Merge branch 'master' into merge-1.8.x

changelog/4826.improvement.rst

@@ -0,0 +1,4 @@
+Add full retrieval intent name to message data
+``ResponseSelector`` will now add the full retrieval intent name
+e.g. ``faq/which_version`` to the prediction, making it accessible
+from the tracker.

rasa/nlu/selectors/response_selector.py

@@ -2,13 +2,16 @@
 
 import numpy as np
 import tensorflow as tf
+from pathlib import Path
 
 from typing import Any, Dict, Optional, Text, Tuple, Union, List, Type
 
+import rasa.utils.io as io_utils
 from rasa.nlu.config import InvalidConfigError
 from rasa.nlu.training_data import TrainingData, Message
 from rasa.nlu.components import Component
 from rasa.nlu.featurizers.featurizer import Featurizer
+from rasa.nlu.model import Metadata
 from rasa.nlu.classifiers.diet_classifier import (
     DIETClassifier,
     DIET,
@@ -66,9 +69,12 @@
 from rasa.nlu.constants import (
     RESPONSE,
     RESPONSE_SELECTOR_PROPERTY_NAME,
+    RESPONSE_KEY_ATTRIBUTE,
+    INTENT,
     DEFAULT_OPEN_UTTERANCE_TYPE,
     TEXT,
 )
+
 from rasa.utils.tensorflow.model_data import RasaModelData
 from rasa.utils.tensorflow.models import RasaModel
 
@@ -203,6 +209,7 @@ def __init__(
         index_label_id_mapping: Optional[Dict[int, Text]] = None,
         index_tag_id_mapping: Optional[Dict[int, Text]] = None,
         model: Optional[RasaModel] = None,
+        retrieval_intent_mapping: Optional[Dict[Text, Text]] = None,
     ) -> None:
 
         component_config = component_config or {}
@@ -211,6 +218,7 @@ def __init__(
         component_config[INTENT_CLASSIFICATION] = True
         component_config[ENTITY_RECOGNITION] = False
         component_config[BILOU_FLAG] = None
+        self.retrieval_intent_mapping = retrieval_intent_mapping or {}
 
         super().__init__(
             component_config, index_label_id_mapping, index_tag_id_mapping, model
@@ -231,6 +239,20 @@ def _check_config_parameters(self) -> None:
         super()._check_config_parameters()
         self._load_selector_params(self.component_config)
 
+    @staticmethod
+    def _create_retrieval_intent_mapping(
+        training_data: TrainingData,
+    ) -> Dict[Text, Text]:
+        """Create response_key dictionary"""
+
+        retrieval_intent_mapping = {}
+        for example in training_data.intent_examples:
+            retrieval_intent_mapping[
+                example.get(RESPONSE)
+            ] = f"{example.get(INTENT)}/{example.get(RESPONSE_KEY_ATTRIBUTE)}"
+
+        return retrieval_intent_mapping
+
     @staticmethod
     def _set_message_property(
         message: Message, prediction_dict: Dict[Text, Any], selector_key: Text
@@ -262,6 +284,9 @@ def preprocess_train_data(self, training_data: TrainingData) -> RasaModelData:
         label_id_index_mapping = self._label_id_index_mapping(
             training_data, attribute=RESPONSE
         )
+        self.retrieval_intent_mapping = self._create_retrieval_intent_mapping(
+            training_data
+        )
 
         if not label_id_index_mapping:
             # no labels are present to train
@@ -288,6 +313,7 @@ def process(self, message: Message, **kwargs: Any) -> None:
 
         out = self._predict(message)
         label, label_ranking = self._predict_label(out)
+        retrieval_intent_name = self.retrieval_intent_mapping.get(label.get("name"))
 
         selector_key = (
             self.retrieval_intent
@@ -299,10 +325,62 @@ def process(self, message: Message, **kwargs: Any) -> None:
             f"Adding following selector key to message property: {selector_key}"
         )
 
-        prediction_dict = {"response": label, "ranking": label_ranking}
+        prediction_dict = {
+            "response": label,
+            "ranking": label_ranking,
+            "full_retrieval_intent": retrieval_intent_name,
+        }
 
         self._set_message_property(message, prediction_dict, selector_key)
 
+    def persist(self, file_name: Text, model_dir: Text) -> Dict[Text, Any]:
+        """Persist this model into the passed directory.
+
+        Return the metadata necessary to load the model again.
+        """
+        if self.model is None:
+            return {"file": None}
+
+        super().persist(file_name, model_dir)
+
+        model_dir = Path(model_dir)
+
+        io_utils.json_pickle(
+            model_dir / f"{file_name}.retrieval_intent_mapping.pkl",
+            self.retrieval_intent_mapping,
+        )
+
+        return {"file": file_name}
+
+    @classmethod
+    def load(
+        cls,
+        meta: Dict[Text, Any],
+        model_dir: Text = None,
+        model_metadata: Metadata = None,
+        cached_component: Optional["ResponseSelector"] = None,
+        **kwargs: Any,
+    ) -> "ResponseSelector":
+        """Loads the trained model from the provided directory."""
+
+        model = super().load(
+            meta, model_dir, model_metadata, cached_component, **kwargs
+        )
+        if model == cls(component_config=meta):
+            model.retrieval_intent_mapping = {}
+            return model  # pytype: disable=bad-return-type
+
+        file_name = meta.get("file")
+        model_dir = Path(model_dir)
+
+        retrieval_intent_mapping = io_utils.json_unpickle(
+            model_dir / f"{file_name}.retrieval_intent_mapping.pkl"
+        )
+
+        model.retrieval_intent_mapping = retrieval_intent_mapping
+
+        return model  # pytype: disable=bad-return-type
+
 
 class DIET2DIET(DIET):
     def _check_data(self) -> None:

tests/nlu/selectors/test_selectors.py

@@ -4,6 +4,7 @@
 from rasa.nlu.training_data import load_data
 from rasa.nlu.train import Trainer, Interpreter
 from rasa.utils.tensorflow.constants import EPOCHS
+from rasa.nlu.constants import RESPONSE_SELECTOR_PROPERTY_NAME
 
 
 @pytest.mark.parametrize(
@@ -33,6 +34,12 @@ def test_train_selector(pipeline, component_builder, tmpdir):
     assert trainer.pipeline
 
     loaded = Interpreter.load(persisted_path, component_builder)
+    parsed = loaded.parse("hello")
 
     assert loaded.pipeline
-    assert loaded.parse("hello") is not None
+    assert parsed is not None
+    assert (
+        parsed.get(RESPONSE_SELECTOR_PROPERTY_NAME)
+        .get("default")
+        .get("full_retrieval_intent")
+    ) is not None