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

Baseline approach for different maps #212

Open
Tarster opened this issue Aug 20, 2024 · 1 comment
Open

Baseline approach for different maps #212

Tarster opened this issue Aug 20, 2024 · 1 comment

Comments

@Tarster
Copy link

Tarster commented Aug 20, 2024

Hi Lucas,

I am implementing different algorithms on different net provided in the library, but I want to simulate the network with a fixed timing and compare the rewards function output for different algorithms.

For ex: I want to simulate the network single intersection with a fixed time control and get the output of the network (reward gathered) as in this scenario the default reward function or queue reward function, at the end I want to show both the rewards on same chart. The rewards obtained by fixed timing and the rewards obtained by Q-Learning. How can I achieve this?

I need help in running the simulation with the fixed timing, I will create the visualization myself.

Thanks for the help once again.

@Everywhere08
Copy link

you can set the fixed_ts to true
here is the documentation https://lucasalegre.github.io/sumo-rl/documentation/sumo_env/

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