"...has only 1 training examples! Minimum is 2, training may fail."

When I go to train my model with

rasa train

and the following pipeline:

language: en

  - name: SpacyNLP
  - name: SpacyTokenizer
  - name: SpacyFeaturizer
  - name: RegexFeaturizer
  - name: LexicalSyntacticFeaturizer
  - name: CountVectorsFeaturizer
  - name: CountVectorsFeaturizer
analyzer: "char_wb"
min_ngram: 1
max_ngram: 4
  - name: DIETClassifier
epochs: 100
  - name: EntitySynonymMapper
  - name: ResponseSelector
epochs: 100

I get the following ERROR:

UserWarning: Intent β€˜immd_family/child_count’ has only 1 training examples! Minimum is 2, training may fail.

This doesn’t make sense, given that my nlu.yml file has the following:

- intent: immd_family/child_count
  examples: |
    - how many kids did mary have?
    - How many children did Joseph have
    - How many babies were born to Elizabeth?

That looks like three examples to me. What’s going on here?

Thanks for the help!

1 Like

yes that seems off. what is the output of rasa data validate ?

2020-11-09 06:35:02 INFO rasa.validator - Validating intents… 2020-11-09 06:35:02 INFO rasa.validator - Validating uniqueness of intents and stories… 2020-11-09 06:35:02 INFO rasa.validator - Validating utterances… 2020-11-09 06:35:02 INFO rasa.validator - Story structure validation… Processed story blocks: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 1/1 [00:00<00:00, 504.79it/s, # trackers=1] Processed rules: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1295.02it/s, # trackers=1] 2020-11-09 06:35:02 INFO rasa.core.training.story_conflict - Considering the preceding 8 turns for conflict analysis. 2020-11-09 06:35:02 INFO rasa.validator - No story structure conflicts found.

Maybe I’m just not quite sure exactly how to use the retrieval intents? I’ve tried reading the docs but I am still pretty confused regarding how and when to use them, why they are different or helpful from intents, and how I need to set up my documents in order for training to go smoothly.

@amn41 Any ideas here? I am still seeing this happen on other intents as well.

@agm this sounds like a bug. Would you be up for creating an issue with a minimal reproducible example? that would help a lot and we can get on fixing it.