I am using LUIS model(exported as json file) to build rasa_nlu using command "rasa train --out models\test.tar.gz"
Luis model has entities of type list and regex, which rasa_nlu model is unable to extract.
for example: LUIS Entity name: "Animal" LUIS Entity Type: "List" Elements in list would be like cat, dog, etc...
If i present string like "How to train dog"?
I am expecting "dog" to be mapped to "Animal" entity just like how LUIS does but rasa_nlu doesn't detect "dog" as entity since it is of type "list". If entity type is changed to "simple" rasa_nlu detects dog and classifies it as "Animal" entity type
Similarly, if luis entity type is of regex identified by "0x[a-f,0-9]{8}/i", rasa_nlu fails to detect it.
Are there any issues with rasa_nlu in detecting luis regex and list entity types or am i doing anything wrong?
below are the contents of json file used to train rasa
{
"luis_schema_version": "3.2.0",
"versionId": "0.1",
"name": "Test",
"desc": "",
"culture": "en-us",
"tokenizerVersion": "1.0.0",
"intents": [
{
"name": "error"
},
{
"name": "None"
},
{
"name": "wild"
}
],
"entities": [],
"composites": [],
"closedLists": [
{
"name": "animal",
"subLists": [
{
"canonicalForm": "dog",
"list": []
},
{
"canonicalForm": "cat",
"list": []
}
],
"roles": []
}
],
"patternAnyEntities": [],
"regex_entities": [
{
"name": "ecode",
"regexPattern": "0x[a-f,0-9]{8}",
"roles": []
}
],
"prebuiltEntities": [],
"model_features": [],
"regex_features": [],
"patterns": [],
"utterances": [
{
"text": "error 0x80070055",
"intent": "error",
"entities": []
},
{
"text": "error while browsing 0x90909090",
"intent": "error",
"entities": []
},
{
"text": "how to train cat",
"intent": "wild",
"entities": []
},
{
"text": "how to train dog",
"intent": "wild",
"entities": []
}
],
"settings": []
}
appreciate your help!
User contributions licensed under CC BY-SA 3.0