Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

How to convert timestep based learning to episodic learning #1175

Open
muk465 opened this issue Sep 12, 2022 · 1 comment
Open

How to convert timestep based learning to episodic learning #1175

muk465 opened this issue Sep 12, 2022 · 1 comment

Comments

@muk465
Copy link

muk465 commented Sep 12, 2022

Question
I am using a custom environment to to path planning using ddpg algorithm , (i am using stable baslines3)

model = DDPG("MlpPolicy", env, action_noise=action_noise, verbose=1) model.learn(total_timesteps=10000, log_interval=1) model.save("sb3_ddpg_model")

here model.learn is used for timestep based learning but i want to convert it to certain like 3000 steps per episode and have multiple episodes,how can i achieve that?

@Missourl
Copy link

Hello,

I also have the same problem, but I have found thoses links. Maybe it could help you.
#352
#776

I still didnt not try the proposed solution, because I need to figure out, how could I combine it (proposed solution), with calling my seed in my env. So everytime I start a new episode "i" , the algo pass the "i" in the seed of my env (seed(i)) so I can sample new reproducible values.

Anyway Good luck .

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants