def _create_session_data(
self,
training_data: "TrainingData",
label_id_dict: Dict[Text, int],
attribute: Text,
) -> "train_utils.SessionData":
"""Prepare data for training and create a SessionData object"""
X = []
label_ids = []
Y = []
for e in training_data.intent_examples:
if e.get(attribute):
X.append(
e.get(
MESSAGE_VECTOR_FEATURE_NAMES[MESSAGE_TEXT_ATTRIBUTE]))
label_ids.append(label_id_dict[e.get(attribute)])
X = np.array(X)
label_ids = np.array(label_ids)
for label_id_idx in label_ids:
Y.append(self._encoded_all_label_ids[label_id_idx])
Y = np.array(Y)
return train_utils.SessionData(X=X, Y=Y, label_ids=label_ids)
def _create_session_data(
self, data_X: "np.ndarray", data_Y: Optional["np.ndarray"] = None
) -> "train_utils.SessionData":
"""Combine all tf session related data into a named tuple"""
if data_Y is not None:
# training time
label_ids = self._label_ids_for_Y(data_Y)
Y = self._label_features_for_Y(label_ids)
# idea taken from sklearn's stratify split
if label_ids.ndim == 2:
# for multi-label y, map each distinct row to a string repr
# using join because str(row) uses an ellipsis if len(row) > 1000
label_ids = np.array([" ".join(row.astype("str")) for row in label_ids])
else:
# prediction time
label_ids = None
Y = None
return train_utils.SessionData(X=data_X, Y=Y, label_ids=label_ids)
