Training Issue

[jeet09 (sarabjeet)]

Hello Everyone,

Everytime i am training the model. My bot is giving different responses … Can anyone help?

Below is the config file details that i am using.

Configuration for Rasa NLU.

https://rasa.com/docs/rasa/nlu/components/

language: en pipeline: supervised_embeddings

Configuration for Rasa Core.

https://rasa.com/docs/rasa/core/policies/

policies:

• name: FormPolicy
• name: MemoizationPolicy max_history: 18
• name: TEDPolicy epochs: 2 max_history: 18
• name: MappingPolicy
• name: “FallbackPolicy” nlu_threshold: 0.06 core_threshold: 0.06 fallback_action_name: “action_default_fallback”

- name: KerasPolicy

[fkoerner (Felicia Koerner)]

Hi! It looks like you’re using a few deprecated policies, and your max_history is quite large, and your fallback threshold is quite low. What version of Rasa are you on? What do your training stories look like? What behaviour are you seeing (different responses for the same user utterances?)