jaehyunkang/pi05-real-workbench-taco-3view-8b7104f0-object-classification-60k

Pi0.5 Real Workbench — taco-3view-8b7104f0-object-classification

Final checkpoint after 60,000 optimization steps. This is a trained policy, not an evaluation result.

• —Dataset: Myungkyu/real_workbench-taco-keyframe-gemini
• —Task scope: object_classification
• —Instruction: Per-frame parquet subtask
• —Views: 3 (exterior and wrist, plus observation.image.keyframe)
• —Global batch: 32; training GPUs: 2; seed: 42
• —State: 8 dimensions; action: delta EEF 7 dimensions (6 Cartesian velocity + gripper)
• —Image storage: 224×126; policy pads to 224×224
• —Action chunk / execution horizon: 50; inference denoising steps: 10
• —Training implementation: RLWRLD/hiwrld-ll-policy, vendored LeRobot Pi0.5. Custom input fields may require the matching implementation.

Policy weights, policy configuration, preprocessing/postprocessing and normalization states are at the repository root. This inference artifact excludes optimizer and training resume state. Host-specific paths were removed from JSON metadata; supply local dataset/output paths when resuming. The tokenizer reference points to google/paligemma-3b-pt-224 (training revision 35e4f46485b4d07967e7e9935bc3786aad50687c).

For subtask models, supply the corresponding per-frame subtask as the policy task text. For 3-view models also supply the dataset-defined keyframe image. No real-robot evaluation metrics are claimed here.

File sizes and SHA-256 hashes are recorded in artifact_manifest.json. Original training checkpoints were preserved.