CODEWITHSUNDEEP
rasa-nlurasarasa-core
dper
dper

Reputation: 904

Lookup table not working after training the model in rasa

I am new to rasa. I am training a model to recognize certain entities using lookup tables. I have multiple entities in the single sentence and I am trying to extract them.

nlu.yml

version: "2.0"
nlu:
- intent: intent_1
  examples : |
    - how many deaths were there last year in [Ohio](Filter-State)?
    - death count of [Florida](Filter-State) this year
    - death count of [Texas](Filter-State) this year
    - what's the death count for this quarter in [CA](Filter-State)?
- lookup: Filter-State
  examples: |
    - Alabama
    - AL
    - Alaska
    - AK
    - Arizona
    - AZ
    - Arkansas
    - AR
    - California
    - CA
    - Colorado
    - CO
    - Connecticut
    - CT
    - Delaware
    - DE
    - District of Columbia
    - DC
    - Florida
    - FL
    - Georgia
    - GA

config.yml

language: en
pipeline:
  - name: WhitespaceTokenizer
  - name: RegexFeaturizer
  - name: LexicalSyntacticFeaturizer
  - name: CountVectorsFeaturizer
  - name: CountVectorsFeaturizer
    analyzer: "char_wb"
    min_ngram: 1
    max_ngram: 4
  - name: DIETClassifier
    epochs: 150
    random_seed: 1
  - name: FallbackClassifier
    threshold: 0.7
  - name: DucklingEntityExtractor
    url: http://duckling.rasa.com:8000
    dimensions:
    - email
    - time
  - name: EntitySynonymMapper
policies:
  - name: AugmentedMemoizationPolicy
    max_history: 4
  - name: TEDPolicy
    max_history: 4
    epochs: 100
  - name: RulePolicy
    core_fallback_threshold: 0.4
    core_fallback_action_name: "action_default_fallback"
    enable_fallback_prediction: True

When I train the model and try using the api, It doesn't recognize cases from the states in the lookup table and as a result can't assign it to slot filter_state.

Can anyone advise me as to what am I doing wrong here for making the lookup table work!

Upvotes: 3

Views: 1842

Answers (2)

Adam Simons
Adam Simons

Reputation: 71

I'm new to Rasa and searching for another issue, but I just ran into and solved this issue last night.

For lookup tables to work, you need to add "RegexEntityExtractor" to your pipeline and possibly remove RegexFeaturizer. You also need to enable lookup tables in the RegexEntityExtractor config.

config.yml

pipeline:
   - name: WhitespaceTokenizer
   - name: LexicalSyntacticFeaturizer
   - name: CountVectorsFeaturizer

   - name: RegexEntityExtractor
     case_sensitive: False
     use_lookup_tables: True
     use_regexes: True
...

Upvotes: 7

SamS
SamS

Reputation: 36

Could you, please, post on the Rasa forum and include more details of your setup? In particular, what version of Rasa Open Source are you using? Is the above your complete NLU data? (I think that you need at least 2 intents for an intent classifier to train.) I also recommend that you test the system using rasa interactive --debug and share e.g. a screenshot, this will help everyone see the exact input message and how it gets processed by Rasa. I'm sure we'll trace the issue down to its roots :-)

Upvotes: 0

Related Questions