rna = RnaModule()
rna.setNumberNeuronsImputLayer(30)
rna.setActivationFunctionImputLayer("tanh")
rna.setImputDimNeurons(30)
rna.setNumberNeuronsHiddenLayer(31)
rna.setActivationFunctionHiddenLayer("tanh")
rna.setNumberNeuronsOutputLayer(1)
rna.setActivationFunctionOutputLayer("tanh")
rna_classifier = RnaClassifier()
rna_classifier.setRna(rna)
#PREPROCESSADOR PARA ATRIBUTOS CATEGORICOS
preprocessor = Preprocessor()
preprocessor.setColumnsCategory(['protocol_type', 'service', 'flag'])
evaluate = EvaluateModule()
cross = CrossValidation()
#DEFINIR A ITERACAO QUE O CROSS VALIDATION ESTA
cross.setIteration(1)
cross.setPreprocessor(preprocessor)
cross.setFilePath("../../bases/sub_bases_nslkdd_30_attribute/")
cross.setResultPath("../../results/30_attribute/rna_oculta_31_time/")
cross.setClassifier(rna_classifier)
cross.setEvaluateModule(evaluate)
def __init__(self):
print("init")
self.evaluate = EvaluateModule()
def __init__(self):
# print("Cross Validation constructor")
self.evaluate = EvaluateModule()
class CrossValidation(object):
dts = None
#metodo utilizado para classifacao
classifier = None
#conjunto de dados de teste
teste_sub_data_set = None
#conjunto de dados de treinamento
training_sub_data_set = None
evaluate = None
#numero de folds
k = 1
file_path = ""
class_name = ""
#caminho da pasta onde serao salvos os resultados
result_path = ""
preprocessor = None
def __init__(self):
print("init")
self.evaluate = EvaluateModule()
def run(self):
self.classifier.setResultPath(self.result_path)
self.classifier.setClass_name(self.class_name)
self.foldExecution()
def foldExecution(self):
i = self.iteration
for self.iteration in range(i, (self.k + 1)):
tempo_inicio = time.time()
self.loadTrainingData()
self.loadTestData()
#executa funcoes para transformacao de dados categoricos
if self.preprocessor:
self.preprocessor.setDataSet(self.training_sub_data_set)
self.preprocessor.setTestDataSet(self.teste_sub_data_set)
self.training_sub_data_set, self.teste_sub_data_set = self.preprocessor.transformCategory(
)
#seta dados de treinamento e teste no classificador
self.classifier.setDataSet(self.training_sub_data_set)
self.classifier.setTestDataSet(self.teste_sub_data_set)
#seta iteracao do cross no classficador
self.classifier.setIteration(self.iteration)
#executa o processo de treino e teste do classificador
self.classifier.run()
del (self.training_sub_data_set)
self.loadTestData()
#seta conjunto de dados original de teste e iteracao atual do cross-validation na classe de avaliacao
self.evaluate.setTestDataSet(self.teste_sub_data_set)
self.evaluate.setIteration(self.iteration)
self.evaluate.setClass_name(self.class_name)
'''
#verifica quel o metodo de classificacao utilziado
if(isinstance(self.classifier, RnaClassifier)):
print("rna")
self.evaluate.setResultPath( self.result_path)
elif(isinstance(self.classifier, KnnClassifier)):
print("knn")
self.evaluate.setResultPath(self.result_path)
elif(isinstance(self.classifier, SvmClassifier)):
print("svm")
self.evaluate.setResultPath(self.result_path)
elif(isinstance(self.classifier, RfClassifier)):
print("rf")
self.evaluate.setResultPath(self.result_path)
elif(isinstance(self.classifier, ClusteredKnnClassifier)):
print("clustered knn")
#self.evaluate.setResultPath(self.result_path)
elif(isinstance(self.classifier, ClusteredDensityKnnClassifier)):
print("clustered density knn")
#self.evaluate.setResultPath(self.result_path)
'''
if (isinstance(self.classifier, HybridClassifier)):
print("hybrid")
self.evaluate.setResultPath(self.result_path +
"final_method_classification/")
else:
self.evaluate.setResultPath(self.result_path)
tempo_execucao = time.time() - tempo_inicio
self.evaluate.setTempoExecucao(tempo_execucao)
self.evaluate.setTrainingTime(self.classifier.getTrainingTime())
self.evaluate.setTestTime(self.classifier.getTestTime())
#executa metodo de avaliacao
self.evaluate.run()
#carrega conjunto de treinamento de acordo coma iteracao atual do cross valiadation
def loadTrainingData(self):
for i in range(1, (self.k + 1)):
if (((self.k + 1) - i) != self.iteration):
new_sub_data_set = DataSet.loadSubDataSet(self.file_path +
"sub_data_set_" +
str(i) + ".csv")
if (i == 1):
self.training_sub_data_set = new_sub_data_set
else:
self.training_sub_data_set = DataSet.concatSubDataSet(
self.training_sub_data_set, new_sub_data_set)
del (new_sub_data_set)
#self.training_sub_data_set = self.training_sub_data_set.reset_index()
print(self.training_sub_data_set)
#carrega conjunto de teste de acordo coma iteracao atual do cross valiadation
def loadTestData(self):
self.teste_sub_data_set = DataSet.loadSubDataSet(self.file_path +
"sub_data_set_" +
str((self.k + 1) -
self.iteration) +
".csv")
print(self.teste_sub_data_set)
#self.teste_sub_data_set = self.teste_sub_data_set.reset_index()
#print(self.teste_sub_data_set)
#exit()
def setIteration(self, iteration):
self.iteration = iteration
def setClassifier(self, classifier):
self.classifier = classifier
def getClassifier(self):
return classifier
def setPreprocessor(self, preprocessor):
self.preprocessor = preprocessor
def getPreprocessor(self):
return preprocessor
def setEvaluateModule(self, evaluate):
self.evaluate = evaluate
def getEvaluateModule(self):
return evaluate
def setFilePath(self, file_path):
self.file_path = file_path
def setResultPath(self, result_path):
self.result_path = result_path
def setK(self, k):
self.k = k
def setClass_name(self, class_name):
self.class_name = class_name
class CrossValidation(object):
dts = None
#metodo utilizado para classifacao
classifier = None
#conjunto de dados de teste
testData = None
#conjunto de dados de treinamento
trainingData = None
evaluate = None
#numero de folds
numberOfFolds = 10
file_path = ""
#caminho da pasta onde serao salvos os resultados
result_path = ""
preprocessor = None
def __init__(self):
# print("Cross Validation constructor")
self.evaluate = EvaluateModule()
def run(self):
self.classifier.setResultPath(self.result_path)
self.foldExecution()
def foldExecution(self):
i = self.iteration
for self.iteration in range(i, (self.numberOfFolds + 1)):
tempo_inicio = time.time()
self.loadTrainingData()
self.loadTestData()
#executa funcoes para transformacao de dados categoricos
if self.preprocessor:
self.preprocessor.setDataSet(self.trainingData)
self.preprocessor.setTestDataSet(self.testData)
self.trainingData, self.testData = self.preprocessor.transformCategory(
)
#seta dados de treinamento e teste no classificador
self.classifier.setDataSet(self.trainingData)
self.classifier.setTestDataSet(self.testData)
#seta iteracao do cross no classficador
self.classifier.setIteration(self.iteration)
#executa o processo de treino e teste do classificador
self.classifier.run()
del (self.trainingData)
# self.loadTestData()
#seta conjunto de dados original de teste e iteracao atual do cross-validation na classe de avaliacao
self.evaluate.setTestDataSet(self.testData)
self.evaluate.setIteration(self.iteration)
#verifica quel o metodo de classificacao utilziado
if (isinstance(self.classifier, RnaClassifier)):
print("rna")
self.evaluate.setResultPath(self.result_path)
elif (isinstance(self.classifier, KnnClassifier)):
print("knn")
# self.evaluate.setResultPath(self.result_path)
elif (isinstance(self.classifier, ClusteredKnnClassifier)):
print("clustered knn")
#self.evaluate.setResultPath(self.result_path)
elif (isinstance(self.classifier, ClusteredDensityKnnClassifier)):
print("clustered density knn")
#self.evaluate.setResultPath(self.result_path)
elif (isinstance(self.classifier, HybridClassifier)):
print("hybrid")
annModel = self.classifier.getRna().getModel()
self.saveModelToFile(annModel, 'ann')
self.evaluate.setResultPath(self.result_path +
"final_method_classification/")
tempo_execucao = time.time() - tempo_inicio
self.evaluate.setTempoExecucao(tempo_execucao)
self.evaluate.setTrainingTime(self.classifier.getTrainingTime())
self.evaluate.setTestTime(self.classifier.getTestTime())
#executa metodo de avaliacao
self.evaluate.run()
#carrega conjunto de treinamento de acordo coma iteracao atual do cross valiadation
def loadTrainingData(self):
#exclude current cross validation iteration corresponding fold
trainFolds = glob.glob(self.file_path + 'fold_[!' +
str(self.iteration) + ']*.csv')
self.trainingData = pd.concat(
(pd.read_csv(fold) for fold in trainFolds))
#carrega conjunto de teste de acordo com a iteracao atual do cross validation
def loadTestData(self):
self.testData = DataSet.loadSubDataSet(self.file_path + "fold_" +
str(self.iteration) + ".csv")
def setIteration(self, iteration):
self.iteration = iteration
def setClassifier(self, classifier):
self.classifier = classifier
def getClassifier(self):
return classifier
def setPreprocessor(self, preprocessor):
self.preprocessor = preprocessor
def getPreprocessor(self):
return preprocessor
def setEvaluateModule(self, evaluate):
self.evaluate = evaluate
def getEvaluateModule(self):
return evaluate
def setFilePath(self, file_path):
self.file_path = file_path
def setResultPath(self, result_path):
directory = os.path.dirname(result_path)
if not os.path.exists(directory):
os.makedirs(directory)
self.result_path = result_path
def setK(self, k):
self.numberOfFolds = k
def saveModelToFile(self, model, prefix):
directory = os.path.dirname(self.file_path + 'models/')
if not os.path.exists(directory):
os.makedirs(directory)
fileName = directory + prefix + '_' + str(self.iteration - 1)
pickle.dump(model, open(fileName, 'wb'))
print('[' + str(datetime.datetime.now()).split('.')[0] + '] ' +
fileName + ' saved [')