Which is the unit of “save_best_after” in “*** PPO.yaml”, “iteration” or “episode”?

Hi
Setting “save_best_after” in “*** PPO.yaml” to ‘1〜15’ does not reflect in my task.
Which is the unit of “save_best_after”, “iteration” or “episode”?

Also, for example, does the “ep” below stand for “episode”?
saving checkpoint ‘runs/FrankaCabinet/nn/last_FrankaCabinet_ep_40_rew_1016.42426.pth’

Also, for example, does the “ep” below stand for “episode”?

I was confused. It’s “epoch”.
It would be better to change from “ep” to “epo”.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.