diff --git a/main.py b/main.py index c7f1cfc4..a1c4bb65 100644 --- a/main.py +++ b/main.py @@ -98,7 +98,8 @@ def run(config): ep_rews = replay_buffer.get_average_rewards( config.episode_length * config.n_rollout_threads) for a_i, a_ep_rew in enumerate(ep_rews): - logger.add_scalar('agent%i/mean_episode_rewards' % a_i, a_ep_rew, ep_i) + logger.add_scalar('agent%i/mean_episode_rewards' % a_i, + a_ep_rew * config.episode_length, ep_i) if ep_i % config.save_interval < config.n_rollout_threads: model.prep_rollouts(device='cpu')