D4RL icon indicating copy to clipboard operation
D4RL copied to clipboard

Maze2d rewards seem odd?

Open dh2shin opened this issue 2 years ago • 1 comments

Screen Shot 2022-06-28 at 11 12 38 AM Since maze2d-umaze-v1 uses a sparse reward type where a distance less than 0.5 yields a reward of 1, I was wondering how the above is possible. The first timeout field is at 91, yet the rewards leading up to it are all zeros. Would appreciate any thoughts!

dh2shin avatar Jun 28 '22 18:06 dh2shin

Hi did you solve this please?

TomQuilter avatar Feb 19 '24 10:02 TomQuilter