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

[Hyper param tuning - early stop / green penalty] #68

Open
wants to merge 8 commits into
base: ppo-hyperparam-tuning
Choose a base branch
from

Conversation

xeviknal
Copy link
Owner

Adding hyperparameter tuning with extra wrappers:

  • early stop - finish the environment when the avg reward of the latest N steps is negative (similar to a curriculum learning)
  • green-penalty - reduce the rewards when the car is in the grass (reward -0.15 instead of -0.1).

See the tuning trails in the following comments.

@xeviknal
Copy link
Owner Author

xeviknal commented Apr 18, 2021

First experiment:
legend-juanjo-train_2021-04-14_22-37-25
tuning-juanjo-train_2021-04-14_22-37-25

@xeviknal
Copy link
Owner Author

xeviknal commented Apr 18, 2021

Second experiment:
legend-juanjo-train_2021-04-15_09-52-01
tuning-juanjo-train_2021-04-15_09-52-01

@xeviknal
Copy link
Owner Author

Third experiment:
legend-juanjo-train_2021-04-16_20-18-59
tuning-juanjo-train_2021-04-16_20-18-59

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

Successfully merging this pull request may close these issues.

1 participant