Rev0kz
(Michael Kwaku Aboagye)
September 25, 2019, 10:37am
1
I have this set of data in the data.json
file.
{
"rasa_nlu_data": {
"common_examples": [
{
"text": "Hello",
"intent": "greeting",
"entities": []
},
{
"text": "Hi",
"intent": "greeting",
"entities": []
},
{
"text": "Goodmorning",
"intent": "greeting",
"entities": []
}
],
"regex_features": [],
"entity_synonyms": []
}
}
However when I train the model tensorflow_embedding
on this model, it says I need to add multiple intents or else training will fail.
How do I do it ?
Rev0kz:
tensorflow_embedding
Why not use pretrained_embeddings_spacy
?
Rev0kz
(Michael Kwaku Aboagye)
September 25, 2019, 1:03pm
3
Thank you. But I am buiding customized chatbot for my domain. That’s why I chose to use tensorflow_embedding
.
This is just a sample of training examples. I have more. But I want to know how to add multiple intents?
1 Like
In that case, just keep on adding other texts with different intent names like I have done below. I added the intent “bye” with texts “Bye” and “Good Bye”. Just make sure to add these intents in the domain.yml file and also create some stories with these intents.
{
"text": "Hi",
"intent": "greeting",
"entities": []
},
{
"text": "Goodmorning",
"intent": "greeting",
"entities": []
} ,
{
"text": "Bye",
"intent": "bye",
"entities": []
},
{
"text": "Good Bye",
"intent": "bye",
"entities": []
}
Rev0kz
(Michael Kwaku Aboagye)
September 26, 2019, 3:06pm
5
Okay. @GauravRoy48 Thank you very much. I will do so.
1 Like