def fromJson(obj, experiment):
binary_model_conf = ClassifierConfFactory.getFactory().fromJson(
obj['models_conf']['binary'], experiment)
conf = UncertaintySamplingConfiguration(obj['auto'], obj['budget'],
obj['batch'],
binary_model_conf)
return conf
def createMulticlassExperiment(self):
exp = self.supervised_exp
name = exp.experiment_name + '_alertsMulticlassClassifier'
multiclass_exp = ClassificationExperiment(
exp.project,
exp.dataset,
exp.db,
exp.cursor,
experiment_name=name,
experiment_label=exp.experiment_label,
parent=exp.experiment_id)
multiclass_exp.setFeaturesFilenames(exp.features_filenames)
params = {}
params['num_folds'] = exp.classification_conf.num_folds
params['sample_weight'] = False
params['families_supervision'] = True
params['optim_algo'] = 'sag'
params['alerts_conf'] = None
test_conf = TestConfiguration()
test_conf.setUnlabeled(labels_annotations='annotations')
params['test_conf'] = test_conf
conf = ClassifierConfFactory.getFactory().fromParam(
'LogisticRegression', params)
multiclass_exp.setClassifierConf(conf)
multiclass_exp.createExperiment()
multiclass_exp.export()
return multiclass_exp
def fromJson(obj, experiment):
binary_model_conf = ClassifierConfFactory.getFactory().fromJson(
obj['models_conf']['binary'], experiment)
conf = CesaBianchiConfiguration(obj['auto'], obj['budget'],
obj['batch'], obj['b'],
binary_model_conf)
return conf
def fromJson(obj, session):
experiment = ClassificationExperiment(obj['project'], obj['dataset'], session)
Experiment.expParamFromJson(experiment, obj)
classification_conf = ClassifierConfFactory.getFactory().fromJson(
obj['classification_conf'], experiment)
experiment.setClassifierConf(classification_conf)
return experiment
def fromJson(obj, experiment):
rare_category_detection_conf = RareCategoryDetectionStrategy.fromJson(obj['rare_category_detection_conf'])
binary_model_conf = ClassifierConfFactory.getFactory().fromJson(obj['models_conf']['binary'], experiment)
conf = IlabConfiguration(obj['auto'], obj['budget'],
rare_category_detection_conf,
obj['num_uncertain'], obj['eps'],
binary_model_conf)
return conf
def fromJson(obj, experiment):
multiclass_model_conf = ClassifierConfFactory.getFactory().fromJson(
obj['models_conf']['multiclass'], experiment)
rare_category_detection_conf = RareCategoryDetectionStrategy.fromJson(
obj['rare_category_detection_conf'])
conf = RareCategoryDetectionConfiguration(
obj['auto'], obj['budget'], rare_category_detection_conf,
multiclass_model_conf)
return conf
def fromJson(obj, experiment):
validation_conf = None
if obj['validation_conf'] is not None:
validation_conf = TestConfiguration.fromJson(obj['validation_conf'],
experiment)
binary_model_conf = ClassifierConfFactory.getFactory().fromJson(
obj['models_conf']['binary'], experiment)
conf = RandomSamplingConfiguration(obj['auto'], obj['budget'], obj['batch'],
binary_model_conf, validation_conf)
return conf
def gornitzBinaryModelConf():
classifier_args = {}
classifier_args['num_folds'] = 4
classifier_args['sample_weight'] = False
classifier_args['families_supervision'] = False
classifier_args['alerts_conf'] = None
binary_model_conf = ClassifierConfFactory.getFactory().fromParam(
'Sssvdd', classifier_args)
binary_model_conf.setUnlabeled(labels_annotations='annotations')
return binary_model_conf
def fromJson(obj, experiment):
validation_conf = None
if obj['validation_conf'] is not None:
validation_conf = TestConfiguration.fromJson(
obj['validation_conf'], experiment)
binary_model_conf = ClassifierConfFactory.getFactory().fromJson(
obj['models_conf']['binary'], experiment)
conf = AladinConfiguration(obj['auto'], obj['budget'],
obj['num_annotations'], binary_model_conf,
validation_conf)
return conf
def generateParser():
parser = argparse.ArgumentParser(
description = 'Learn a detection model. ' +
'The labels must be stored in labels/true_labels.csv.')
Experiment.projectDatasetFeturesParser(parser)
models = ['LogisticRegression', 'Svc', 'GaussianNaiveBayes',
'DecisionTree', 'RandomForest', 'GradientBoosting']
subparsers = parser.add_subparsers(dest = 'model')
factory = ClassifierConfFactory.getFactory()
for model in models:
model_parser = subparsers.add_parser(model)
factory.generateParser(model, model_parser)
return parser
def setExperimentFromArgs(self, args):
self.setFeaturesFilenames(args.features_files)
factory = ClassifierConfFactory.getFactory()
conf = factory.fromArgs(args.model, args, self)
self.setClassifierConf(conf)
try:
self.initLabels(args.labels)
except Exception as e:
message = 'The ground truth labels must be provided in true_labels.csv '
message += 'to run SecuML_classification.'
print message
raise e
self.export()
def generateParamsFromArgs(args, experiment):
supervised_args = {}
supervised_args['num_folds'] = 4
supervised_args['sample_weight'] = False
supervised_args['families_supervision'] = False
test_conf = TestConfiguration()
test_conf.setUnlabeled(labels_annotations='annotations')
supervised_args['test_conf'] = test_conf
binary_model_conf = ClassifierConfFactory.getFactory().fromParam(
'LogisticRegression', supervised_args)
params = ActiveLearningConfiguration.generateParamsFromArgs(
args, experiment, binary_model_conf=binary_model_conf)
params['num_annotations'] = args.num_annotations
return params
def aladinMulticlassModelConf():
classifier_args = {}
classifier_args['num_folds'] = 4
classifier_args['sample_weight'] = False
classifier_args['families_supervision'] = True
classifier_args['alerts_conf'] = None
classifier_args['optim_algo'] = 'liblinear'
test_conf = TestConfiguration()
test_conf.setUnlabeled(labels_annotations='annotations')
classifier_args['test_conf'] = test_conf
factory = ClassifierConfFactory.getFactory()
multiclass_model_conf = factory.fromParam('LogisticRegression',
classifier_args)
return multiclass_model_conf
def generateParamsFromArgs(args):
params = ActiveLearningConfiguration.generateParamsFromArgs(args)
multiclass_classifier_args = {}
multiclass_classifier_args['num_folds'] = args.num_folds
multiclass_classifier_args['sample_weight'] = False
multiclass_classifier_args['families_supervision'] = True
test_conf = TestConfiguration()
test_conf.setUnlabeled(labels_annotations = 'annotations')
multiclass_classifier_args['test_conf'] = test_conf
multiclass_conf = ClassifierConfFactory.getFactory().fromParam(
'LogisticRegression', multiclass_classifier_args)
rare_category_detection_conf = RareCategoryDetectionStrategy(multiclass_conf,
args.cluster_strategy, args.num_annotations, 'uniform')
params['rare_category_detection_conf'] = rare_category_detection_conf
params['num_uncertain'] = args.num_uncertain
params['eps'] = 0.49
return params
def generateParamsFromArgs(args):
supervised_args = {}
supervised_args['num_folds'] = args.num_folds
supervised_args['sample_weight'] = args.sample_weight
supervised_args['families_supervision'] = False
test_conf = TestConfiguration()
test_conf.setUnlabeled(labels_annotations='annotations')
supervised_args['test_conf'] = test_conf
binary_model_conf = ClassifierConfFactory.getFactory().fromParam(
args.model_class, supervised_args)
active_learning_params = {}
active_learning_params['auto'] = args.auto
active_learning_params['budget'] = args.budget
active_learning_params['binary_model_conf'] = binary_model_conf
return active_learning_params
def generateParamsFromArgs(args, experiment):
params = ActiveLearningConfiguration.generateParamsFromArgs(
args, experiment)
multiclass_classifier_args = {}
multiclass_classifier_args['num_folds'] = args.num_folds
multiclass_classifier_args['sample_weight'] = False
multiclass_classifier_args['families_supervision'] = True
multiclass_classifier_args['alerts_conf'] = None
test_conf = TestConfiguration()
test_conf.setUnlabeled(labels_annotations='annotations')
multiclass_classifier_args['test_conf'] = test_conf
multiclass_conf = ClassifierConfFactory.getFactory().fromParam(
args.model_class, multiclass_classifier_args)
rare_category_detection_conf = RareCategoryDetectionStrategy(
multiclass_conf, args.cluster_strategy, args.num_annotations,
'uniform')
params['rare_category_detection_conf'] = rare_category_detection_conf
params['num_annotations'] = args.num_annotations
params['multiclass_model_conf'] = multiclass_conf
return params
def generateParamsFromArgs(args, experiment, binary_model_conf=None):
if binary_model_conf is None:
supervised_args = {}
supervised_args['num_folds'] = args.num_folds
supervised_args['sample_weight'] = args.sample_weight
supervised_args['families_supervision'] = False
test_conf = TestConfiguration()
test_conf.setUnlabeled(labels_annotations='annotations')
supervised_args['test_conf'] = test_conf
binary_model_conf = ClassifierConfFactory.getFactory().fromParam(
args.model_class, supervised_args)
active_learning_params = {}
active_learning_params['auto'] = args.auto
active_learning_params['budget'] = args.budget
active_learning_params['binary_model_conf'] = binary_model_conf
validation_conf = None
if args.validation_dataset is not None:
validation_conf = TestConfiguration()
validation_conf.setTestDataset(args.validation_dataset, experiment)
active_learning_params['validation_conf'] = validation_conf
return active_learning_params
def fromJson(obj):
classification_conf = ClassifierConfFactory.getFactory().fromJson(obj['classification_conf'], None)
conf = RareCategoryDetectionStrategy(classification_conf, obj['cluster_strategy'],
obj['num_annotations'], obj['cluster_weights'])
return conf
