Cannot retrieve path from "tokenizer_whitespace"

Rasa Open Source
1

Hey everyone!. Just started using rasa… My pipeline looks like this…

language: es

pipeline:

  • name: “tokenizer_whitespace”
  • name: “ner_crf”
  • name: “ner_synonyms”
  • name: “intent_featurizer_count_vectors”
  • name: “intent_classifier_tensorflow_embedding”

am I doing something wrong? versions: -rasa 2.4.3 -spacy 3.0.5 -nltk 3.6.1 -tensorflow 2.3.2 -python 3.6.13 (Using Anaconda)

thanks in advance!

2

I think you’ve gotten a configuration from an older version of Rasa. Where did you get this pipeline definition from?

Here’s a more standard pipeline template that should work. It’s the one that comes with rasa init. The only thing I’ve changed is the language setting from en → es.

language: es

pipeline:
  - name: WhitespaceTokenizer
  - name: RegexFeaturizer
  - name: LexicalSyntacticFeaturizer
  - name: CountVectorsFeaturizer
  - name: CountVectorsFeaturizer
    analyzer: char_wb
    min_ngram: 1
    max_ngram: 4
  - name: DIETClassifier
    epochs: 100
    constrain_similarities: true

policies:
  - name: MemoizationPolicy
  - name: TEDPolicy
    max_history: 5
    epochs: 100
    constrain_similarities: true
  - name: RulePolicy

Also, how did you install Rasa? Rasa 2.4.3, technically, doesn’t support spaCy 3.0.