DLR-RM / stable-baselines3

PyTorch version of Stable Baselines, reliable implementations of reinforcement learning algorithms.
https://stable-baselines3.readthedocs.io
MIT License
8.38k stars 1.61k forks source link

[Question] Control PPO training #1872

Closed mwalidcharrwi closed 3 months ago

mwalidcharrwi commented 3 months ago

❓ Question

Hi I wanted to inquire is there a way to control the training of the MaskablePPO when it finishes the data set that is to be trained on (when ep_len_mean reaches 1 and ep_rew_mean reaches 100 in the logger) or at any specific point during the learning?

Checklist