Octo: WidowX real-robot policy evaluation

Pinned didactic Octo evaluation entrypoint for a separately configured real WidowX robot and matching fine-tuned checkpoint. Statically reviewed source only; not run and no Hub preparation adapter.

octo-examples-04-eval-finetuned-on-robot-py | Record version: source-dfaab1cd9a067860d36c8c10ff43f8d80cde29220482b88dba3c40714cb2f89c

Record JSON | Markdown | Pinned upstream source

Access

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

Hub preparation adapter: false. Follow the upstream setup and invocation contract.

Source and license

Repository: https://github.com/octo-models/octo
Revision: 241fb3514b7c40957a86d869fecb7c7fc353f540
Path: examples/04_eval_finetuned_on_robot.py
Source SHA256: dfaab1cd9a067860d36c8c10ff43f8d80cde29220482b88dba3c40714cb2f89c
Original authors: Octo Model Team, Robotic AI & Learning Lab Berkeley
License: MIT
License scope: Selected first-party Python workflow source only. Preserve upstream copyright and per-file notices. Model weights, datasets, third-party code, robot drivers/firmware and services retain separate terms; no rights for those assets are inferred.
License SHA256: 8c891a7dd06fdaa873aa011ff0b2773a3ccfe6d123f6b9a9fc589db718aec1d1
Hub commit: 1d6d9eb134e397d910265a8d6e7d8f257f9e8d7f

Artifact

{
  "kind": "workflow",
  "format": "python",
  "workflow_role": "policy-evaluation"
}

Requirements

{
  "python": "Python3.10 in README setup; controller environment remains separate.",
  "packages": [
    "Pinned Octo checkout",
    "JAX0.4.20, Flax0.7.5 and TensorFlow2.15.0 from requirements.txt",
    "Per-example simulator/robot-controller dependencies and reviewed checkpoints/data"
  ],
  "display": "Hardware: Docstring requires reproducing BridgeData robot setup and installing bridge_data_robot controller. Source creates WidowXClient, initializes the environment and issues movement/actions; exact setup is didactic and may not apply to another robot. Weights: Requires checkpoint_weights_path and checkpoint_step, then loads OctoModel.load_pretrained. Matching model observations/actions are required by the chosen source setup. Datasets: Consumes live robot/camera observations and can save videos; no training dataset fetch is performed by the selected entrypoint. Services/auth: Requires the separate WidowX controller service at configured IP/port. No cloud API key or W&B initialization is declared by this selected file.",
  "execution_domain": "hardware",
  "source_imports": [
    "absl",
    "click",
    "cv2",
    "datetime",
    "envs.widowx_env",
    "functools",
    "imageio",
    "jax",
    "jax.numpy",
    "numpy",
    "octo.model.octo_model",
    "octo.utils.gym_wrappers",
    "octo.utils.train_callbacks",
    "os",
    "time",
    "widowx_envs.widowx_env_service"
  ],
  "notes": "WidowX is a source-declared discovery label. No compatibility combination, adapter support, automatic execution or robot-success claim is created. Code licensing does not establish model, checkpoint, dataset or service permission.",
  "declared_requirements": {
    "python": "Python3.10 in README setup; controller environment remains separate.",
    "weights": "Requires checkpoint_weights_path and checkpoint_step, then loads OctoModel.load_pretrained. Matching model observations/actions are required by the chosen source setup.",
    "datasets": "Consumes live robot/camera observations and can save videos; no training dataset fetch is performed by the selected entrypoint.",
    "hardware": "Docstring requires reproducing BridgeData robot setup and installing bridge_data_robot controller. Source creates WidowXClient, initializes the environment and issues movement/actions; exact setup is didactic and may not apply to another robot.",
    "external_services_and_auth": "Requires the separate WidowX controller service at configured IP/port. No cloud API key or W&B initialization is declared by this selected file.",
    "evidence": [
      {
        "anchor": "WidowXClient(host=FLAGS.ip",
        "line": 92
      },
      {
        "anchor": "OctoModel.load_pretrained",
        "line": 101
      },
      {
        "anchor": "env.step(action)",
        "line": 215
      }
    ],
    "license_scopes": {
      "code": "MIT for selected first-party workflow source.",
      "weights": "README links separately hosted rail-berkeley Octo checkpoints. This bounded repository review did not inspect model cards or establish a weight license; repository MIT is limited to selected code.",
      "datasets": "README references Open X-Embodiment and an example ALOHA simulation dataset. Their individual terms were not inspected and are not covered by the code-license assertion.",
      "third_party_code": "External dependencies retain their own licenses."
    },
    "interpretation": "Pinned upstream declarations only; no dependency resolution, model/data access, compatibility inference or runtime validation was performed."
  }
}

Invocation

{
  "entrypoint": "cd examples && python 04_eval_finetuned_on_robot.py --checkpoint_weights_path=<approved-checkpoint> --checkpoint_step=<step> --ip=<controller-address>",
  "notes": "WidowX is a source-declared discovery label. No compatibility combination, adapter support, automatic execution or robot-success claim is created."
}

Explicit compatibility

[]

Missing environment fields remain unknown. 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.