What is the meaning of the episode_reward_mean metric? Is it the sum of the reward obtained in each time step of the episode? What is the difference between episode_reward_mean and episode_reward_min?

Hi @carlorop , RLLIB collects a number of metrics. One of those is the episode_reward. When creating the summary it will compute and store the mean, min , and max of those metrics. The metrics summarize all of the collected values during the previous iteration. What defines an iteration varies. If …

Meaning of episode_reward_mean

RLlib

Roller44 May 10, 2022, 1:42pm 5

If I want to obtain the reward in the latest episode, what setting should I change?

Topic		Replies	Views
[RLlib, Tune, PPO] episode_reward_mean based on new episodes for each iteration Configure Algorithm, Training, Evaluation, Scaling	1	25	November 25, 2024
How rllib train log the reward on tensorboard? RLlib	1	517	March 25, 2022
Why the episode reward mean is always the same number for a while(about 10 iters)? RLlib	1	726	May 4, 2021
Any other metric other than "episode_reward_mean" Configure Algorithm, Training, Evaluation, Scaling	3	55	October 16, 2024
How to obtain single episode reward? RLlib	6	1408	March 19, 2024

Meaning of episode_reward_mean

Related topics