def trainingMethod(self, kindDataSet):
self.model = GaussianNB()
self.GaussianNBAlgorithm = self.model.fit(self.dataset, self.target)
if kindDataSet == 1:
params = 'Params:default'
self.performanceData = responseTraining.responseTraining(
self.GaussianNBAlgorithm, 'GaussianNB', params,
self.validation)
self.performanceData.estimatedMetricsPerformance(
self.dataset, self.target)
else:
params = 'Params:default'
self.performanceData = responseTraining.responseTraining(
self.GaussianNBAlgorithm, 'GaussianNB', params,
self.validation)
self.performanceData.estimatedMetricsPerformanceMultilabels(
self.dataset, self.target)
def trainingMethod(self, kindDataSet):
self.model = AdaBoostClassifier(n_estimators=self.n_estimators,
algorithm=self.algorithm)
self.AdaBoostAlgorithm = self.model.fit(self.dataset, self.target)
if kindDataSet == 1: #binary
params = "algorithm:%s-n_estimators:%d" % (self.algorithm,
self.n_estimators)
self.performanceData = responseTraining.responseTraining(
self.AdaBoostAlgorithm, 'AdaBoost', params, self.validation)
self.performanceData.estimatedMetricsPerformance(
self.dataset, self.target)
else:
params = "algorithm:%s-n_estimators:%d" % (self.algorithm,
self.n_estimators)
self.performanceData = responseTraining.responseTraining(
self.AdaBoostAlgorithm, 'AdaBoost', params, self.validation)
self.performanceData.estimatedMetricsPerformanceMultilabels(
self.dataset, self.target)
def trainingMethod(self, kindDataSet):
self.model = BaggingClassifier(n_estimators=self.n_estimators,
bootstrap=self.bootstrap,
n_jobs=-1)
self.BagginAlgorithm = self.model.fit(self.dataset, self.target)
if kindDataSet == 1: #binary
params = "n_estimators:%d-bootstrap:%s" % (self.n_estimators,
self.bootstrap)
self.performanceData = responseTraining.responseTraining(
self.BagginAlgorithm, 'Baggin', params, self.validation)
self.performanceData.estimatedMetricsPerformance(
self.dataset, self.target)
else:
params = "n_estimators:%d-bootstrap:%s" % (self.n_estimators,
self.bootstrap)
self.performanceData = responseTraining.responseTraining(
self.BagginAlgorithm, 'Baggin', params, self.validation)
self.performanceData.estimatedMetricsPerformanceMultilabels(
self.dataset, self.target)
def trainingMethod(self, kindDataSet):
self.model=MLPClassifier(hidden_layer_sizes=self.hidden_layer_sizes,activation=self.activation,solver=self.solver,learning_rate=self.learning_rate)
self.MLPAlgorithm=self.model.fit(self.dataset,self.target)
params = "activation:%s-learning_rate:%s-solver:%s-hidden_layer_sizes_a:%d-hidden_layer_sizes_b:%d-hidden_layer_sizes_c:%d-alpha:%f-max_iter:%d-shuffle:%s" % (self.activation, self.learning_rate, self.solver,self.hidden_layer_sizes[0], self.hidden_layer_sizes[1], self.hidden_layer_sizes[2], self.alpha, self.max_iter, self.shuffle)
self.performanceData = responseTraining.responseTraining(self.MLPAlgorithm, 'MLP', params, self.validation)
if kindDataSet == 1:
self.performanceData.estimatedMetricsPerformance(self.dataset, self.target)
else:
self.performanceData.estimatedMetricsPerformanceMultilabels(self.dataset, self.target)
def trainingMethod(self, kindDataSet):
self.model = tree.DecisionTreeClassifier(criterion=self.criterion,
splitter=self.splitter)
self.DecisionTreeAlgorithm = self.model.fit(self.dataset, self.target)
if kindDataSet == 1:
params = "criterion:%s-splitter:%s" % (self.criterion,
self.splitter)
self.performanceData = responseTraining.responseTraining(
self.DecisionTreeAlgorithm, 'DecisionTree', params,
self.validation)
self.performanceData.estimatedMetricsPerformance(
self.dataset, self.target)
else:
params = "criterion:%s-splitter:%s" % (self.criterion,
self.splitter)
self.performanceData = responseTraining.responseTraining(
self.DecisionTreeAlgorithm, 'DecisionTree', params,
self.validation)
self.performanceData.estimatedMetricsPerformanceMultilabels(
self.dataset, self.target)
def trainingMethod(self, kindDataSet):
self.model = GradientBoostingClassifier(n_estimators=self.n_estimators)
self.GradientAlgorithm = self.model.fit(self.dataset, self.target)
params = "n_estimators:%d-loss:%s-min_samples_leaf:%d-min_samples_split:%d" % (
self.n_estimators, self.loss, self.min_samples_leaf,
self.min_samples_split)
self.performanceData = responseTraining.responseTraining(
self.GradientAlgorithm, 'Gradient', params, self.validation)
if kindDataSet == 1:
self.performanceData.estimatedMetricsPerformance(
self.dataset, self.target)
else:
self.performanceData.estimatedMetricsPerformanceMultilabels(
self.dataset, self.target)
def trainingMethod(self, kindDataSet):
self.model = NuSVC(kernel=self.kernel,
degree=self.degree,
gamma=self.gamma,
nu=self.nu,
probability=True)
self.NuSVMAlgorithm = self.model.fit(self.dataset, self.target)
params = "kernel:%s-degree:%f-gamma:%f-nu:%f-probability:True" % (
self.kernel, self.degree, self.gamma, self.nu)
self.performanceData = responseTraining.responseTraining(
self.NuSVMAlgorithm, 'NuSVM', params, self.validation)
if kindDataSet == 1:
self.performanceData.estimatedMetricsPerformance(
self.dataset, self.target)
else:
self.performanceData.estimatedMetricsPerformanceMultilabels(
self.dataset, self.target)
def trainingMethod(self, kindDataSet):
self.model = KNeighborsClassifier(n_neighbors=self.n_neighbors,
weights=self.weights,
algorithm=self.algorithm,
metric=self.metric,
n_jobs=-1) #instancia
self.knnAlgorithm = self.model.fit(self.dataset, self.response)
#training...
params = "algorithm:%s-metric:%s-neighbors:%d-weights:%s" % (
self.algorithm, self.metric, self.n_neighbors, self.weights)
self.performanceData = responseTraining.responseTraining(
self.knnAlgorithm, 'KNN', params, self.validation)
if kindDataSet == 1:
self.performanceData.estimatedMetricsPerformance(
self.dataset, self.response)
else:
self.performanceData.estimatedMetricsPerformanceMultilabels(
self.dataset, self.response)
def trainingMethod(self, kindDataSet):
self.model = RandomForestClassifier(
n_estimators=self.n_estimators,
criterion=self.criterion,
min_samples_leaf=self.min_samples_leaf,
min_samples_split=self.min_samples_split,
bootstrap=self.bootstrap,
n_jobs=-1)
self.RandomForestAlgorithm = self.model.fit(self.dataset, self.target)
params = "criterion:%s-n_estimators:%d-min_samples_leaf:%d-min_samples_split:%d-bootstrap:%s" % (
self.criterion, self.n_estimators, self.min_samples_leaf,
self.min_samples_split, self.bootstrap)
self.performanceData = responseTraining.responseTraining(
self.RandomForestAlgorithm, 'RandomForest', params,
self.validation)
if kindDataSet == 1:
self.performanceData.estimatedMetricsPerformance(
self.dataset, self.target)
else:
self.performanceData.estimatedMetricsPerformanceMultilabels(
self.dataset, self.target)
