# Isaac Lab: RSL-RL policy training

Train an RL policy with the RSL-RL backend and an explicitly selected Isaac Lab task. Inspect the pinned source and setup before use.

ID: `isaaclab-scripts-reinforcement-learning-rsl-rl-train-py`

Record version: `source-e12938bd4807d10d3972bb257a2edd90a156824b326c5f345f48ac8d09a84a21`

Hub commit: `66a13cd6ae8ab828219a4916587fb83d6dc71809`

## Access

Fetch the pinned repository, verify this source checksum, and use upstream installation and invocation. This record does not install or execute code.

Optional inert file retrieval: `python /PATH_TO_SKILL/scripts/rsh.py fetch-source isaaclab-scripts-reinforcement-learning-rsl-rl-train-py --expected-source-revision 37ddf626871758333d6ed89cf64ad702aef127d0 --expected-source-sha256 e12938bd4807d10d3972bb257a2edd90a156824b326c5f345f48ac8d09a84a21 --project ./experiment --json`. This verifies one file, preserves provenance and never installs or runs it. See [source-file retrieval](https://hub.superfastrobotics.com/source-files.md).

Hub preparation adapter: **false**. Runtime verification: **not_run**.

Upstream setup: https://github.com/isaac-sim/IsaacLab/blob/37ddf626871758333d6ed89cf64ad702aef127d0/README.md

## Source and license

- Repository: https://github.com/isaac-sim/IsaacLab
- Pinned source: https://github.com/isaac-sim/IsaacLab/blob/37ddf626871758333d6ed89cf64ad702aef127d0/scripts/reinforcement_learning/rsl_rl/train.py
- Revision: `37ddf626871758333d6ed89cf64ad702aef127d0`
- Path: `scripts/reinforcement_learning/rsl_rl/train.py`
- Source SHA256: `e12938bd4807d10d3972bb257a2edd90a156824b326c5f345f48ac8d09a84a21`
- Source bytes: 9112
- Original authors: The Isaac Lab Project Developers
- License: [BSD-3-Clause](https://github.com/isaac-sim/IsaacLab/blob/37ddf626871758333d6ed89cf64ad702aef127d0/LICENSE)
- License scope: Selected first-party workflow Python source only; preserve all upstream copyright, author and per-file notices. This does not relicense model weights, datasets, Gemma or other third-party model code, robot firmware, simulator binaries, meshes, dependencies or external services; their separate terms and access requirements remain applicable.

License SHA256: `df351ebfa8af317bf5a9ad83df63a8b68e4f5c04b71f30767cdb27aa434eac62`

## Artifact

```json
{
  "kind": "workflow",
  "format": "python",
  "workflow_role": "policy-training"
}
```

## Requirements

```json
{
  "python": "Use the Python environment required by the pinned Isaac Lab and matching Isaac Sim installation",
  "packages": [
    "Pinned Isaac Lab checkout",
    "Compatible Isaac Sim installation",
    "Selected RL-Games, RSL-RL, Stable Baselines3 or skrl backend"
  ],
  "display": "A supported NVIDIA GPU, matching Isaac Sim/Isaac Lab environment and explicitly selected --task configuration are required. Play scripts also require a matching trained checkpoint; robot morphology and assets are determined by that task.",
  "execution_domain": "simulation",
  "source_imports": [
    "argparse",
    "cli_args",
    "datetime",
    "gymnasium",
    "importlib.metadata",
    "isaaclab.app",
    "isaaclab.envs",
    "isaaclab.utils.dict",
    "isaaclab.utils.io",
    "isaaclab_rl.rsl_rl",
    "isaaclab_tasks",
    "isaaclab_tasks.utils",
    "isaaclab_tasks.utils.hydra",
    "logging",
    "os",
    "packaging",
    "platform",
    "rsl_rl.runners",
    "sys",
    "time",
    "torch"
  ]
}
```

## Invocation

```json
{
  "entrypoint": "./isaaclab.sh -p scripts/reinforcement_learning/rsl_rl/train.py",
  "notes": "Use the pinned upstream Isaac Lab launcher with the selected --task and RL backend configuration; play requires a compatible checkpoint. Simulator assets and third-party components retain their own terms. No simulator, package, robot model or checkpoint was loaded during source indexing."
}
```

## Compatibility

```json
[]
```

Missing environment fields remain unknown. Source discovery labels do not establish robot compatibility.

## Evidence

Level: **source_reviewed**. Runtime verification: **not_run**.

Verified exact source, applicable license and author/setup document bytes at a full Git commit; parsed the Python AST and reviewed this guarded operational entrypoint and its explicit license notices. No upstream code, installation, robot command, dataset, mesh or checkpoint was executed or loaded.

Source review and contributor reports do not establish task completion. Preserve upstream attribution and licenses.