👋 Introducing the Rasa Playground

Which is the difference between two policies format with the same policies but written differently

post by Asmacats on Sep 3, 2019

AsmacatsAsma Trabelsi

Sep 2019

Hi all,

Versions:
rasa 1.2.2
rasa_core 0.14.5
rasa_nlu 0.15.1

Using the rasa core, the next predicted action changes according to the defined policies within the config file. I have two config file with the same policies but written in a different manner. The config files are given below:

– Config1.yml –
policies:

• batch_size: 5
  epochs: 100
  name: KerasPolicy
• fallback_action_name: ‘utter_action_default_fallback’
  name: FallbackPolicy
  nlu_threshold: 0.4
  core_threshold: 0.4
• max_history: 3
  name: MemoizationPolicy
• name: FormPolicy
• name: MappingPolicy

– Config2.yml –
policies:

• name: KerasPolicy
  batch_size: 5
  epochs: 100
• name: FallbackPolicy
  fallback_action_name: ‘utter_action_default_fallback’
  nlu_threshold: 0.4
  core_threshold: 0.4
• name: MemoizationPolicy
  max_history: 3
• name: FormPolicy
• name: MappingPolicy

known that the first configuration has given the most accurate results, I would like to know the difference between these two configurations.

I would also like to know the main difference between the latest rasa version and rasa-for-botfront

Thanks for your help!!