Hi,
I am facing the above warning from RASA while training my model. I don’t have any idea how should I fix this. Note: I am creating a chatbot for English and another for Bangla language. However, I am not getting this warning for the english model. I am using DIETClassifier for entity extraction.
In nlu.yml, I have almost 60 examples for this particular entity. Hope to get an answer soon from you guys and perdon my English (English is not my mother tongue).
Here is my config.yml file:
version: "2.0"
language: en
pipeline:
- name: WhitespaceTokenizer
- name: RegexFeaturizer #RegexEntityExtractor the name of the regular expression should match the name of the entity you want to extract.
- name: LexicalSyntacticFeaturizer
- name: CountVectorsFeaturizer
- name: CountVectorsFeaturizer
analyzer: "char_wb"
min_ngram: 1
max_ngram: 4
- name: DIETClassifier
epochs: 100
- name: FallbackClassifier
threshold: 0.7
- name: DucklingEntityExtractor
url: http://localhost:8000
dimensions:
- amount-of-money
- time
- number
- name: SpacyNLP
model: "en_core_web_md"
case_sensitive: false
- name: "SpacyEntityExtractor"
# Note: It is not possible to use the SpacyTokenizer + SpacyFeaturizer in
# combination with the WhitespaceTokenizer, and as a result the
# PERSON extraction by Spacy is not very robust.
# Because of this, the nlu training data is annotated as well, and the
# DIETClassifier will also extract PERSON entities.
dimensions: ["USERNAME"] #Replaced PERSON to USERNAME (by Rifat)
- name: EntitySynonymMapper
policies:
- name: AugmentedMemoizationPolicy
- name: TEDPolicy
epochs: 40
- name: RulePolicy
core_fallback_threshold: 0.4
core_fallback_action_name: "action_default_fallback"
enable_fallback_prediction: True
I have already check the existing questions and answer but I think those are not answer my problem.
@Rifat Hi, Please provide more training examples for cheque_number, as the error saying it has only 1 training example, but its should have more the 2 or max at least 10 examples for the same.
@Rifat You another question regarding Bangla language chatbot using spacy, please check this link for supported languages https://spacy.io/models
Hi, I have provided 60 training examples for cheque_number. But, still it’s not working and I have trained it by deleting all the older trained models too. No improvement at all.
The full warning is
/home/financial_demo_ban/financial_demo_ban/ban_env/lib/python3.7/site-packages/rasa/shared/utils/io.py:97: UserWarning: Entity entity ‘cheque_number’ has only 1 training examples! The minimum is 2, because of this the training may fail.
