Hi, I started to use RASA recently. My goal is to try different ML models for the policies and study their qualities . For that, I need to understand what is the data fed to the policy models such as KerasPolicy and how it is structured. Can somebody elaborate preferably with an example? [Edit]
Basically, after digging more, I found that the model is trained with a (3541, 5, 95) Matrix. I could understand that 95 is related to the number of states. Unfortunately, I didn’t get it for 3541 and 5. I am using the policy with 100 epochs and batches of size 50. Any clues?
Thanks.