meksikann
November 1, 2018, 10:23am
1
Hey guys. Is there predefined email entity extractor in pipeline? Or shell I teach nlu to find the entity?
I am using rasa-core 0.11
here is nlu pipeline:
language: "en"
pipeline:
- name: "nlp_spacy"
- name: "tokenizer_spacy"
- name: "intent_featurizer_spacy"
- name: "intent_classifier_sklearn"
- name: "ner_crf"
- name: "ner_synonyms"
- name: "ner_duckling_http"
url: "http://0.0.0.0:8000"
dimensions: ["time", "duration"]
thanks:)
Neofita
November 1, 2018, 12:36pm
2
I am not an expert,
however, from this demo GitHub - RasaHQ/rasa-demo: An example of a cool Rasa bot
in domain you must have this
enter_data: {use_entities: false}
You need to create an action
# -*- coding: utf-8 -*-
import logging
from rasa_core_sdk import Action
from rasa_core_sdk.events import SlotSet, UserUtteranceReverted, \
ConversationPaused
from demo.api import MailChimpAPI
from demo import config
from demo.gdrive_service import GDriveService
logger = logging.getLogger(__name__)
class ActionSubscribeNewsletter(Action):
""" This action calls our newsletter API and subscribes the user with
their email address"""
def name(self):
show original
voir action_store_email
and a story rasa-demo/chitchat.md at master · RasaHQ/rasa-demo · GitHub
signup_newsletter
utter_great
utter_ask_email
enter_data{“email”: “maxmeier@firma.de ”} OR enter_data{“number”:“1”}
action_store_email
slot{“email”: “maxmeier@firma.de ”}
meksikann
November 2, 2018, 10:28am
3
pipeline:
name: “nlp_spacy”
name: “tokenizer_spacy”
name: “intent_featurizer_spacy”
name: “intent_classifier_sklearn”
name: “ner_crf”
name: “ner_synonyms”
name: “ner_duckling_http”
url: “http://0.0.0.0:8000 ”
dimensions: [“time”, “duration”, “email”]
used “email” in duckling config. seems it worked:)
1 Like