huggingface / deep-rl-class

This repo contains the syllabus of the Hugging Face Deep Reinforcement Learning Course.
Apache License 2.0
3.92k stars 603 forks source link

[HANDS-ON BUG] boolean is a digital number other than 1 or 0 #532

Closed nanxusu closed 5 months ago

nanxusu commented 6 months ago

Describe the bug

https://colab.research.google.com/github/huggingface/deep-rl-class/blob/master/notebooks/unit1/unit1.ipynb?authuser=1#scrollTo=ZNPG0g_UGCfh

For unit1.ipynb, the final two values of the 8-dimention observation should be: If the left leg contact point has touched the land (boolean) and If the right leg contact point has touched the land (boolean).

However, the env.observation_space.sample() returned [-66.26277 87.800964 4.018273 3.077604 -0.1319175 -2.6315386 0.20755419 0.5502597 ], where the final two values are not boolean.

Material

If not:

simoninithomas commented 5 months ago

Hi, I'm agree with you but the official document seems to say https://gymnasium.farama.org/environments/box2d/lunar_lander/ that it's two boolean, I think the best is to open an issue to https://github.com/Farama-Foundation/Gymnasium