.DataFile()
input_file.load(args.input)
reference_file = commonlib.DataFile()
reference_file.load(args.reference)
if input_file.features_count != reference_file.features_count:
print("feature_count differs")
sys.exit(1)
if input_file.uim_count != reference_file.uim_count:
print("uim_count differs")
sys.exit(1)
if input_file.uim_weights:
if input_file.uim_weights != reference_file.uim_weights:
print("uim_weights differs")
sys.exit(1)
for i in range(args.features):
features_count[i % len(features_count)] += 1
for i in range(len(features_count)):
for j in range(features_count[i]):
generator.add_feature(
genlib.Feature(genlib.Feature.kinds[i],
maximal=random.randint(1, args.maximal),
absence_probability=random.random() *
args.absence_probability))
generator.set_patterns_length(1)
for i in range(args.patterns):
generator.add_pattern(genlib.Pattern([i]))
generator.initialize(True)
features = []
pfeatures = []
for obj in generator.generate(args.objects):
features.append(obj[0])
pfeatures.append(obj[1])
datafile = commonlib.DataFile()
datafile.features = features
datafile.pfeatures = pfeatures
datafile.features_min = [x.minimal for x in generator.features]
datafile.features_max = [x.maximal for x in generator.features]
datafile.bake()
datafile.save(args.output)
# -*- coding: utf-8 -*-
import argparse
import json
import commonlib
import sys
parser = argparse.ArgumentParser()
parser.add_argument('input', help='input file for statistic calculation')
parser.add_argument('--format',
help='output format',
choices=['plain', 'json'],
default='plain')
args = parser.parse_args()
data = commonlib.DataFile()
data.load(args.input)
result = []
result.append(('lset_count', data.lset_count))
result.append(('features_count', data.features_count))
result.append(('pfeatures_count', data.pfeatures_count))
result.append(('patterns', len(data.patterns)))
result.append(('features_min', data.features_min))
result.append(('features_max', data.features_max))
#
skips = [0 for x in range(data.features_count)]
for obj in data.features:
for i in range(data.features_count):