How can i save intermediate model checkpoints during rasa train

Rasa Open Source
1

rasa_version: 2.8.3 python_version: 3.8.0

I want to store all the models after every evaluate_every_number_of_epochs . so that i can use this models afterwards. Is there any way to make this happen other than checkpoint_model : True

Below is the config.yml is used:

language: en

pipeline:

# No configuration for the NLU pipeline was provided. The following default pipeline was used to train your model.

# If you’d like to customize it, uncomment and adjust the pipeline.

# See Tuning Your NLU Model for more information.

  • name: WhitespaceTokenizer
  • name: RegexFeaturizer
  • name: LexicalSyntacticFeaturizer
  • name: CountVectorsFeaturizer
  • name: CountVectorsFeaturizer analyzer: char_wb min_ngram: 1 max_ngram: 4
  • name: DIETClassifier epochs: 300 evaluate_on_number_of_examples: 40 evaluate_every_number_of_epochs: 5 tensorboard_log_directory: “tensorboard” tensorboard_log_level: “epoch” checkpoint_model: True constrain_similarities: True
  • name: EntitySynonymMapper

- name: ResponseSelector

epochs: 100

- name: FallbackClassifier

threshold: 0.3

ambiguity_threshold: 0.1

policies:

  • name: MemoizationPolicy
  • name: TEDPolicy max_history: 5 epochs: 300 evaluate_on_number_of_examples: 40 evaluate_every_number_of_epochs: 5 tensorboard_log_directory: “tensorboard” tensorboard_log_level: “epoch” checkpoint_model: True constrain_similarities: True
  • name: RulePolicy
2

If you want to do that you will have to modify the Rasa library. I think there is no easier way to save models one by one.