def getTopWeightedFeatures(experiment_id, inst_exp_id, instance_id, size,
fold_id):
if fold_id == 'all':
return None
instance_id = int(instance_id)
exp = ExperimentFactory.getFactory().fromJson(experiment_id, session)
validation_experiment = ExperimentFactory.getFactory().fromJson(
inst_exp_id, session)
# get the features
features_names, features_values = validation_experiment.getFeatures(
instance_id)
features_values = [float(value) for value in features_values]
# get the pipeline with scaler and logistic model
experiment_dir = exp.getOutputDirectory()
if fold_id != 'None':
experiment_dir = path.join(experiment_dir, fold_id)
pipeline = joblib.load(path.join(experiment_dir, 'model', 'model.out'))
# scale the features
scaled_values = pipeline.named_steps['scaler'].transform(
np.reshape(features_values, (1, -1)))
weighted_values = np.multiply(scaled_values,
pipeline.named_steps['model'].coef_)
features = list(
map(lambda name, value, w_value: (name, value, w_value),
features_names, features_values, weighted_values[0]))
features.sort(key=lambda tup: abs(tup[2]))
features = features[:-int(size) - 1:-1]
features_names = [x[0] for x in features]
features_values = [x[1] for x in features]
features_weighted_values = [x[2] for x in features]
max_length = max([len(f) for f in features_names])
if max_length > 30:
labels = [str(i) for i in range(len(features_names))]
tooltips = [
features_names[i] + ' (' + str(features_values[i]) + ')'
for i in range(len(features_names))
]
else:
labels = features_names
tooltips = features_values
barplot = BarPlot(labels)
dataset = PlotDataset(features_weighted_values, None)
dataset.setColor(colors_tools.red)
barplot.addDataset(dataset)
return jsonify(barplot.toJson(tooltip_data=tooltips))
def getTopModelFeatures(experiment_id, size, train_test, fold_id):
size = int(size)
exp = ExperimentFactory.getFactory().fromJson(experiment_id, session)
directory = exp.getOutputDirectory()
if fold_id != 'None' and fold_id != 'all':
directory = path.join(directory, fold_id)
directory = path.join(directory, train_test)
filename = path.join(directory, 'model_coefficients.csv')
with open(filename, 'r') as f:
coefficients_df = pd.read_csv(f, header=0, index_col=0)
model_coefficients = list(coefficients_df['mean'])
features_names = list(map(str, coefficients_df.index))
coefficients = list(
map(lambda name, coef: (name, coef), features_names,
model_coefficients))
coefficients.sort(key=lambda tup: abs(tup[1]))
coefficients = coefficients[:-size - 1:-1]
coefficients_names = [coef[0] for coef in coefficients]
coefficients_values = [coef[1] for coef in coefficients]
max_length = max([len(coef) for coef in coefficients_names])
if max_length > 30:
coefficients_ids = [str(i) for i in range(len(coefficients_names))]
coefficients_names = [
name.replace(' WHERE', '\nWHERE')
for name in coefficients_names
]
barplot = BarPlot(coefficients_ids)
dataset = PlotDataset(coefficients_values, None)
if exp.conf.featureImportance() == 'weight':
dataset.setColor(colors_tools.red)
barplot.addDataset(dataset)
return jsonify(barplot.toJson(tooltip_data=coefficients_names))
else:
barplot = BarPlot(coefficients_names)
dataset = PlotDataset(coefficients_values, None)
if exp.conf.featureImportance() == 'weight':
dataset.setColor(colors_tools.red)
barplot.addDataset(dataset)
return jsonify(barplot.toJson())
def updateCurrentExperiment(experiment_id):
experiment = ExperimentFactory.getFactory().fromJson(
experiment_id, secuml_conf, session)
return experiment
formatter_class=argparse.RawTextHelpFormatter)
Experiment.projectDatasetFeturesParser(parser)
subparsers = parser.add_subparsers(dest='strategy')
subparsers.required = True
factory = ActiveLearningConfFactory.getFactory()
strategies = factory.getStrategies()
for strategy in strategies:
strategy_parser = subparsers.add_parser(strategy)
factory.generateParser(strategy, strategy_parser)
return parser
def webTemplate(self):
return 'ActiveLearning/active_learning.html'
def getCurrentIteration(self):
return experiment_db_tools.getCurrentIteration(self.session,
self.experiment_id)
def setExperimentFromArgs(self, args):
self.initExperiment(args.project, args.dataset,
experiment_name=args.exp_name)
factory = ActiveLearningConfFactory.getFactory()
conf = factory.fromArgs(args.strategy, args, logger=self.logger)
self.setConf(conf, args.features_file,
annotations_filename=args.init_annotations_file)
self.export()
ExperimentFactory.getFactory().registerClass('ActiveLearningExperiment',
ActiveLearningExperiment)
clustering_conf = ClusteringConfFactory.getFactory().fromParam(
args.clustering_algo, params, logger=logger)
alerts_conf = AlertsConfiguration(args.top_n_alerts,
args.detection_threshold,
clustering_conf,
logger=logger)
test_conf = ValidationDatasetConf(args.validation_dataset,
alerts_conf=alerts_conf,
logger=logger)
conf.test_conf = test_conf
return conf
def setExperimentFromArgs(self, args):
self.initExperiment(args.project,
args.dataset,
experiment_name=args.exp_name)
factory = ClassifierConfFactory.getFactory()
if args.model == 'AlreadyTrained':
self.already_trained = args.model_exp_id
conf = self.generateAlreadyTrainedConf(factory, args, self.logger)
else:
conf = factory.fromArgs(args.model, args, logger=self.logger)
self.setConf(conf,
args.features_file,
annotations_filename='ground_truth.csv')
self.export()
ExperimentFactory.getFactory().registerClass('ClassificationExperiment',
ClassificationExperiment)
class ValidationExperiment(Experiment):
def getKind(self):
return 'Validation'
def generateSuffix(self):
suffix = ''
return suffix
def _checkConf(self):
if not db_tables.hasGroundTruth(self):
self.conf.logger.warning(
'The validation dataset does not have ground-truth.')
@staticmethod
def fromJson(obj, secuml_conf):
experiment = ValidationExperiment(secuml_conf)
experiment.initExperiment(obj['project'], obj['dataset'], create=False)
Experiment.expParamFromJson(experiment, obj, Configuration())
return experiment
def toJson(self):
conf = Experiment.toJson(self)
conf['__type__'] = 'ValidationExperiment'
return conf
ExperimentFactory.getFactory().registerClass('ValidationExperiment',
ValidationExperiment)
obj['dataset'],
session,
create=False)
Experiment.expParamFromJson(experiment, obj, None)
return experiment
def toJson(self):
conf = Experiment.toJson(self)
conf['__type__'] = 'DescriptiveStatisticsExperiment'
return conf
@staticmethod
def generateParser():
parser = argparse.ArgumentParser(
description='Descriptive Statistics of the Dataset')
Experiment.projectDatasetFeturesParser(parser)
return parser
def webTemplate(self):
return 'DescriptiveStatistics/descriptive_statistics.html'
def setExperimentFromArgs(self, args):
self.setConf(None,
args.features_file,
annotations_filename='ground_truth.csv')
self.export()
ExperimentFactory.getFactory().registerClass('DescriptiveStatisticsExperiment',
DescriptiveStatisticsExperiment)
def createDimensionReductionExperiment(self, num_features):
name = '-'.join([self.experiment_name, 'projection'])
projection_conf = self.conf.projection_conf
if projection_conf.num_components is not None:
if projection_conf.num_components > num_features:
projection_conf.num_components = num_features
projection_exp = DimensionReductionExperiment(
self.project,
self.dataset,
self.session,
experiment_name=name,
annotations_id=self.annotations_id,
parent=self.experiment_id,
logger=self.logger)
projection_exp.setConf(projection_conf)
projection_exp.setFeaturesFilenames(self.features_filename)
projection_exp.createExperiment()
projection_exp.export()
return projection_exp
def projectInstances(self, instances):
projection_exp = self.createDimensionReductionExperiment(
instances.numFeatures())
projected_instances = projection_exp.run(instances=instances,
export=False)
return projected_instances
ExperimentFactory.getFactory().registerClass('ClusteringExperiment',
ClusteringExperiment)
def toJson(self):
conf = DimensionReductionExperiment.toJson(self)
conf['__type__'] = 'FeatureSelectionExperiment'
conf['conf'] = self.conf.toJson()
return conf
@staticmethod
def generateParser():
parser = argparse.ArgumentParser(description='Feature Selection.')
algos = [
'ChiSquare', 'Fclassif', 'MutualInfoClassif', 'VarianceFilter',
'RecursiveFeatureElimination', 'NoneFilter'
]
DimensionReductionExperiment.projectDatasetFeturesParser(parser)
DimensionReductionExperiment.generateDimensionReductionParser(parser)
subparsers = parser.add_subparsers(dest='algo')
subparsers.required = True
factory = DimensionReductionConfFactory.getFactory()
for algo in algos:
algo_parser = subparsers.add_parser(algo)
factory.generateParser(algo, algo_parser)
return parser
def webTemplate(self):
return 'DimensionReduction/feature_selection.html'
ExperimentFactory.getFactory().registerClass('FeatureSelectionExperiment',
FeatureSelectionExperiment)
@staticmethod
def generateParser():
parser = argparse.ArgumentParser(
description='Rare Category Detection',
formatter_class=argparse.RawTextHelpFormatter)
ActiveLearningExperiment.projectDatasetFeturesParser(parser)
factory = ActiveLearningConfFactory.getFactory()
factory.generateParser('RareCategoryDetection', parser)
return parser
def webTemplate(self):
return 'ActiveLearning/active_learning.html'
def getCurrentIteration(self):
return experiment_db_tools.getCurrentIteration(self.session,
self.experiment_id)
def setExperimentFromArgs(self, args):
factory = ActiveLearningConfFactory.getFactory()
conf = factory.fromArgs('RareCategoryDetection',
args,
logger=self.logger)
self.setConf(conf,
args.features_file,
annotations_filename=args.init_annotations_file)
self.export()
ExperimentFactory.getFactory().registerClass('RareCategoryDetectionExperiment',
RareCategoryDetectionExperiment)
DimensionReductionExperiment.expParamFromJson(experiment, obj, conf)
return experiment
def toJson(self):
conf = DimensionReductionExperiment.toJson(self)
conf['__type__'] = 'ProjectionExperiment'
conf['conf'] = self.conf.toJson()
return conf
@staticmethod
def generateParser():
parser = argparse.ArgumentParser(
description='Projection of the data for data visualization.')
DimensionReductionExperiment.projectDatasetFeturesParser(parser)
DimensionReductionExperiment.generateDimensionReductionParser(parser)
algos = ['Pca', 'Rca', 'Lda', 'Lmnn', 'Nca', 'Itml']
subparsers = parser.add_subparsers(dest='algo')
subparsers.required = True
factory = DimensionReductionConfFactory.getFactory()
for algo in algos:
algo_parser = subparsers.add_parser(algo)
factory.generateParser(algo, algo_parser)
return parser
def webTemplate(self):
return 'DimensionReduction/projection.html'
ExperimentFactory.getFactory().registerClass('ProjectionExperiment',
ProjectionExperiment)