Could you please share the training model for the data used in your paper? i want to check actions from agent