brax
brax copied to clipboard
google
Massively parallel rigidbody physics simulation on accelerator hardware.
I'm currently doing a project related to transferring control policies from simulator to simulator. This lead me to try to standardize action and observation spaces between brax and mujoco. I...
currently, policy_params_fn exists only in ppo algorithm. are you going to update the policy clalback function to other algorithms too?
Hi, I am trying to run the [tutorial training notebook](https://colab.research.google.com/github/google/brax/blob/main/notebooks/training.ipynb) locally. However the HTML rendering seems to be quite flickering. The first render stacked on top of each other [simplescreenrecorder-2024-03-16_11.38.05.webm](https://github.com/google/brax/assets/3454586/156e5221-d388-436d-9b33-8a637b0e07d7)...
I'm on MacOS Sonoma, and when training completes, I get: ``` File "/Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/os.py", line 225, in makedirs mkdir(name, mode) OSError: [Errno 30] Read-only file system: '/saved_videos' ``` I've done `chmod...
It would be nice to have a wrapper under `brax.envs.wrappers` which takes an Env and a context length, and wraps the Env so that the obs space of the new...
When I try to render the scene without any contacts, I encounter the following error: ``` Traceback (most recent call last): File "/home/pcy/Research/code/crazyswarm2-adaptive/utils/redis_minimum/mbd_core.py", line 201, in main(tyro.cli(Args)) File "/home/pcy/Research/code/crazyswarm2-adaptive/utils/redis_minimum/mbd_core.py", line...
Fix `flags.DEFINE_integer('policy updates', None, Number of policy updates in APG.')` to have an underscore `flags.DEFINE_integer('policy_updates', None, Number of policy updates in APG.')`, and maybe add a test to check if...
Hi ! I'm doing research on value-based deep RL. I've been enjoying Brax very much lately ! My question is about which of the 4 backends (spring, positional, generalized, mjx)...
Hello, I have been encountering an issue where my training runs error free and learns well, but then control values of NaN are generated at inference when collecting a trajectory...
I have the following test file, which I use to test whether my domain randomization implementation works. The idea is that I condition (randomized) environments on the same action and...
