rasa train报错：tensorflow.python.framework.errors_impl.InvalidArgumentError: Incompatible shapes:

rasa 3.2

- name: JiebaTokenizer
- name: LanguageModelFeaturizer
  model_name: bert
  model_weights: rasa/LaBSE
- name: RegexFeaturizer
- name: DIETClassifier

解决方法：

清理nlu data中的空格

见Misalignment between LanguageModelFeaturizer and DIETClassifier in Chinese nlu data · Issue #10578 · RasaHQ/rasa · GitHub