search

RasaHQ / rasa

💬 Open source machine learning framework to automate text- and voice-based conversations: NLU, dialogue management, connect to Slack, Facebook, and more - Create chatbots and voice assistants
https://rasa.com/docs/rasa/
Apache License 2.0
18.81k stars 4.62k forks source link

case sensitive #1355

Closed AdelePh closed 6 years ago

AdelePh commented 6 years ago
**Rasa NLU version**: 0.13.0a2 **Operating system** (windows, osx, ...): Ubuntu **Content of model configuration file**: ```yml pipeline: - name: "nlp_spacy" - name: "tokenizer_spacy" - name: "ner_crf" - name: "ner_synonyms" - name: "sentiment_textblob" - name: "intent_featurizer_count_vectors" - name: "intent_classifier_tensorflow_embedding" intent_tokenization_flag: true intent_split_symbol: "+" ``` **Issue**: in my training data, I have examples: ``` "I am a doctor" - intent : inform - entity : job (doctor) "I am Adele" - intent : intro - entity : name (Adele) ``` I want the bot to capture the "name" slot to greet user : Hi {name} but Rasa NLU is getting confused with this case.
AdelePh commented 6 years ago

anyone help me please !!!

akelad commented 6 years ago

If those are your only two training examples, then it's impossible for the model to pick up the entities correctly. As a first step, please add more training examples. And since this is more of a question than an issue, please post it in our forum :)