1 data_handling.ase_efm

data_handling.ase_efm

Raw ASE/EFM data interfaces used by aria_nbv.data_handling.

This module provides the stable raw snippet layer for the data-handling package. It keeps the existing raw dataset and typed view implementations intact while presenting a single import surface for:

the iterable raw ASE/EFM dataset,
typed raw and VIN snippet views, and
the worker-local snippet loader used for live snippet attachment.

1.1 Package guide

aria_nbv.data_handling.ase_efm streams tensorized ASE snippets from ATEK WebDataset shards and presents them as typed EFM views. It is the live input used by candidate generation, immutable VIN-store construction, and small rollout-generation smoke tests.

ATEK shards carry calibrated observations, trajectory, and MPS semidense points. Optional ASE meshes and GT annotations are attached as oracle/evaluation assets and must not become ordinary actor inputs.

1.2 Read Snippets

from aria_nbv.data_handling.ase_efm import AseEfmDatasetConfig

dataset = AseEfmDatasetConfig(
    scene_ids=["81286"],
    load_meshes=True,
    require_mesh=True,
    device="cpu",
).setup_target()

snippet = next(iter(dataset))
print(snippet.scene_id, snippet.snippet_id)

AseEfmDatasetConfig resolves scene, shard, and sample filters; applies the ATEK-to-EFM taxonomy; and optionally pairs ASE meshes. EfmSnippetView exposes typed camera, trajectory, semidense-point, OBB, and GT surfaces without adding a parallel field schema. EfmSnippetLoader is the worker-local adapter for reattaching a known raw snippet to an immutable store row.

Use the downloader before constructing the dataset:

cd aria_nbv
uv run nbv-downloader -m list -c efm -n 10
uv run nbv-downloader -m download -c efm --ns 1 --max-shards 1

Detailed field, frame, and padding contracts live in the module docstrings and generated API reference.

1.3 Verification

cd aria_nbv
uv run ruff check aria_nbv/data_handling/ase_efm
uv run pytest tests/data_handling/test_efm_dataset_snippet.py \
  tests/data_handling/test_efm_snippet_loader.py

1.4 Classes

Name	Description
AseEfmDataset	Stream zero-copy EFM views from ATEK shards with optional ASE meshes.
AseEfmDatasetConfig	Configure the canonical ATEK-shard to EFM-snippet streaming path.
EfmCameraView	Expose one calibrated Aria camera stream view without copying EFM tensors.
EfmGTView	Parse nested `gt_data/efm_gt` annotations for evaluation-only access.
EfmObbView	Expose snippet-level GT boxes in the EFM `ObbTW` layout.
EfmPointsView	Expose actor-visible MPS semidense reconstruction evidence.
EfmSnippetLoader	Resolve source snippets through worker-owned, per-scene dataset caches.
EfmSnippetView	Own the typed boundary around one adapted ATEK/ASE snippet.
EfmTrajectoryView	Expose the MPS/EFM world-from-rig trajectory for one snippet.

1.4.1 AseEfmDataset

data_handling.ase_efm.AseEfmDataset(config)

Stream zero-copy EFM views from ATEK shards with optional ASE meshes.

Each iteration adapts one WebDataset sample into the consumed EFM key families, infers stable scene/snippet lineage, and attaches processed mesh tensors when configured. The iterable owns per-instance mesh memoization; yielded EfmSnippetView objects retain the backing EFM dictionary. Meshes never become actor-visible observations.

1.4.2 AseEfmDatasetConfig

data_handling.ase_efm.AseEfmDatasetConfig()

Configure the canonical ATEK-shard to EFM-snippet streaming path.

The config resolves scene/shard/sample filters, the ATEK-to-EFM taxonomy, and optional ASE mesh pairing before constructing AseEfmDataset. EfmModelAdaptor owns flattened-key remapping; this config does not introduce a parallel schema.

1.4.2.1 Attributes

Name	Description
cache_exclude_fields	Fields omitted from cache snapshots because they are large or derived.
target_type	Factory target for `BaseConfig.setup_target`.
paths	Project path resolver (data roots, mesh locations, etc.).
atek_variant	ATEK payload variant and on-disk shard subdirectory under the data root.
scene_ids	Optional list of ASE scene IDs to include.
snippet_ids	Optional snippet IDs to include for each scene.
snippet_key_filter	Optional sample key filter applied after loading shards.
tar_urls	ATEK WebDataset shard paths or globs carrying tensorized ASE snippets.
scene_to_mesh	Optional mapping of `scene_id` to ASE GT-mesh evaluation asset.
taxonomy_csv_filename	Filename for the ATEK→EFM taxonomy mapping CSV.
batch_size	WebDataset batch size inside ATEK loader.
wds_shuffle	Enable shuffled sampling in the underlying WebDataset pipeline.
wds_repeat	Repeat the underlying WebDataset stream (infinite iterator).
snippet_length_s	Snippet duration in seconds used by the EFM adaptor.
freq_hz	Sampling frequency (Hz) used by the EFM adaptor.
semidense_points_pad	Pad/clip semi-dense points to this count per snippet frame.
load_meshes	Attach GT meshes to snippets when available.
require_mesh	Raise if a GT mesh is missing for any requested scene.
mesh_simplify_ratio	Fraction of faces to keep when simplifying meshes. `None` disables.
crop_mesh	If `True`, crop meshes to snippet bounds before simplification.
mesh_crop_margin_m	Margin added to semidense bounds when cropping meshes. `None` disables cropping.
cache_meshes	Cache loaded meshes in memory per dataset instance.
mesh_crop_min_keep_ratio	Minimum face keep ratio tolerated during mesh cropping.
device	Device to move snippet tensors onto.
verbosity	Verbosity level for dataset logging.
is_debug	Enable debug logging and extra diagnostics.
taxonomy_csv	Resolve the ATEK-to-EFM semantic taxonomy used by `EfmModelAdaptor`.

1.4.2.2 Methods

Name	Description
setup_target	Instantiate the configured raw EFM dataset.

1.4.2.2.1 setup_target

data_handling.ase_efm.AseEfmDatasetConfig.setup_target()

Instantiate the configured raw EFM dataset.

1.4.3 EfmCameraView

data_handling.ase_efm.EfmCameraView(
    images,
    calib,
    time_ns,
    frame_ids,
    distance_m=None,
    distance_time_ns=None,
)

Expose one calibrated Aria camera stream view without copying EFM tensors.

The view consumes matching entries from ARIA_IMG, ARIA_CALIB, ARIA_IMG_TIME_NS, ARIA_FRAME_ID, and optional ARIA_DISTANCE_M/ARIA_DEPTH_TIME_NS families. VRS and online calibration are upstream provenance; runtime access is through ATEK/EFM.

1.4.3.1 Attributes

Name	Description
images	`Tensor["F C H W", float32]` normalized RGB or monochrome frames.
calib	Per-frame `CameraTW` intrinsics/extrinsics; backing shape is `(F, 34)`.
time_ns	`Tensor["F", int64]` Aria device-clock timestamps aligned to `images`.
frame_ids	`Tensor["F", int64\\|float32]` frame ids within the snippet.
distance_m	Optional `Tensor["F 1 H W", float32]` per-pixel ray distances in metres.
distance_time_ns	Optional `Tensor["F", int64]` timestamps for `distance_m`.
num_frames	Return the number of frames in the camera stream.

1.4.3.2 Methods

Name	Description
to	Move the camera view tensors to the requested device and dtype.
get_fov	Return `Tensor["F 2", float32]` horizontal/vertical FOV in degrees.
select_frame_indices	Resolve user-provided frame indices, supporting negatives and defaults.
nearest_traj_indices	Return selected camera indices and nearest trajectory indices.

1.4.3.2.1 to

data_handling.ase_efm.EfmCameraView.to(device, *, dtype=None)

Move the camera view tensors to the requested device and dtype.

1.4.3.2.2 get_fov

data_handling.ase_efm.EfmCameraView.get_fov()

Return Tensor["F 2", float32] horizontal/vertical FOV in degrees.

1.4.3.2.3 select_frame_indices

data_handling.ase_efm.EfmCameraView.select_frame_indices(
    frame_indices,
    *,
    default_last,
)

Resolve user-provided frame indices, supporting negatives and defaults.

1.4.3.2.4 nearest_traj_indices

data_handling.ase_efm.EfmCameraView.nearest_traj_indices(
    traj_ts_ns,
    frame_indices=None,
    *,
    default_last=True,
)

Return selected camera indices and nearest trajectory indices.

1.4.4 EfmGTView

data_handling.ase_efm.EfmGTView(raw)

Parse nested gt_data/efm_gt annotations for evaluation-only access.

1.4.4.1 Attributes

Name	Description
raw	Zero-copy backing GT payload from the adapted ATEK sample.
efm_gt	Timestamp-indexed per-camera GT views; never actor-visible observations.
timestamps	Return sorted string keys for available `efm_gt` timestamps.

1.4.4.2 Methods

Name	Description
cameras_at	Return evaluation-only per-camera GT OBBs at one timestamp.

1.4.4.2.1 cameras_at

data_handling.ase_efm.EfmGTView.cameras_at(ts)

Return evaluation-only per-camera GT OBBs at one timestamp.

1.4.4.2.1.1 Parameters

Name	Type	Description	Default
ts	str \| int	Timestamp key accepted as text or an integer.	required

1.4.4.2.1.2 Returns

Name	Type	Description
	dict[str, EfmGtCameraObbView]	Camera-name mapping for the requested GT slice.

1.4.5 EfmObbView

data_handling.ase_efm.EfmObbView(obbs, hz=None)

Expose snippet-level GT boxes in the EFM ObbTW layout.

obbs may be shaped (T, K, 34), (1, K, 34), or (K, 34) depending on the source. K is padded; padded rows are all PAD_VAL and must be removed with ObbTW.get_padding_mask() before matching label targets.

The 34 payload columns are the EFM OBB contract: object-frame 3D bounds, three 2D boxes, a 12-value pose, semantic id, instance id, confidence, and movable flag. In EfmSnippetView these boxes originate from ARIA_OBB_PADDED and are GT label/evaluation data. Actor-visible target hypotheses must come from EVL predictions or tracking instead. Downstream target selection transforms the latest valid slice into world coordinates before constructing Oracle target tasks.

1.4.5.1 Attributes

Name	Description
obbs	`Tensor["... K 34", float32]` GT boxes in the source EFM frame.
hz	Optional `Tensor["1", int32]` detection frequency.

1.4.6 EfmPointsView

data_handling.ase_efm.EfmPointsView(
    points_world,
    dist_std,
    inv_dist_std,
    time_ns,
    volume_min,
    volume_max,
    lengths,
)

Expose actor-visible MPS semidense reconstruction evidence.

The ARIA_POINTS_* tensors are padded per frame; lengths defines the valid prefix and non-finite rows remain invalid. Positions, uncertainties, and volume bounds preserve the EFM world frame and metric units.

1.4.6.1 Attributes

Name	Description
points_world	`Tensor["F N_pad 3", float32]` padded world-frame MPS points in metres.
dist_std	`Tensor["F N_pad", float32]` per-point distance standard deviation in metres.
inv_dist_std	`Tensor["F N_pad", float32]` inverse distance uncertainty in metres⁻¹.
time_ns	`Tensor["F", int64]` timestamps aligned to point slices.
volume_min	`Tensor["3", float32]` snippet AABB minimum in world frame, metres.
volume_max	`Tensor["3", float32]` snippet AABB maximum in world frame, metres.
lengths	`Tensor["F", int64]` valid point-prefix length for each padded frame.

1.4.6.2 Methods

Name	Description
to	Move the semidense point tensors to the requested device and dtype.
collapse_points	Collapse points across time and optionally subsample.
last_frame_points_np	Return points from the latest frame with valid points, subsampled if needed.

1.4.6.2.1 to

data_handling.ase_efm.EfmPointsView.to(device, *, dtype=None)

Move the semidense point tensors to the requested device and dtype.

1.4.6.2.2 collapse_points

data_handling.ase_efm.EfmPointsView.collapse_points(
    max_points=None,
    include_inv_dist_std=False,
    include_obs_count=False,
)

Collapse points across time and optionally subsample.

1.4.6.2.2.1 Parameters

Name	Type	Description	Default
max_points	int \| None	Optional cap on the number of returned points.	`None`
include_inv_dist_std	bool	If True, append inverse depth std per point.	`False`
include_obs_count	bool	If True, append the number of snippet frames that observed each point (observation count).	`False`

1.4.6.2.2.2 Returns

Name	Type	Description
	Tensor	`Tensor["K 3", float32]` if no extra features are requested.
	Tensor	`Tensor["K 4", float32]` if exactly one extra feature is requested
	Tensor	(XYZ + inv_dist_std or XYZ + obs_count).
	Tensor	`Tensor["K 5", float32]` if both extras are requested
	Tensor	(XYZ + inv_dist_std + obs_count).

1.4.6.2.3 last_frame_points_np

data_handling.ase_efm.EfmPointsView.last_frame_points_np(max_points=None)

Return points from the latest frame with valid points, subsampled if needed.

1.4.7 EfmSnippetLoader

data_handling.ase_efm.EfmSnippetLoader(
    dataset_payload,
    device,
    paths,
    include_gt_mesh,
)

Resolve source snippets through worker-owned, per-scene dataset caches.

Instances are intentionally not a persisted store component. They reopen ATEK WebDataset shards when a reader requests raw camera, MPS, calibration, or optional GT-mesh context that was not duplicated into offline blocks.

1.4.7.1 Methods

Name	Description
load	Load one source snippet, rebuilding a stale scene iterator once.

1.4.7.1.1 load

data_handling.ase_efm.EfmSnippetLoader.load(scene_id, snippet_id)

Load one source snippet, rebuilding a stale scene iterator once.

1.4.7.1.1.1 Parameters

Name	Type	Description	Default
scene_id	str	ASE scene identifier used to locate ATEK shards.	required
snippet_id	str	Compact or raw ATEK sample identifier to match.	required

1.4.7.1.1.2 Returns

Name	Type	Description
Live	EfmSnippetView	class:`EfmSnippetView` backed by the source shard.

1.4.8 EfmSnippetView

data_handling.ase_efm.EfmSnippetView(
    efm,
    scene_id,
    snippet_id,
    mesh=None,
    crop_bounds=None,
    mesh_verts=None,
    mesh_faces=None,
    mesh_cache_key=None,
    mesh_specs=None,
)

Own the typed boundary around one adapted ATEK/ASE snippet.

efm retains the zero-copy key/value payload produced by EfmModelAdaptor. Camera/calibration, world-from-rig trajectory, and MPS semidense fields are actor-visible observations. Attached ASE meshes, ARIA_OBB_PADDED, and nested gt_data are oracle/evaluation assets. Mesh tensors are optional CPU/cache products and are not moved into the EFM dictionary.

1.4.8.1 Attributes

Name	Description
efm	Zero-copy adapted ATEK payload containing the consumed `ARIA_*` key families.
scene_id	ASE scene identifier used for shard lineage and GT-mesh pairing.
snippet_id	Compact ATEK sample identifier derived from the WebDataset sample key.
mesh	Optional ASE GT mesh used only by oracle labeling and evaluation.
crop_bounds	Optional pair of `Tensor["3", float32]` world-frame bounds in metres.
mesh_verts	Optional `Tensor["V 3", float32]` GT-mesh vertices in world frame, metres.
mesh_faces	Optional `Tensor["F_mesh 3", int64]` triangle vertex indices.
mesh_cache_key	Stable key (spec hash) for shared mesh caches across components.
mesh_specs	Optional mesh processing specification used to derive attached mesh tensors.
camera_rgb	RGB stream (fisheye624, 240x240 in ASE preprocessing).
camera_slam_left	Left SLAM mono stream (fisheye624, ~320x240).
camera_slam_right	Right SLAM mono stream (fisheye624, ~320x240).
trajectory	Rig trajectory (world←rig) aligned to snippet frames.
semidense	Semi-dense SLAM points (padded to fixed length).
obbs	Return padded ASE GT OBBs for label/evaluation use when present.
gt	Ground-truth dict `gt_data` (EFM OBBs per timestamp/camera).
has_mesh	Return whether a ground-truth mesh is attached to the snippet.

1.4.8.2 Methods

Name	Description
from_cache_efm	Construct a snippet view from an offline-cache EFM dict.
get_camera	Return the requested camera stream view from the backing EFM dict.
get_occupancy_extend	Return `[xmin, xmax, ymin, ymax, zmin, zmax]` AABB in world frame.
to	Return a shallow copy with EFM tensors and wrappers moved to a device.
prune_efm	Return a view with the EFM dict pruned to the requested keys.

1.4.8.2.1 from_cache_efm

data_handling.ase_efm.EfmSnippetView.from_cache_efm(
    efm,
    *,
    mesh=None,
    crop_bounds=None,
    mesh_verts=None,
    mesh_faces=None,
    mesh_cache_key=None,
    mesh_specs=None,
)

Construct a snippet view from an offline-cache EFM dict.

1.4.8.2.1.1 Parameters

Name	Type	Description	Default
efm	dict[str, Any]	Raw cache sample dict that includes `"__key__"`.	required
mesh	`Trimesh` \| None	Optional GT mesh to attach.	`None`
crop_bounds	tuple[torch.Tensor, torch.Tensor] \| None	Optional world-space AABB override.	`None`
mesh_verts	torch.Tensor \| None	Optional cached mesh vertices.	`None`
mesh_faces	torch.Tensor \| None	Optional cached mesh faces.	`None`
mesh_cache_key	str \| None	Optional mesh cache key for shared caches.	`None`
mesh_specs	MeshProcessSpec \| None	Optional mesh processing spec.	`None`

1.4.8.2.1.2 Returns

Name	Type	Description
	'EfmSnippetView'	Parsed `EfmSnippetView` instance.

1.4.8.2.2 get_camera

data_handling.ase_efm.EfmSnippetView.get_camera(prefix)

Return the requested camera stream view from the backing EFM dict.

1.4.8.2.3 get_occupancy_extend

data_handling.ase_efm.EfmSnippetView.get_occupancy_extend()

Return [xmin, xmax, ymin, ymax, zmin, zmax] AABB in world frame.

1.4.8.2.4 to

data_handling.ase_efm.EfmSnippetView.to(device, *, dtype=None)

Return a shallow copy with EFM tensors and wrappers moved to a device.

The backing dictionary is rebuilt without mutating this view. Plain tensors honor dtype; EFM wrappers preserve their own dtype contract. The CPU Trimesh object is shared, while cached mesh tensors and world-frame crop bounds are moved.

1.4.8.2.5 prune_efm

data_handling.ase_efm.EfmSnippetView.prune_efm(keep_keys, *, keep_prefixes=())

Return a view with the EFM dict pruned to the requested keys.

1.4.8.2.5.1 Parameters

Name	Type	Description	Default
keep_keys	set[str] \| None	Exact EFM keys to retain. `None` returns `self` unchanged.	required
keep_prefixes	tuple[str, …]	Optional prefixes; keys starting with any prefix are retained.	`()`

1.4.9 EfmTrajectoryView

data_handling.ase_efm.EfmTrajectoryView(t_world_rig, time_ns, gravity_in_world)

Expose the MPS/EFM world-from-rig trajectory for one snippet.

1.4.9.1 Attributes

Name	Description
t_world_rig	`PoseTW["F 12"]` world←rig transforms; translation is metres.
time_ns	`Tensor["F", int64]` pose timestamps.
gravity_in_world	`Tensor["3", float32]` world-frame gravity in metres per second squared.
final_pose	Return the final `PoseTW["12"]` world←rig pose in the snippet.

1.4.9.2 Methods

Name	Description
to	Move the trajectory tensors to the requested device and dtype.

1.4.9.2.1 to

data_handling.ase_efm.EfmTrajectoryView.to(device, *, dtype=None)

Move the trajectory tensors to the requested device and dtype.

1.5 Functions

Name	Description
infer_semidense_bounds	Infer snippet world-space AABB from semidense metadata or points.
is_efm_snippet_view_instance	Return whether `value` behaves like an `EfmSnippetView`.

1.5.1 infer_semidense_bounds

data_handling.ase_efm.infer_semidense_bounds(efm_dict)

Infer snippet world-space AABB from semidense metadata or points.

Metadata bounds and finite-point bounds are compared when both exist; the smaller finite volume is retained to avoid an unnecessarily loose crop. Padded and non-finite semidense rows never contribute.

1.5.1.1 Parameters

Name	Type	Description	Default
efm_dict	Mapping[str, Any]	Adapted EFM payload containing world-frame semidense points and/or explicit volume bounds.	required

1.5.1.2 Returns

Name	Type	Description
	tuple[torch.Tensor, torch.Tensor] \| None	Pair of `Tensor["3", float32]` world-frame AABB minima and maxima in
	tuple[torch.Tensor, torch.Tensor] \| None	metres on CPU, or `None` when no finite bounds are available.

1.5.2 is_efm_snippet_view_instance

data_handling.ase_efm.is_efm_snippet_view_instance(value)

Return whether value behaves like an EfmSnippetView.

The v2 stack accepts both the local data-handling view classes and legacy view objects that expose the same public attributes.

# data_handling.ase_efm { #aria_nbv.data_handling.ase_efm } `data_handling.ase_efm` Raw ASE/EFM data interfaces used by ``aria_nbv.data_handling``. This module provides the stable raw snippet layer for the data-handling package. It keeps the existing raw dataset and typed view implementations intact while presenting a single import surface for: - the iterable raw ASE/EFM dataset, - typed raw and VIN snippet views, and - the worker-local snippet loader used for live snippet attachment.  ## Package guide `aria_nbv.data_handling.ase_efm` streams tensorized ASE snippets from ATEK WebDataset shards and presents them as typed EFM views. It is the live input used by candidate generation, immutable VIN-store construction, and small rollout-generation smoke tests. ATEK shards carry calibrated observations, trajectory, and MPS semidense points. Optional ASE meshes and GT annotations are attached as oracle/evaluation assets and must not become ordinary actor inputs. ## Read Snippets ```python from aria_nbv.data_handling.ase_efm import AseEfmDatasetConfig dataset = AseEfmDatasetConfig( scene_ids=["81286"], load_meshes=True, require_mesh=True, device="cpu", ).setup_target() snippet = next(iter(dataset)) print(snippet.scene_id, snippet.snippet_id) ``` `AseEfmDatasetConfig` resolves scene, shard, and sample filters; applies the ATEK-to-EFM taxonomy; and optionally pairs ASE meshes. `EfmSnippetView` exposes typed camera, trajectory, semidense-point, OBB, and GT surfaces without adding a parallel field schema. `EfmSnippetLoader` is the worker-local adapter for reattaching a known raw snippet to an immutable store row. Use the downloader before constructing the dataset: ```sh cd aria_nbv uv run nbv-downloader -m list -c efm -n 10 uv run nbv-downloader -m download -c efm --ns 1 --max-shards 1 ``` Detailed field, frame, and padding contracts live in the module docstrings and [generated API reference](index.qmd). ## Verification ```sh cd aria_nbv uv run ruff check aria_nbv/data_handling/ase_efm uv run pytest tests/data_handling/test_efm_dataset_snippet.py \ tests/data_handling/test_efm_snippet_loader.py ```  ## Classes | Name | Description | | --- | --- | | [AseEfmDataset](#aria_nbv.data_handling.ase_efm.AseEfmDataset) | Stream zero-copy EFM views from ATEK shards with optional ASE meshes. | | [AseEfmDatasetConfig](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig) | Configure the canonical ATEK-shard to EFM-snippet streaming path. | | [EfmCameraView](#aria_nbv.data_handling.ase_efm.EfmCameraView) | Expose one calibrated Aria camera stream view without copying EFM tensors. | | [EfmGTView](#aria_nbv.data_handling.ase_efm.EfmGTView) | Parse nested ``gt_data/efm_gt`` annotations for evaluation-only access. | | [EfmObbView](#aria_nbv.data_handling.ase_efm.EfmObbView) | Expose snippet-level GT boxes in the EFM ``ObbTW`` layout. | | [EfmPointsView](#aria_nbv.data_handling.ase_efm.EfmPointsView) | Expose actor-visible MPS semidense reconstruction evidence. | | [EfmSnippetLoader](#aria_nbv.data_handling.ase_efm.EfmSnippetLoader) | Resolve source snippets through worker-owned, per-scene dataset caches. | | [EfmSnippetView](#aria_nbv.data_handling.ase_efm.EfmSnippetView) | Own the typed boundary around one adapted ATEK/ASE snippet. | | [EfmTrajectoryView](#aria_nbv.data_handling.ase_efm.EfmTrajectoryView) | Expose the MPS/EFM world-from-rig trajectory for one snippet. | ### AseEfmDataset { #aria_nbv.data_handling.ase_efm.AseEfmDataset } ```python data_handling.ase_efm.AseEfmDataset(config) ``` Stream zero-copy EFM views from ATEK shards with optional ASE meshes. Each iteration adapts one WebDataset sample into the consumed EFM key families, infers stable scene/snippet lineage, and attaches processed mesh tensors when configured. The iterable owns per-instance mesh memoization; yielded :class:`EfmSnippetView` objects retain the backing EFM dictionary. Meshes never become actor-visible observations. ### AseEfmDatasetConfig { #aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig } ```python data_handling.ase_efm.AseEfmDatasetConfig() ``` Configure the canonical ATEK-shard to EFM-snippet streaming path. The config resolves scene/shard/sample filters, the ATEK-to-EFM taxonomy, and optional ASE mesh pairing before constructing :class:`AseEfmDataset`. ``EfmModelAdaptor`` owns flattened-key remapping; this config does not introduce a parallel schema. #### Attributes | Name | Description | | --- | --- | | [cache_exclude_fields](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.cache_exclude_fields) | Fields omitted from cache snapshots because they are large or derived. | | [target_type](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.target_type) | Factory target for `BaseConfig.setup_target`. | | [paths](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.paths) | Project path resolver (data roots, mesh locations, etc.). | | [atek_variant](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.atek_variant) | ATEK payload variant and on-disk shard subdirectory under the data root. | | [scene_ids](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.scene_ids) | Optional list of ASE scene IDs to include. | | [snippet_ids](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.snippet_ids) | Optional snippet IDs to include for each scene. | | [snippet_key_filter](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.snippet_key_filter) | Optional sample key filter applied after loading shards. | | [tar_urls](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.tar_urls) | ATEK WebDataset shard paths or globs carrying tensorized ASE snippets. | | [scene_to_mesh](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.scene_to_mesh) | Optional mapping of ``scene_id`` to ASE GT-mesh evaluation asset. | | [taxonomy_csv_filename](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.taxonomy_csv_filename) | Filename for the ATEK→EFM taxonomy mapping CSV. | | [batch_size](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.batch_size) | WebDataset batch size inside ATEK loader. | | [wds_shuffle](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.wds_shuffle) | Enable shuffled sampling in the underlying WebDataset pipeline. | | [wds_repeat](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.wds_repeat) | Repeat the underlying WebDataset stream (infinite iterator). | | [snippet_length_s](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.snippet_length_s) | Snippet duration in seconds used by the EFM adaptor. | | [freq_hz](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.freq_hz) | Sampling frequency (Hz) used by the EFM adaptor. | | [semidense_points_pad](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.semidense_points_pad) | Pad/clip semi-dense points to this count per snippet frame. | | [load_meshes](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.load_meshes) | Attach GT meshes to snippets when available. | | [require_mesh](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.require_mesh) | Raise if a GT mesh is missing for any requested scene. | | [mesh_simplify_ratio](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.mesh_simplify_ratio) | Fraction of faces to keep when simplifying meshes. ``None`` disables. | | [crop_mesh](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.crop_mesh) | If ``True``, crop meshes to snippet bounds before simplification. | | [mesh_crop_margin_m](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.mesh_crop_margin_m) | Margin added to semidense bounds when cropping meshes. ``None`` disables cropping. | | [cache_meshes](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.cache_meshes) | Cache loaded meshes in memory per dataset instance. | | [mesh_crop_min_keep_ratio](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.mesh_crop_min_keep_ratio) | Minimum face keep ratio tolerated during mesh cropping. | | [device](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.device) | Device to move snippet tensors onto. | | [verbosity](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.verbosity) | Verbosity level for dataset logging. | | [is_debug](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.is_debug) | Enable debug logging and extra diagnostics. | | [taxonomy_csv](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.taxonomy_csv) | Resolve the ATEK-to-EFM semantic taxonomy used by ``EfmModelAdaptor``. | #### Methods | Name | Description | | --- | --- | | [setup_target](#aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.setup_target) | Instantiate the configured raw EFM dataset. | ##### setup_target { #aria_nbv.data_handling.ase_efm.AseEfmDatasetConfig.setup_target } ```python data_handling.ase_efm.AseEfmDatasetConfig.setup_target() ``` Instantiate the configured raw EFM dataset. ### EfmCameraView { #aria_nbv.data_handling.ase_efm.EfmCameraView } ```python data_handling.ase_efm.EfmCameraView( images, calib, time_ns, frame_ids, distance_m=None, distance_time_ns=None, ) ``` Expose one calibrated Aria camera stream view without copying EFM tensors. The view consumes matching entries from ``ARIA_IMG``, ``ARIA_CALIB``, ``ARIA_IMG_TIME_NS``, ``ARIA_FRAME_ID``, and optional ``ARIA_DISTANCE_M``/``ARIA_DEPTH_TIME_NS`` families. VRS and online calibration are upstream provenance; runtime access is through ATEK/EFM. #### Attributes | Name | Description | | --- | --- | | [images](#aria_nbv.data_handling.ase_efm.EfmCameraView.images) | ``Tensor["F C H W", float32]`` normalized RGB or monochrome frames. | | [calib](#aria_nbv.data_handling.ase_efm.EfmCameraView.calib) | Per-frame ``CameraTW`` intrinsics/extrinsics; backing shape is ``(F, 34)``. | | [time_ns](#aria_nbv.data_handling.ase_efm.EfmCameraView.time_ns) | ``Tensor["F", int64]`` Aria device-clock timestamps aligned to ``images``. | | [frame_ids](#aria_nbv.data_handling.ase_efm.EfmCameraView.frame_ids) | ``Tensor["F", int64\|float32]`` frame ids within the snippet. | | [distance_m](#aria_nbv.data_handling.ase_efm.EfmCameraView.distance_m) | Optional ``Tensor["F 1 H W", float32]`` per-pixel ray distances in metres. | | [distance_time_ns](#aria_nbv.data_handling.ase_efm.EfmCameraView.distance_time_ns) | Optional ``Tensor["F", int64]`` timestamps for ``distance_m``. | | [num_frames](#aria_nbv.data_handling.ase_efm.EfmCameraView.num_frames) | Return the number of frames in the camera stream. | #### Methods | Name | Description | | --- | --- | | [to](#aria_nbv.data_handling.ase_efm.EfmCameraView.to) | Move the camera view tensors to the requested device and dtype. | | [get_fov](#aria_nbv.data_handling.ase_efm.EfmCameraView.get_fov) | Return ``Tensor["F 2", float32]`` horizontal/vertical FOV in degrees. | | [select_frame_indices](#aria_nbv.data_handling.ase_efm.EfmCameraView.select_frame_indices) | Resolve user-provided frame indices, supporting negatives and defaults. | | [nearest_traj_indices](#aria_nbv.data_handling.ase_efm.EfmCameraView.nearest_traj_indices) | Return selected camera indices and nearest trajectory indices. | ##### to { #aria_nbv.data_handling.ase_efm.EfmCameraView.to } ```python data_handling.ase_efm.EfmCameraView.to(device, *, dtype=None) ``` Move the camera view tensors to the requested device and dtype. ##### get_fov { #aria_nbv.data_handling.ase_efm.EfmCameraView.get_fov } ```python data_handling.ase_efm.EfmCameraView.get_fov() ``` Return ``Tensor["F 2", float32]`` horizontal/vertical FOV in degrees. ##### select_frame_indices { #aria_nbv.data_handling.ase_efm.EfmCameraView.select_frame_indices } ```python data_handling.ase_efm.EfmCameraView.select_frame_indices( frame_indices, *, default_last, ) ``` Resolve user-provided frame indices, supporting negatives and defaults. ##### nearest_traj_indices { #aria_nbv.data_handling.ase_efm.EfmCameraView.nearest_traj_indices } ```python data_handling.ase_efm.EfmCameraView.nearest_traj_indices( traj_ts_ns, frame_indices=None, *, default_last=True, ) ``` Return selected camera indices and nearest trajectory indices. ### EfmGTView { #aria_nbv.data_handling.ase_efm.EfmGTView } ```python data_handling.ase_efm.EfmGTView(raw) ``` Parse nested ``gt_data/efm_gt`` annotations for evaluation-only access. #### Attributes | Name | Description | | --- | --- | | [raw](#aria_nbv.data_handling.ase_efm.EfmGTView.raw) | Zero-copy backing GT payload from the adapted ATEK sample. | | [efm_gt](#aria_nbv.data_handling.ase_efm.EfmGTView.efm_gt) | Timestamp-indexed per-camera GT views; never actor-visible observations. | | [timestamps](#aria_nbv.data_handling.ase_efm.EfmGTView.timestamps) | Return sorted string keys for available ``efm_gt`` timestamps. | #### Methods | Name | Description | | --- | --- | | [cameras_at](#aria_nbv.data_handling.ase_efm.EfmGTView.cameras_at) | Return evaluation-only per-camera GT OBBs at one timestamp. | ##### cameras_at { #aria_nbv.data_handling.ase_efm.EfmGTView.cameras_at } ```python data_handling.ase_efm.EfmGTView.cameras_at(ts) ``` Return evaluation-only per-camera GT OBBs at one timestamp. ###### Parameters {.doc-section .doc-section-parameters} | Name | Type | Description | Default | |--------|------------------------------|-----------------------------------------------|------------| | ts | [str](`str`) \| [int](`int`) | Timestamp key accepted as text or an integer. | _required_ | ###### Returns {.doc-section .doc-section-returns} | Name | Type | Description | |--------|-----------------------------------------------------------------------------------------------------------------|-------------------------------------------------| | | [dict](`dict`)\[[str](`str`), [EfmGtCameraObbView](`aria_nbv.data_handling.ase_efm.views.EfmGtCameraObbView`)\] | Camera-name mapping for the requested GT slice. | ### EfmObbView { #aria_nbv.data_handling.ase_efm.EfmObbView } ```python data_handling.ase_efm.EfmObbView(obbs, hz=None) ``` Expose snippet-level GT boxes in the EFM ``ObbTW`` layout. `obbs` may be shaped `(T, K, 34)`, `(1, K, 34)`, or `(K, 34)` depending on the source. ``K`` is padded; padded rows are all ``PAD_VAL`` and must be removed with ``ObbTW.get_padding_mask()`` before matching label targets. The `34` payload columns are the EFM OBB contract: object-frame 3D bounds, three 2D boxes, a 12-value pose, semantic id, instance id, confidence, and movable flag. In :class:`EfmSnippetView` these boxes originate from ``ARIA_OBB_PADDED`` and are GT label/evaluation data. Actor-visible target hypotheses must come from EVL predictions or tracking instead. Downstream target selection transforms the latest valid slice into world coordinates before constructing Oracle target tasks. #### Attributes | Name | Description | | --- | --- | | [obbs](#aria_nbv.data_handling.ase_efm.EfmObbView.obbs) | ``Tensor["... K 34", float32]`` GT boxes in the source EFM frame. | | [hz](#aria_nbv.data_handling.ase_efm.EfmObbView.hz) | Optional ``Tensor["1", int32]`` detection frequency. | ### EfmPointsView { #aria_nbv.data_handling.ase_efm.EfmPointsView } ```python data_handling.ase_efm.EfmPointsView( points_world, dist_std, inv_dist_std, time_ns, volume_min, volume_max, lengths, ) ``` Expose actor-visible MPS semidense reconstruction evidence. The ``ARIA_POINTS_*`` tensors are padded per frame; ``lengths`` defines the valid prefix and non-finite rows remain invalid. Positions, uncertainties, and volume bounds preserve the EFM world frame and metric units. #### Attributes | Name | Description | | --- | --- | | [points_world](#aria_nbv.data_handling.ase_efm.EfmPointsView.points_world) | ``Tensor["F N_pad 3", float32]`` padded world-frame MPS points in metres. | | [dist_std](#aria_nbv.data_handling.ase_efm.EfmPointsView.dist_std) | ``Tensor["F N_pad", float32]`` per-point distance standard deviation in metres. | | [inv_dist_std](#aria_nbv.data_handling.ase_efm.EfmPointsView.inv_dist_std) | ``Tensor["F N_pad", float32]`` inverse distance uncertainty in metres⁻¹. | | [time_ns](#aria_nbv.data_handling.ase_efm.EfmPointsView.time_ns) | ``Tensor["F", int64]`` timestamps aligned to point slices. | | [volume_min](#aria_nbv.data_handling.ase_efm.EfmPointsView.volume_min) | ``Tensor["3", float32]`` snippet AABB minimum in world frame, metres. | | [volume_max](#aria_nbv.data_handling.ase_efm.EfmPointsView.volume_max) | ``Tensor["3", float32]`` snippet AABB maximum in world frame, metres. | | [lengths](#aria_nbv.data_handling.ase_efm.EfmPointsView.lengths) | ``Tensor["F", int64]`` valid point-prefix length for each padded frame. | #### Methods | Name | Description | | --- | --- | | [to](#aria_nbv.data_handling.ase_efm.EfmPointsView.to) | Move the semidense point tensors to the requested device and dtype. | | [collapse_points](#aria_nbv.data_handling.ase_efm.EfmPointsView.collapse_points) | Collapse points across time and optionally subsample. | | [last_frame_points_np](#aria_nbv.data_handling.ase_efm.EfmPointsView.last_frame_points_np) | Return points from the latest frame with valid points, subsampled if needed. | ##### to { #aria_nbv.data_handling.ase_efm.EfmPointsView.to } ```python data_handling.ase_efm.EfmPointsView.to(device, *, dtype=None) ``` Move the semidense point tensors to the requested device and dtype. ##### collapse_points { #aria_nbv.data_handling.ase_efm.EfmPointsView.collapse_points } ```python data_handling.ase_efm.EfmPointsView.collapse_points( max_points=None, include_inv_dist_std=False, include_obs_count=False, ) ``` Collapse points across time and optionally subsample. ###### Parameters {.doc-section .doc-section-parameters} | Name | Type | Description | Default | |----------------------|----------------------|--------------------------------------------------------------------------------------------|-----------| | max_points | [int](`int`) \| None | Optional cap on the number of returned points. | `None` | | include_inv_dist_std | [bool](`bool`) | If True, append inverse depth std per point. | `False` | | include_obs_count | [bool](`bool`) | If True, append the number of snippet frames that observed each point (observation count). | `False` | ###### Returns {.doc-section .doc-section-returns} | Name | Type | Description | |--------|--------------------------|----------------------------------------------------------------------| | | [Tensor](`torch.Tensor`) | ``Tensor["K 3", float32]`` if no extra features are requested. | | | [Tensor](`torch.Tensor`) | ``Tensor["K 4", float32]`` if exactly one extra feature is requested | | | [Tensor](`torch.Tensor`) | (XYZ + inv_dist_std or XYZ + obs_count). | | | [Tensor](`torch.Tensor`) | ``Tensor["K 5", float32]`` if both extras are requested | | | [Tensor](`torch.Tensor`) | (XYZ + inv_dist_std + obs_count). | ##### last_frame_points_np { #aria_nbv.data_handling.ase_efm.EfmPointsView.last_frame_points_np } ```python data_handling.ase_efm.EfmPointsView.last_frame_points_np(max_points=None) ``` Return points from the latest frame with valid points, subsampled if needed. ### EfmSnippetLoader { #aria_nbv.data_handling.ase_efm.EfmSnippetLoader } ```python data_handling.ase_efm.EfmSnippetLoader( dataset_payload, device, paths, include_gt_mesh, ) ``` Resolve source snippets through worker-owned, per-scene dataset caches. Instances are intentionally not a persisted store component. They reopen ATEK WebDataset shards when a reader requests raw camera, MPS, calibration, or optional GT-mesh context that was not duplicated into offline blocks. #### Methods | Name | Description | | --- | --- | | [load](#aria_nbv.data_handling.ase_efm.EfmSnippetLoader.load) | Load one source snippet, rebuilding a stale scene iterator once. | ##### load { #aria_nbv.data_handling.ase_efm.EfmSnippetLoader.load } ```python data_handling.ase_efm.EfmSnippetLoader.load(scene_id, snippet_id) ``` Load one source snippet, rebuilding a stale scene iterator once. ###### Parameters {.doc-section .doc-section-parameters} | Name | Type | Description | Default | |------------|--------------|--------------------------------------------------|------------| | scene_id | [str](`str`) | ASE scene identifier used to locate ATEK shards. | _required_ | | snippet_id | [str](`str`) | Compact or raw ATEK sample identifier to match. | _required_ | ###### Returns {.doc-section .doc-section-returns} | Name | Type | Description | |--------|-------------------------------------------------------------------------|----------------------------------------------------| | Live | [EfmSnippetView](`aria_nbv.data_handling.ase_efm.views.EfmSnippetView`) | class:`EfmSnippetView` backed by the source shard. | ### EfmSnippetView { #aria_nbv.data_handling.ase_efm.EfmSnippetView } ```python data_handling.ase_efm.EfmSnippetView( efm, scene_id, snippet_id, mesh=None, crop_bounds=None, mesh_verts=None, mesh_faces=None, mesh_cache_key=None, mesh_specs=None, ) ``` Own the typed boundary around one adapted ATEK/ASE snippet. ``efm`` retains the zero-copy key/value payload produced by ``EfmModelAdaptor``. Camera/calibration, world-from-rig trajectory, and MPS semidense fields are actor-visible observations. Attached ASE meshes, ``ARIA_OBB_PADDED``, and nested ``gt_data`` are oracle/evaluation assets. Mesh tensors are optional CPU/cache products and are not moved into the EFM dictionary. #### Attributes | Name | Description | | --- | --- | | [efm](#aria_nbv.data_handling.ase_efm.EfmSnippetView.efm) | Zero-copy adapted ATEK payload containing the consumed ``ARIA_*`` key families. | | [scene_id](#aria_nbv.data_handling.ase_efm.EfmSnippetView.scene_id) | ASE scene identifier used for shard lineage and GT-mesh pairing. | | [snippet_id](#aria_nbv.data_handling.ase_efm.EfmSnippetView.snippet_id) | Compact ATEK sample identifier derived from the WebDataset sample key. | | [mesh](#aria_nbv.data_handling.ase_efm.EfmSnippetView.mesh) | Optional ASE GT mesh used only by oracle labeling and evaluation. | | [crop_bounds](#aria_nbv.data_handling.ase_efm.EfmSnippetView.crop_bounds) | Optional pair of ``Tensor["3", float32]`` world-frame bounds in metres. | | [mesh_verts](#aria_nbv.data_handling.ase_efm.EfmSnippetView.mesh_verts) | Optional ``Tensor["V 3", float32]`` GT-mesh vertices in world frame, metres. | | [mesh_faces](#aria_nbv.data_handling.ase_efm.EfmSnippetView.mesh_faces) | Optional ``Tensor["F_mesh 3", int64]`` triangle vertex indices. | | [mesh_cache_key](#aria_nbv.data_handling.ase_efm.EfmSnippetView.mesh_cache_key) | Stable key (spec hash) for shared mesh caches across components. | | [mesh_specs](#aria_nbv.data_handling.ase_efm.EfmSnippetView.mesh_specs) | Optional mesh processing specification used to derive attached mesh tensors. | | [camera_rgb](#aria_nbv.data_handling.ase_efm.EfmSnippetView.camera_rgb) | RGB stream (fisheye624, 240x240 in ASE preprocessing). | | [camera_slam_left](#aria_nbv.data_handling.ase_efm.EfmSnippetView.camera_slam_left) | Left SLAM mono stream (fisheye624, ~320x240). | | [camera_slam_right](#aria_nbv.data_handling.ase_efm.EfmSnippetView.camera_slam_right) | Right SLAM mono stream (fisheye624, ~320x240). | | [trajectory](#aria_nbv.data_handling.ase_efm.EfmSnippetView.trajectory) | Rig trajectory (world←rig) aligned to snippet frames. | | [semidense](#aria_nbv.data_handling.ase_efm.EfmSnippetView.semidense) | Semi-dense SLAM points (padded to fixed length). | | [obbs](#aria_nbv.data_handling.ase_efm.EfmSnippetView.obbs) | Return padded ASE GT OBBs for label/evaluation use when present. | | [gt](#aria_nbv.data_handling.ase_efm.EfmSnippetView.gt) | Ground-truth dict ``gt_data`` (EFM OBBs per timestamp/camera). | | [has_mesh](#aria_nbv.data_handling.ase_efm.EfmSnippetView.has_mesh) | Return whether a ground-truth mesh is attached to the snippet. | #### Methods | Name | Description | | --- | --- | | [from_cache_efm](#aria_nbv.data_handling.ase_efm.EfmSnippetView.from_cache_efm) | Construct a snippet view from an offline-cache EFM dict. | | [get_camera](#aria_nbv.data_handling.ase_efm.EfmSnippetView.get_camera) | Return the requested camera stream view from the backing EFM dict. | | [get_occupancy_extend](#aria_nbv.data_handling.ase_efm.EfmSnippetView.get_occupancy_extend) | Return ``[xmin, xmax, ymin, ymax, zmin, zmax]`` AABB in world frame. | | [to](#aria_nbv.data_handling.ase_efm.EfmSnippetView.to) | Return a shallow copy with EFM tensors and wrappers moved to a device. | | [prune_efm](#aria_nbv.data_handling.ase_efm.EfmSnippetView.prune_efm) | Return a view with the EFM dict pruned to the requested keys. | ##### from_cache_efm { #aria_nbv.data_handling.ase_efm.EfmSnippetView.from_cache_efm } ```python data_handling.ase_efm.EfmSnippetView.from_cache_efm( efm, *, mesh=None, crop_bounds=None, mesh_verts=None, mesh_faces=None, mesh_cache_key=None, mesh_specs=None, ) ``` Construct a snippet view from an offline-cache EFM dict. ###### Parameters {.doc-section .doc-section-parameters} | Name | Type | Description | Default | |----------------|------------------------------------------------------------------------------------------------------------------|----------------------------------------------------|------------| | efm | [dict](`dict`)\[[str](`str`), [Any](`typing.Any`)\] | Raw cache sample dict that includes ``"__key__"``. | _required_ | | mesh | [Trimesh](`trimesh.Trimesh`) \| None | Optional GT mesh to attach. | `None` | | crop_bounds | [tuple](`tuple`)\[[torch](`torch`).[Tensor](`torch.Tensor`), [torch](`torch`).[Tensor](`torch.Tensor`)\] \| None | Optional world-space AABB override. | `None` | | mesh_verts | [torch](`torch`).[Tensor](`torch.Tensor`) \| None | Optional cached mesh vertices. | `None` | | mesh_faces | [torch](`torch`).[Tensor](`torch.Tensor`) \| None | Optional cached mesh faces. | `None` | | mesh_cache_key | [str](`str`) \| None | Optional mesh cache key for shared caches. | `None` | | mesh_specs | [MeshProcessSpec](`aria_nbv.data_handling.mesh_cache.MeshProcessSpec`) \| None | Optional mesh processing spec. | `None` | ###### Returns {.doc-section .doc-section-returns} | Name | Type | Description | |--------|--------------------|-----------------------------------| | | \'EfmSnippetView\' | Parsed `EfmSnippetView` instance. | ##### get_camera { #aria_nbv.data_handling.ase_efm.EfmSnippetView.get_camera } ```python data_handling.ase_efm.EfmSnippetView.get_camera(prefix) ``` Return the requested camera stream view from the backing EFM dict. ##### get_occupancy_extend { #aria_nbv.data_handling.ase_efm.EfmSnippetView.get_occupancy_extend } ```python data_handling.ase_efm.EfmSnippetView.get_occupancy_extend() ``` Return ``[xmin, xmax, ymin, ymax, zmin, zmax]`` AABB in world frame. ##### to { #aria_nbv.data_handling.ase_efm.EfmSnippetView.to } ```python data_handling.ase_efm.EfmSnippetView.to(device, *, dtype=None) ``` Return a shallow copy with EFM tensors and wrappers moved to a device. The backing dictionary is rebuilt without mutating this view. Plain tensors honor ``dtype``; EFM wrappers preserve their own dtype contract. The CPU ``Trimesh`` object is shared, while cached mesh tensors and world-frame crop bounds are moved. ##### prune_efm { #aria_nbv.data_handling.ase_efm.EfmSnippetView.prune_efm } ```python data_handling.ase_efm.EfmSnippetView.prune_efm(keep_keys, *, keep_prefixes=()) ``` Return a view with the EFM dict pruned to the requested keys. ###### Parameters {.doc-section .doc-section-parameters} | Name | Type | Description | Default | |---------------|---------------------------------------|----------------------------------------------------------------|------------| | keep_keys | [set](`set`)\[[str](`str`)\] \| None | Exact EFM keys to retain. ``None`` returns ``self`` unchanged. | _required_ | | keep_prefixes | [tuple](`tuple`)\[[str](`str`), ...\] | Optional prefixes; keys starting with any prefix are retained. | `()` | ### EfmTrajectoryView { #aria_nbv.data_handling.ase_efm.EfmTrajectoryView } ```python data_handling.ase_efm.EfmTrajectoryView(t_world_rig, time_ns, gravity_in_world) ``` Expose the MPS/EFM world-from-rig trajectory for one snippet. #### Attributes | Name | Description | | --- | --- | | [t_world_rig](#aria_nbv.data_handling.ase_efm.EfmTrajectoryView.t_world_rig) | ``PoseTW["F 12"]`` world←rig transforms; translation is metres. | | [time_ns](#aria_nbv.data_handling.ase_efm.EfmTrajectoryView.time_ns) | ``Tensor["F", int64]`` pose timestamps. | | [gravity_in_world](#aria_nbv.data_handling.ase_efm.EfmTrajectoryView.gravity_in_world) | ``Tensor["3", float32]`` world-frame gravity in metres per second squared. | | [final_pose](#aria_nbv.data_handling.ase_efm.EfmTrajectoryView.final_pose) | Return the final ``PoseTW["12"]`` world←rig pose in the snippet. | #### Methods | Name | Description | | --- | --- | | [to](#aria_nbv.data_handling.ase_efm.EfmTrajectoryView.to) | Move the trajectory tensors to the requested device and dtype. | ##### to { #aria_nbv.data_handling.ase_efm.EfmTrajectoryView.to } ```python data_handling.ase_efm.EfmTrajectoryView.to(device, *, dtype=None) ``` Move the trajectory tensors to the requested device and dtype. ## Functions | Name | Description | | --- | --- | | [infer_semidense_bounds](#aria_nbv.data_handling.ase_efm.infer_semidense_bounds) | Infer snippet world-space AABB from semidense metadata or points. | | [is_efm_snippet_view_instance](#aria_nbv.data_handling.ase_efm.is_efm_snippet_view_instance) | Return whether ``value`` behaves like an `EfmSnippetView`. | ### infer_semidense_bounds { #aria_nbv.data_handling.ase_efm.infer_semidense_bounds } ```python data_handling.ase_efm.infer_semidense_bounds(efm_dict) ``` Infer snippet world-space AABB from semidense metadata or points. Metadata bounds and finite-point bounds are compared when both exist; the smaller finite volume is retained to avoid an unnecessarily loose crop. Padded and non-finite semidense rows never contribute. #### Parameters {.doc-section .doc-section-parameters} | Name | Type | Description | Default | |----------|---------------------------------------------------------------------------|--------------------------------------------------------------------------------------------|------------| | efm_dict | [Mapping](`collections.abc.Mapping`)\[[str](`str`), [Any](`typing.Any`)\] | Adapted EFM payload containing world-frame semidense points and/or explicit volume bounds. | _required_ | #### Returns {.doc-section .doc-section-returns} | Name | Type | Description | |--------|------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------| | | [tuple](`tuple`)\[[torch](`torch`).[Tensor](`torch.Tensor`), [torch](`torch`).[Tensor](`torch.Tensor`)\] \| None | Pair of ``Tensor["3", float32]`` world-frame AABB minima and maxima in | | | [tuple](`tuple`)\[[torch](`torch`).[Tensor](`torch.Tensor`), [torch](`torch`).[Tensor](`torch.Tensor`)\] \| None | metres on CPU, or ``None`` when no finite bounds are available. | ### is_efm_snippet_view_instance { #aria_nbv.data_handling.ase_efm.is_efm_snippet_view_instance } ```python data_handling.ase_efm.is_efm_snippet_view_instance(value) ``` Return whether ``value`` behaves like an `EfmSnippetView`. The v2 stack accepts both the local data-handling view classes and legacy view objects that expose the same public attributes.