def evaluate(self):
classifier = Recurrent()
classifier.retrieve_samples(Path.FEATURE_PATH)
classifier.evaluate()
classifier.relevance.output_confusion_matrix("{}confusion_matrix{}.png".format(Path.RESULT_PATH, self.run_name))
classifier.relevance.output_statistics("{}statistics.md".format(Path.RESULT_PATH), self.run_name)
def predict(self, session_id):
self.process(session_id)
classifier = Recurrent()
classifier.retrieve_sample("{}{}.data".format(Path.FEATURE_PATH,
session_id),
is_labelled=False)
classifier.evaluate(is_labelled=False)
def evaluate(self):
classifier = Recurrent()
classifier.retrieve_samples(Path.FEATURE_PATH)
classifier.evaluate()
classifier.relevance.output_confusion_matrix(
"{}confusion_matrix{}.png".format(Path.RESULT_PATH, self.run_name))
classifier.relevance.output_statistics(
"{}statistics.md".format(Path.RESULT_PATH), self.run_name)
def predict(self, session_id):
self.process(session_id)
classifier = Recurrent()
classifier.retrieve_sample("{}{}.data".format(Path.FEATURE_PATH, session_id), is_labelled=False)
classifier.evaluate(is_labelled=False)
def get_classifier(self, neurons_per_layer):
classifier = Recurrent(neurons_per_layer)
classifier.retrieve_samples(Path.FEATURE_PATH)
return classifier
def get_classifier(self, neurons_per_layer):
classifier = Recurrent(neurons_per_layer)
classifier.retrieve_samples(Path.FEATURE_PATH)
return classifier