Break Yang

> You can use `gin.add_config_file_search_path('path/to/config.gin')` to add **folders** that you want gin to use when doing `gin.parse_config_files_and_bindings(...)` Thanks for sharing this. The problem can definitely be solved by automatically add...

> > Lastly multi agent learning works with ML-Agents with the num-envs argument when running from build. This wastes some resources due to running multiple instances of mujoco in the...

I was reading https://github.com/stanfordnlp/dspy/blob/main/dspy/teleprompt/bootstrap.py#L145 and it seems to suggest that the metric is treated as a `bool` anyway. Did I miss anything? Thanks!

At this moment, I just implement them with jax functions following the logic implemented in c++: https://github.com/google-deepmind/mujoco/blob/main/src/engine/engine_sensor.c#L525

> Does anyone want to send us a PR implementing this in the MJX codebase? That would be welcome. Never contributed to MuJoCo before, but I would like to give...

A height map with random pattern (e.g. Perlin noise) is usually used as a terrain to train locomotion of robot e.g. with RL so that the produced policy is robust....

I see. Thanks for the explanation @btaba ! In that case, maybe an alternative would be to approximate uneven plane with something like "stairs"?

Yes using boxes should also work. Thanks for sharing the code snippet! Didn't know about those two options, TIL.

> Thanks for responding @kevinzakka! I'm not using Docker, just a conda environment. I think the error occurs on every GPU I've tried so far: A100 SXM4, RTX 2080Ti, T4...

> This one, on all machines: `| NVIDIA-SMI 525.60.13 Driver Version: 525.60.13 CUDA Version: 12.0 |` I also ran into this problem consistently on all my machines, with different Nvidia...