Hello, I would like to create a function for early stopping. I’ve observed that the number of simulation steps required varies for different functionalities during the search process. Therefore, I’m thinking of implementing a mechanism where, if step - best_reward_step > threshold, the search process will stop. How can I implement this function? Thank you.
Hello, I would like to create a function for early stopping. I’ve observed that the number of simulation steps required varies for different functionalities during the search process. Therefore, I’m thinking of implementing a mechanism where, if step - best_reward_step > threshold, the search process will stop. How can I implement this function? Thank you.