def _predict_failure(self, ip_feature_db): """ Turn the ip_feature_db into two dimensional array and feed it to all classifiers. """ failList = list() ip_set = TrainingSet() for cur_ip in ip_feature_db: ip_set.add_ip(cur_ip, ip_feature_db[cur_ip]) for fail_classifier in self.__fail_classifiers: bad_ip_prediction = fail_classifier.predict(ip_set._ip_feature_list) failList.extend([ip_set._ip_index[i] for i in range(0, len(bad_ip_prediction)) if bad_ip_prediction[i] == ip_set.BAD_TARGET]) return failList
def _predict_failure(self, ip_feature_db): """ Turn the ip_feature_db into two dimensional array and feed it to all classifiers. """ failList = list() ip_set = TrainingSet() for cur_ip in ip_feature_db: ip_set.add_ip(cur_ip, ip_feature_db[cur_ip]) for fail_classifier in self.__fail_classifiers: bad_ip_prediction = fail_classifier.predict( ip_set._ip_feature_list) failList.extend([ ip_set._ip_index[i] for i in range(0, len(bad_ip_prediction)) if bad_ip_prediction[i] == ip_set.BAD_TARGET ]) return failList
def _predict_failure(self, ip_feature_db): """ Turn the ip_feature_db into two dimensional array and feed it to all classifiers. """ failList = list() ip_set = TrainingSet() for fail_model in self._fail_models: ip_set._normalisation_data = fail_model.getNormalisationData() ip_set._normalisation_function = ip_set.normalise_individual if ip_set._normalisation_data[TrainingSet.NORMALISATION_TYPE] == 'sparse': ip_set._normalisation_function = ip_set.normalise_sparse ip_set = ip_set.precook_to_predict(ip_feature_db) print ip_set._ip_feature_array bad_ip_prediction = fail_model.getClassifier().predict(ip_set._ip_feature_array) failList.extend([ip_set._ip_index[i] for i in range(0, len(bad_ip_prediction)) if bad_ip_prediction[i] == ip_set.BAD_TARGET]) return failList
def _predict_failure(self, ip_feature_db): """ Turn the ip_feature_db into two dimensional array and feed it to all classifiers. """ failList = list() ip_set = TrainingSet() for fail_model in self._fail_models: ip_set._normalisation_data = fail_model.getNormalisationData() ip_set._normalisation_function = ip_set.normalise_individual if ip_set._normalisation_data[ TrainingSet.NORMALISATION_TYPE] == 'sparse': ip_set._normalisation_function = ip_set.normalise_sparse ip_set = ip_set.precook_to_predict(ip_feature_db) print ip_set._ip_feature_array bad_ip_prediction = fail_model.getClassifier().predict( ip_set._ip_feature_array) failList.extend([ ip_set._ip_index[i] for i in range(0, len(bad_ip_prediction)) if bad_ip_prediction[i] == ip_set.BAD_TARGET ]) return failList