def run(self, train_instances, cv_monitoring=False):
Experiment.run(self)
self._train(train_instances)
if cv_monitoring:
self._cv_monitoring(train_instances)
else:
self.cv_monitoring = None
self._export()
def add_to_db(self):
from secuml.exp.diadem import add_diadem_exp_to_db
Experiment.add_to_db(self)
add_diadem_exp_to_db(self.session,
self.exp_conf.exp_id,
self.exp_conf.fold_id,
'train',
classifier_conf=self.exp_conf.core_conf)
def run(self, instances=None, drop_annotated_instances=False, quick=False):
Experiment.run(self)
instances = self.get_instances()
core_conf = self.exp_conf.core_conf
clustering = core_conf.algo(instances, core_conf)
clustering.fit()
clustering.generate(drop_annotated_instances=drop_annotated_instances)
clustering.export(self.output_dir(), quick=quick)
def __init__(self, exp_conf, create=True, session=None):
self.kind = exp_conf.kind
Experiment.__init__(self, exp_conf, create=create, session=session)
self.test_instances = None
self.predictions = None
self.prediction_time = None
self.classifier = None
alerts_conf = self.exp_conf.core_conf
self.monitoring = DetectionMonitoring(self, alerts_conf=alerts_conf)
def add_to_db(self):
Experiment.add_to_db(self)
from secuml.exp.diadem import add_diadem_exp_to_db
add_diadem_exp_to_db(self.session,
self.exp_conf.exp_id,
self.exp_conf.fold_id,
self.kind,
alerts_conf=self.alerts_conf,
classifier_conf=self.exp_conf.core_conf)
def set_clusters(self, instances, assigned_clusters, centroids,
drop_annotated_instances, cluster_labels):
Experiment.run(self)
clustering = Clusters(instances, assigned_clusters)
clustering.generate(centroids,
drop_annotated_instances=drop_annotated_instances,
cluster_labels=cluster_labels)
clustering.export(self.output_dir(),
drop_annotated_instances=drop_annotated_instances)
def add_to_db(self):
Experiment.add_to_db(self)
features_set_id = self.exp_conf.features_conf.set_id
annotations_file = self.exp_conf.annotations_conf.annotations_filename
stats_exp = FeaturesAnalysisExpAlchemy(
id=self.exp_id,
features_set_id=features_set_id,
annotations_filename=annotations_file)
self.session.add(stats_exp)
self.session.flush()
def add_to_db(self):
Experiment.add_to_db(self)
has_ground_truth = self.exp_conf.dataset_conf.has_ground_truth
dataset_id = self.exp_conf.dataset_conf.dataset_id
self.session.add(
DiademExpAlchemy(exp_id=self.exp_conf.exp_id,
dataset_id=dataset_id,
fold_id=self.exp_conf.fold_id,
type=self.kind,
perf_monitoring=has_ground_truth))
self.session.flush()
def __init__(self,
exp_conf,
classifier=None,
alerts_conf=None,
create=True,
session=None):
self.classifier = classifier
self.kind = exp_conf.kind
self.alerts_conf = alerts_conf
self.test_instances = None
self.predictions = None
self.monitoring = None
Experiment.__init__(self, exp_conf, create=create, session=session)
def run(self):
Experiment.run(self)
instances = self.get_instances()
with_density = instances.num_instances() < 150000
if not with_density:
self.exp_conf.logger.warning('There are more than 150.000, so '
'the density plots are not '
'displayed. ')
stats = FeaturesAnalysis(instances,
self.exp_conf.multiclass,
self.exp_conf.logger,
with_density=with_density)
stats.gen_plots(self.output_dir())
stats.gen_scoring(self.output_dir())
def run(self, instances=None, export=True):
Experiment.run(self)
instances = self.get_instances()
core_conf = self.exp_conf.core_conf
dimension_reduction = core_conf.algo(core_conf)
# Fit
dimension_reduction.fit(instances)
if export:
dimension_reduction.export_fit(self.output_dir(), instances)
# Transformation
projected_instances = dimension_reduction.transform(instances)
if export:
dimension_reduction.export_transform(self.output_dir(), instances,
projected_instances)
return projected_instances
def run(self):
Experiment.run(self)
datasets = Datasets(self.get_instances())
active_learning = ActiveLearning(self, datasets)
if not self.exp_conf.core_conf.auto:
from secuml.exp.celery_app.app import secumlworker
from secuml.exp.active_learning.celery_tasks import IterationTask
options = {}
# bind iterations object to IterationTask class
active_learning.run_next_iter(output_dir=self.output_dir())
IterationTask.iteration_object = active_learning
# Start worker
secumlworker.enable_config_fromcmdline = False
secumlworker.run(**options)
else:
active_learning.run_iterations(output_dir=self.output_dir())
def create_exp(self):
Experiment.create_exp(self)
# create projection experiment
self.projection_exp = None
if self.exp_conf.core_conf is None:
return
projection_core_conf = self.exp_conf.core_conf.projection_conf
if projection_core_conf is not None:
features_conf = FeaturesConf(
self.exp_conf.features_conf.input_features,
self.exp_conf.secuml_conf.logger)
projection_conf = ProjectionConf(self.exp_conf.secuml_conf,
self.exp_conf.dataset_conf,
features_conf,
self.exp_conf.annotations_conf,
projection_core_conf,
name='-'.join(
[self.exp_conf.name, 'proj']),
parent=self.exp_id)
self.projection_exp = ProjectionExperiment(projection_conf,
session=self.session)
def get_instances(self, instances=None):
instances = Experiment.get_instances(self, instances=instances)
if self.exp_conf.label != 'all':
selected_ids = instances.ground_truth.get_annotated_ids(
label=self.exp_conf.label)
instances = instances.get_from_ids(selected_ids)
if self.projection_exp is not None:
try:
instances = self.projection_exp.run(instances=instances,
export=False)
except FewerThanTwoLabels:
self.conf.logger.warning('There are too few class labels.'
'The instances are not projected '
'before building the clustering.')
return instances
def __init__(self, exp_conf, create=True, session=None):
Experiment.__init__(self, exp_conf, create=create, session=session)
self.classifier = None
self.train_time = None
self.monitoring = TrainMonitoring(self)
def add_to_db(self):
Experiment.add_to_db(self)
al_exp = ActiveLearningExpAlchemy(id=self.exp_id, current_iter=0,
finished=False)
self.session.add(al_exp)
self.session.flush()
def run(self):
Experiment.run(self)
stats = FeaturesAnalysis(self.get_instances())
stats.compute()
stats.export(self.output_dir())
def run(self, classifier, test_instances):
Experiment.run(self)
self.test_instances = self.get_instances(test_instances)
self.classifier = classifier
self._test()
self._export()
def __init__(self, exp_conf, create=True, session=None):
Experiment.__init__(self, exp_conf, create=create, session=session)
self.classifier = None
self.predictions = None
self.monitoring = None
self.cv_monitoring = None
def __init__(self, exp_conf, create=True, session=None):
Experiment.__init__(self, exp_conf, create, session)
self.test_conf = self.exp_conf.core_conf.test_conf
self.validation_conf = self.exp_conf.core_conf.validation_conf
self._train_test_exps = {}
def add_to_db(self):
Experiment.add_to_db(self)
self.session.add(DiademExpAlchemy(exp_id=self.exp_conf.exp_id,
fold_id=self.exp_conf.fold_id,
type=self.kind))
self.session.flush()
def run(self, train_instances, cv_monitoring=False):
Experiment.run(self)
self._train(train_instances)
if cv_monitoring:
self._cv_monitoring(train_instances)
self.monitoring.display(self.output_dir())
def run(self, test_instances):
Experiment.run(self)
self.test_instances = self.get_instances(test_instances)
self._test()
self._export()