def load_object(self, fname, category="", classifier=None):
"""Load an object, that was serialized with dump_object.
Args:
fname: Name of the serialized file.
category: Category name, that was used for serialization.
classifier: Classifier object, that was used for serialization.
"""
self.logger.info("Reading object from disk")
t2 = time.time()
if classifier == None:
classifier = self.classifier
try:
folder = util.folder_name(self.datamanager.PATHS["CLASSIFIER"],
category, classifier)
if not os.path.isdir(folder):
self.logger.info("Object's path doesn't exist")
return None
obj = joblib.load(os.path.join(folder, fname))
self.logger.info("%f seconds\n" % (time.time() - t2))
return obj
except Exception as e:
self.logger.error("Joblib failed: %s" % e)
return None
def load_object(self, fname, category="", classifier=None):
"""Load an object, that was serialized with dump_object.
Args:
fname: Name of the serialized file.
category: Category name, that was used for serialization.
classifier: Classifier object, that was used for serialization.
"""
self.logger.info("Reading object from disk")
t2 = time.time()
if classifier == None:
classifier = self.classifier
try:
folder = util.folder_name(self.datamanager.PATHS["CLASSIFIER"], category, classifier)
if not os.path.isdir(folder):
self.logger.info("Object's path doesn't exist")
return None
obj = joblib.load(os.path.join(folder, fname))
self.logger.info("%f seconds\n" % (time.time() - t2))
return obj
except Exception as e:
self.logger.error("Joblib failed: %s" % e)
return None
def test_folder_name(self):
clf = AdaBoostClassifier(n_estimators=23)
clf.base_estimator.max_depth = 42
base = "/hello/world/"
category = "testing"
params_path = util.params_to_path(clf.get_params())
self.assertEqual(
util.folder_name(base, category, clf),
os.path.join("/hello/world/AdaBoostClassifier/testing/",
params_path))
def dump_object(self, obj, fname, category="", classifier=None):
"""Serialize an object to disk.
This is most often used to serialize trained classifiers,
so they can be reloaded for later processing (e.g. visualization).
The object is serialized into a file in a subfolder
of the classifier-folder of self.datamanager.
The path will mirror the parameter settings of
the classifier object.
Objects that were serialized with this method can
be re-loaded with load_object.
Args:
obj: A python object, that should be serialized (e.g. a classifier).
fname: Name of the serialized file.
category: String that specifies, for which category the
classifier was trained. This introduces another directory level
to separate classifiers for different categories.
classifier: Classifier object, that determines the path
of the serialized file. The directory structure
mirrors the parameters of this classifier.
By default, self.classifier is used.
"""
self.logger.info("Writing object to disk")
t2 = time.time()
if classifier == None:
classifier = self.classifier
try:
folder = util.folder_name(self.datamanager.PATHS["CLASSIFIER"],
category, classifier)
if not os.path.isdir(folder):
os.makedirs(folder)
joblib.dump(obj, os.path.join(folder, fname), compress=3)
except Exception as e:
self.logger.error("Joblib failed: %s" % e)
self.logger.info("%f seconds\n" % (time.time() - t2))
def dump_object(self, obj, fname, category="", classifier=None):
"""Serialize an object to disk.
This is most often used to serialize trained classifiers,
so they can be reloaded for later processing (e.g. visualization).
The object is serialized into a file in a subfolder
of the classifier-folder of self.datamanager.
The path will mirror the parameter settings of
the classifier object.
Objects that were serialized with this method can
be re-loaded with load_object.
Args:
obj: A python object, that should be serialized (e.g. a classifier).
fname: Name of the serialized file.
category: String that specifies, for which category the
classifier was trained. This introduces another directory level
to separate classifiers for different categories.
classifier: Classifier object, that determines the path
of the serialized file. The directory structure
mirrors the parameters of this classifier.
By default, self.classifier is used.
"""
self.logger.info("Writing object to disk")
t2 = time.time()
if classifier == None:
classifier = self.classifier
try:
folder = util.folder_name(self.datamanager.PATHS["CLASSIFIER"], category, classifier)
if not os.path.isdir(folder):
os.makedirs(folder)
joblib.dump(obj, os.path.join(folder, fname), compress=3)
except Exception as e:
self.logger.error("Joblib failed: %s" % e)
self.logger.info("%f seconds\n" % (time.time() - t2))