Reinforcement Learning Scripts
We provide wrappers to different reinforcement libraries. These wrappers convert the data from the environments into the respective libraries function argument and return types.
RL-Games
When using RL-Games with the Ray workflow for distributed training or hyperparameter tuning, please be aware that due to security risks associated with Ray, this workflow is not intended for use outside of a strictly controlled network environment.
-
Training an agent with RL-Games on
Isaac-Ant-v0:Linux
Windows
RSL-RL
-
Training an agent with RSL-RL on
Isaac-Reach-Franka-v0:Linux
Windows
-
Training and distilling an agent with RSL-RL on
Isaac-Velocity-Flat-Anymal-D-v0:Linux
Windows
SKRL
-
Training an agent with SKRL on
Isaac-Reach-Franka-v0:PyTorch
JAX
Linux
Windows
- Training the multi-agent environment
Isaac-Shadow-Hand-Over-Direct-v0with skrl:
Linux
Windows
- Training the multi-agent environment
Stable-Baselines3
-
Training an agent with Stable-Baselines3 on
Isaac-Velocity-Flat-Unitree-A1-v0:Linux
Windows
All the scripts above log the training progress to Tensorboard in the logs directory in the root of
the repository. The logs directory follows the pattern logs/<library>/<task>/<date-time>, where <library>
is the name of the learning framework, <task> is the task name, and <date-time> is the timestamp at
which the training script was executed.
To view the logs, run: