| --- |
| tags: |
| - lerobot |
| - robotics |
| - libero |
| - libero-plus |
| pretty_name: LIBERO Plus libero_plus_object (detailed LeRobot v3.0) |
| --- |
| |
| # libero_plus_object: detailed LeRobot v3.0 |
|
|
| This dataset was converted from the LIBERO Plus LeRobot v2.1 `libero_plus_object` partition. |
| The original 8D state and 7D action vectors are preserved exactly as |
| `raw_state.ref_state` and `raw_action.ref_action`. Canonical low-dimensional fields follow |
| `failure_rollout_data/dataset.md`; `debug.gripper_eef_*` contains the ground-truth next-step |
| relative EEF motion for inspection. |
|
|
| ## Required camera transform for canonical training |
|
|
| The source `observation.images.front` and `observation.images.wrist` videos are preserved |
| unchanged. **For canonical training, horizontally flip both camera views at load time.** The |
| videos in this repository are deliberately not rewritten or re-encoded. |
|
|
| The source-pipeline root cause is a composition of two image transforms: robosuite returns a |
| vertically inverted render, and the original dataset writer then rotates it by 180 degrees (flips |
| both image axes). The vertical flips cancel, leaving the stored image horizontally mirrored. This |
| is also tracked in [LeRobot issue #3830](https://github.com/huggingface/lerobot/issues/3830). |
|
|
| For an array whose layout ends in `(height, width, channels)`: |
|
|
| ```python |
| image = np.flip(image, axis=-2) |
| ``` |
|
|
| For a tensor whose width is the last dimension, such as `(..., channels, height, width)`: |
|
|
| ```python |
| image = torch.flip(image, dims=(-1,)) |
| ``` |
|
|
| Apply the image transform only to the camera pixels. Do not flip or negate `raw_state.*`, |
| `raw_target.*`, `state.*`, `target.*`, or `debug.*`; those fields remain proper right-handed |
| coordinate representations. |
|
|
| ## Conversion notes |
|
|
| - No frames were filtered. The historical LIBERO no-op predicate is audit-only; see |
| `meta/noop_audit.json` and `meta/noop_audit_episodes.jsonl`. |
| - Rotation and reconstruction checks are recorded in `meta/conversion_validation.json`. |
| - Alignment and controller-scale assumptions are recorded in `meta/conversion_config.json`. |
| - `meta/stats.json` includes `q01` and `q99` for every numeric and video feature. Numeric |
| quantiles use every frame; video quantiles use a deterministic uniform sample of stored frames. |
|
|
