Error: embedding_intent_classifier

pipeline:
#   "supervised_embeddings"
  - name: "WhitespaceTokenizer"
    intent_tokenization_flag: true
    intent_split_symbol: "+"
  - name: "RegexFeaturizer"
  - name: "CRFEntityExtractor"
  - name: "EntitySynonymMapper"
  - name: "CountVectorsFeaturizer"
  - name: "CountVectorsFeaturizer"
    analyzer: "char_wb"
    min_ngram: 1
    max_ngram: 4
  - name: "EmbeddingIntentClassifier"
  - name: ResponseSelector

policies:
  - name: MemoizationPolicy
  - name: KerasPolicy
  - name: MappingPolicy

Why I’m getting the following error, what I’m doing wrong :

2019-11-01 14:38:36 INFO     rasa.nlu.model  - Starting to train component ResponseSelector
2019-11-01 14:38:36 ERROR    rasa.nlu.classifiers.embedding_intent_classifier  - Can not train a classifier. Need at least 2 different classes. Skipping training of classifier.
2019-11-01 14:38:36 INFO     rasa.nlu.model  - Finished training component.
2019-11-01 14:38:36 INFO     rasa.nlu.model  - Successfully saved model into '/tmp/tmpm1pb9sq0/nlu'
NLU model training completed.

Hi @sten,

as the error you posted shows:

you don’t seem to have at least two different intents. Can you copy/paste your nlu.md content?

Regards Julian

nlu.md (2.6 KB)

Hi @sten,

could you please remove the ResponseSelector and try to train again?

Regards Julian

that was it :frowning: , btw i would need it for FAQ handling

quick question :

if I have multiple *.md files in /data, does rasa train pick them up … or there is some way to point them

Hi @sten,

glad that I could help. You actually didnt define any response intents (they need to be provided in a special way) but used their classifier which would not work.

Jeah - rasa picks those files together.

Regards Julian

i have responses.md file

how do i make it work with ResponseSelector ? I followed :

i made it work

I have same question ,can you tell me how to make it work?

as far as I can remember you probably have #intent which have only one item …

 #intent: abc
 - abcd

should have more than one :

#intent: abc
 - abcd
 - defg