mcpField
P
physicalai-runtime-loading-exported-policies

physicalai-runtime-loading-exported-policies

github.com/openvinotoolkitDevelopment4080 installsUpdated today
Repository

About physicalai-runtime-loading-exported-policies

A development skill for loading exported policies in physicalai Runtime. It covers local export directories and Hub snapshots, backend selection, artifact validation, and smoke inference.

Key features

  1. 01Load local Studio exports
  2. 02Load Hub snapshots
  3. 03Detect or select inference backends
  4. 04Validate manifests and artifacts
  5. 05Run a smoke inference

Use cases

  1. 01Loading a locally exported policy
  2. 02Testing a policy downloaded from Hub
  3. 03Checking backend and manifest compatibility

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/openvinotoolkit/physicalai --skill physicalai-runtime-loading-exported-policies

After installing, type/physicalai-runtime-loading-exported-policiesin Claude Code to confirm it loaded.

Bundled files

References and scripts installed along with SKILL.md.

references/
export-load-contract.md
onnx.md
openvino.md

Details

From the project README

Loading Exported Policies

Runtime loads Studio export directories (or Hub snapshots that mirror them) through InferenceModel in src/physicalai/inference/model.py. Manifest parsing lives in src/physicalai/inference/manifest.py; backends register in src/physicalai/inference/adapters/registry.py (onnx → .onnx, openvino → .xml). Hub downloads use src/physicalai/inference/utils/_hub.py.

Workflow

  1. Identify the artifact: local export directory or Hub repo_id, expected backend, and whether the user needs select_action vs predict_action_chunk.

    • Done when: load path and API entry point are chosen before editing code.

  2. Load with auto-detection first (local):

    from physicalai.inference import InferenceModel
model = InferenceModel("./exports/act_policy")
    • Done when: model constructs without explicit backend= when artifacts match a registered extension.

  3. Hub load when the package is published:

    model = InferenceModel.from_pretrained("OpenVINO/act-fp16-ov", revision="<commit-sha>")
    • Done when: revision is pinned for reproducibility when security or CI matters.

  4. Explicit backend only when auto-detection is ambiguous:

    model = InferenceModel("./exports/act_policy", backend="openvino", device="CPU")

  5. Validate structure against references/export-load-contract.md and backend notes (references/onnx.md, references/openvino.md).

    • Done when: manifest, model file, and processor artifacts resolve under the export directory.

  6. Smoke inference without owning robot timing:

    model.reset()
action = model.select_action(observation)
    • Done when: one forward pass succeeds on representative observation keys/shapes. For hardware loops, hand off to physicalai-runtime-running-policy-on-robot.

Validation loop

uv run pytest tests/unit/inference/test_model.py tests/unit/inference/test_manifest.py -q

For adapter changes, add or run targeted tests under tests/unit/inference/.

Required checks

  • Export directory contains backend model file(s) and manifest.json consistent with docs/reference/manifest-schema.md.
  • Metadata input/output/feature names align with preprocessors in the manifest.
  • Optional backends fail with clear install guidance (onnxruntime, OpenVINO).
  • Do not document a backend unless an adapter is registered in src/physicalai/inference/adapters/.
  • Hub loads must not log tokens; prefer pinned revision= for production docs.

References

  • references/export-load-contract.md — consumer-side contract (coordinate with Studio export skill).
  • references/onnx.md, references/openvino.md — adapter constraints.

Comments

Markdown supported