InvalidConfigException: Can't load class for name 'HFTransformersNLP'. in rasa

Question

how to implement BERT in rasa with huggingface transformers and what are needed for running the Bert model in rasa ?

recipe: default.v1
*# Configuration for Rasa NLU.
# https://rasa.com/docs/rasa/nlu/components/*
language: en
pipeline:
*# how to implement this BERT in rasa* 
  - name: HFTransformersNLP
    model_weights: "bert-base-uncased"
    model_name: "bert"
  - name: LanguageModelTokenizer
  - name: LanguageModelFeaturizer
  - name: DIETClassifier
    epochs: 200

Answer 1

This error could be due to the Rasa version you're using (output of rasa --version). In the current versions (>2.1), HFTransformersNLP and LanguageModelTokenizer are deprecated. Using a BERT model can be achieved with any tokenizer and

pipeline:
  - name: LanguageModelFeaturizer
    model_name: "bert"
    model_weights: "rasa/LaBSE"

See the documentation for further details.