JamesHeald

[Feature Request] Recurrent policies

16

There are recurrent (LSTM) policy options for sb3 (e.g. [RecurrentPPO](https://github.com/Stable-Baselines-Team/stable-baselines3-contrib/blob/master/sb3_contrib/ppo_recurrent/ppo_recurrent.py)). It would be great to have recurrent PPO implemented for sbx.

I have implemented a diffusion probabilistic model for the first time using a cosine noise schedule, and I am finding that the variance/standard deviation of samples drawn from the reverse...

Basic arm reaching environment created, myoArmReachRandom-v0

11

I have created a basic environment for reaching to random targets in space with the MyoSuite arm. - I created a new myoarm_reach.xml file by modifying the existing myoarm_relocate.xml file...

Solution to issue #2489

1

I believe this PR resolves issue #2489. The solution is to perform code branching using the conditional control flow jax.lax.cond instead of jp.where. From what I understand, jax.lax.cond only evaluates...

Draft PR for MJX-compatible MyoSuite

8

This is a draft PR for the integration of MuJoCo XLA (MJX) into MyoSuite. Our goal is to create MJX-compatible versions of the MyoSuite environments, which will allow the environments...

I am training a policy using SAC on the myoHandReorient environments. Regularly throughout training, messages appear saying that the simulation is unstable: ``` WARNING:absl:Nan, Inf or huge value in QACC...

Brax PPO can be passed a policy_params_fn: policy_params_fn: a user-defined callback function that can be used for saving custom policy checkpoints or creating policy rollouts and videos I personally find...