Utilities (.util)

Code Statistics (.util.codestats)

compute

The model is already indexed into topological order by setup.py.

• choice_probs[choice_id] gives branch probabilities for one discrete choice

• choice_branches[choice_id] gives the branch ids for that choice

• branch_vars[branch_id] is a bitmask of variables activated by that branch

• var_kind[var_id] is “choice” or “continuous”

infinigen2.util.codestats.compute.shannon_entropy_bits(probabilities) → float

Parameters:

probabilities

infinigen2.util.codestats.compute.cyclomatic_complexity(choice_probs) → int

Cyclomatic complexity where only control flow is pf.control.hcoice Return 1 + sum(branch_count - 1) over discrete choices.

Parameters:

choice_probs (dict[int, list[float]])

infinigen2.util.codestats.compute.solve_tree(indexed_model) → dict

Tree-style recursion.

child subtrees inside one branch are treated as disjoint and summed independently.

Parameters:

indexed_model (dict)

infinigen2.util.codestats.compute.solve_dp(indexed_model) → dict

Exact DP

Parameters:

indexed_model (dict)

setup

Compute-graph extraction and stat summaries

Representation after setup:

• choice_probs[choice_id] gives branch probabilities for one discrete choice

• choice_branches[choice_id] gives the branch ids for that choice

• branch_vars[branch_id] gives the variables activated by that branch

• var_kind[var_id] is “choice” or “continuous”

This code handles: - extracting this model from a compute graph - topologically reindexing variables so the solver can use bitmasks - calling either the plain recursion or the memoized DP

infinigen2.util.codestats.setup.build_model_from_compute_graph(graph) → dict

Extract the compact branch-first model from a traced compute graph.

Parameters:

graph (ComputeGraph)

infinigen2.util.codestats.setup.build_indexed_model(model) → dict

Reindex variables into topological order and convert branches to masks.

Parameters:

model (dict)

infinigen2.util.codestats.setup.compute_stats(...) → dict

Extract the compact model and run either the tree or DP solver.

Parameters:

• graph (ComputeGraph)

• method (str) -- (default: 'dp')

• continuous_bits (float) -- (default: 3.0)

• trace_level (TraceLevel | None) -- (default: None)

Module contents

infinigen2.util.codestats.build_indexed_model(model) → dict

Reindex variables into topological order and convert branches to masks.

Parameters:

model (dict)

infinigen2.util.codestats.build_model_from_compute_graph(graph) → dict

Extract the compact branch-first model from a traced compute graph.

Parameters:

graph (ComputeGraph)

infinigen2.util.codestats.compute_stats(...) → dict

Extract the compact model and run either the tree or DP solver.

Parameters:

• graph (ComputeGraph)

• method (str) -- (default: 'dp')

• continuous_bits (float) -- (default: 3.0)

• trace_level (TraceLevel | None) -- (default: None)

infinigen2.util.codestats.cyclomatic_complexity(choice_probs) → int

Cyclomatic complexity where only control flow is pf.control.hcoice Return 1 + sum(branch_count - 1) over discrete choices.

Parameters:

choice_probs (dict[int, list[float]])

infinigen2.util.codestats.solve_dp(indexed_model) → dict

Exact DP

Parameters:

indexed_model (dict)

infinigen2.util.codestats.solve_tree(indexed_model) → dict

Tree-style recursion.

child subtrees inside one branch are treated as disjoint and summed independently.

Parameters:

indexed_model (dict)

camera_projection

infinigen2.util.camera_projection.adjust_camera_sensor(camera)

Parameters:

camera (CameraObject)

infinigen2.util.camera_projection.bpy_resolution() → tuple[int, int]

infinigen2.util.camera_projection.get_calibration_matrix_K_from_blender(camera) → ndarray

Build intrinsic camera parameters from Blender camera data

Based on https://blender.stackexchange.com/questions/15102/what-is-blenders-camera-projection-matrix-model

Parameters:

camera (CameraObject) -- Blender camera object

Returns:

3x3 intrinsic camera matrix

Return type:

K

infinigen2.util.camera_projection.get_3x4_RT_matrix_from_blender(camera) → ndarray

Get camera rotation and translation matrices from Blender.

We return a camera matrix in the usual computer vision convention: +x is horizontal, +y is down, +z is forward.

This is different than blender’s camera typical convention, which is +x is horizontal, +y is up, -z is forward.

Parameters:

camera (CameraObject) -- Blender camera object

Returns:

3x4 camera matrix

Return type:

RT

infinigen2.util.camera_projection.get_3x4_P_matrix_from_blender(cam)

Parameters:

cam

infinigen2.util.camera_projection.project_points(camera, points_N3) → ndarray

Project points onto the camera image plane.

Parameters:

• camera (CameraObject)

• points_N3 (ndarray)

infinigen2.util.camera_projection.is_projection_within_image(projected, resolution=None) → ndarray

Check if the projection is within the image.

Parameters:

• projected (ndarray)

• resolution (tuple[int, int] | None) -- (default: None)

infinigen2.util.camera_projection.camera_depth_raycast(cam, colliders, meshgridcoords) → ndarray

Parameters:

• cam (CameraObject)

• colliders (CollisionSet)

• meshgridcoords (ndarray)

infinigen2.util.camera_projection.get_camera_parameters(camera, frame=None, use_dof=False) → dict

Parameters:

• camera (CameraObject)

• frame (int | None) -- (default: None)

• use_dof (bool | None) -- (default: False)

curve

class infinigen2.util.curve.CurveToMeshWithUvResult(mesh, vector)

Bases: NamedTuple

mesh: pf.ProcNode[pf.MeshObject]

vector: pf.ProcNode[pf.Vector]

class infinigen2.util.curve.FilletMaskResult(...)

Bases: NamedTuple

curve: pf.ProcNode[CurveObject]

is_fillet_point: pf.ProcNode[bool]

is_flat_point: pf.ProcNode[bool]

orig_point_idx: pf.ProcNode[int]

infinigen2.util.curve.curve_to_mesh_with_uv(

curve,

profile,

fill_caps=False,

) → CurveToMeshWithUvResult

Parameters:

• curve (pf.ProcNode[CurveObject])

• profile (pf.ProcNode[CurveObject])

• fill_caps (pf.ProcNode[bool] | bool) -- (default: False)

infinigen2.util.curve.curve_to_noncyclic(input) → pf.ProcNode[CurveObject]

Parameters:

input (pf.ProcNode[CurveObject])

infinigen2.util.curve.fillet_mask(...) → FilletMaskResult

Parameters:

• geometry (pf.ProcNode[CurveObject])

• selection (pf.ProcNode[bool] | bool)

• fillet_vertices (pf.ProcNode[int] | int)

• radius (pf.ProcNode[float] | float)

errors

exception infinigen2.util.errors.RejectedScene

Bases: Exception

external_assets

infinigen2.util.external_assets.distribution_from_asset_glob(folder)

Parameters:

folder (Path)

infinigen2.util.external_assets.pregenerated_asset_rand(relative_glob) → Callable[[pf.RNG], pf.MeshObject | None]

Parameters:

relative_glob (str)

hardware_info

infinigen2.util.hardware_info.get_hardware_info() → dict

import_utils

Utilities for dynamic imports and finding generator functions.

infinigen2.util.import_utils.module_path()

instance

infinigen2.util.instance.instance(...) → pf.ProcNode

Parameters:

• parent (pf.ProcNode[pf.MeshObject])

• child (pf.ProcNode[pf.Collection])

• seed (pf.ProcNode[int]) -- (default: 0)

• density (pf.ProcNode[float]) -- (default: 2.0)

• distance_min (pf.ProcNode[float]) -- (default: 0.0)

• offset (pf.ProcNode[pf.Vector]) -- (default: (0, 0, 0.002))

infinigen2.util.instance.instanced_objects(...) → list[pf.MeshObject]

Parameters:

• rng (Generator)

• parent (pf.MeshObject)

• child (pf.Collection)

• density (float) -- (default: 2.0)

• distance_min (float) -- (default: 0.0)

• offset (tuple[float, float, float]) -- (default: (0, 0, 0.002))

math

infinigen2.util.math.offset_spacing(...) → pf.ProcNode[float]

Parameters:

• seed (pf.ProcNode[int] | int)

• dimension (pf.ProcNode[float] | float)

• index (pf.ProcNode[int] | int)

• vertices (pf.ProcNode[int] | int)

• max_offset_pct (pf.ProcNode[float] | float)

mesh

class infinigen2.util.mesh.CubeWithVertexIndicesResult(...)

Bases: NamedTuple

mesh: pf.ProcNode[pf.MeshObject]

index_x: pf.ProcNode[int]

index_y: pf.ProcNode[int]

index_z: pf.ProcNode[int]

class infinigen2.util.mesh.ExtrudeSeamlessResult(mesh, top, side)

Bases: NamedTuple

mesh: pf.ProcNode[pf.MeshObject]

top: pf.ProcNode[bool]

side: pf.ProcNode[bool]

class infinigen2.util.mesh.WallCutoutResult(wall, sill, lightblocker)

Bases: NamedTuple

wall: pf.ProcNode[pf.MeshObject]

sill: pf.ProcNode[pf.MeshObject]

lightblocker: pf.ProcNode[pf.MeshObject]

infinigen2.util.mesh.corner_box(...) → CubeWithVertexIndicesResult

Parameters:

• size (pf.ProcNode[pf.Vector] | pf.Vector) -- (default: (1, 1, 1))

• loops_x (pf.ProcNode[int] | int) -- (default: 0)

• loops_y (pf.ProcNode[int] | int) -- (default: 0)

• loops_z (pf.ProcNode[int] | int) -- (default: 0)

• support_loop_offset (pf.ProcNode[pf.Vector] | pf.Vector) -- (default: (0.05, 0.05, 0.05))

infinigen2.util.mesh.crease_sharp(

mesh,

threshold_degrees,

) → pf.ProcNode[pf.MeshObject]

Parameters:

• mesh (pf.ProcNode[pf.MeshObject])

• threshold_degrees (pf.ProcNode[float] | float)

infinigen2.util.mesh.extrude_mesh_seamless_uvs(...) → ExtrudeSeamlessResult

Extrude faces and continue source UVs onto the new side faces seamlessly.

Each side corner gets uv0 plus a perpendicular offset proportional to its extrusion depth. The boundary edge driving that offset is found per corner via its own next-corner edge (the one with both ends on the depth-0 ring), then broadcast across the side face with accumulate_field.

Parameters:

• mesh (pf.ProcNode[pf.MeshObject])

• selection (pf.ProcNode[bool] | bool)

• offset_scale (pf.ProcNode[float] | float)

• uv_winding_sign (pf.ProcNode[float] | float) -- (default: 1.0)

infinigen2.util.mesh.grid_from_corners(...) → pf.ProcNode

Parameters:

• point_1 (pf.ProcNode[pf.Vector] | pf.Vector)

• point_2 (pf.ProcNode[pf.Vector] | pf.Vector)

• vertices_x (pf.ProcNode[int] | int)

• vertices_y (pf.ProcNode[int] | int)

infinigen2.util.mesh.uv_winding_sign(obj) → float

Sign of the first face’s UV winding (+1/-1), for un-mirrored continuation.

Parameters:

obj (pf.MeshObject)

infinigen2.util.mesh.wall_cutout_split(...) → WallCutoutResult

Split a flat cutout_sel-tagged surface into holed wall, sill tunnels, and lightblocker backing.

The mouth of each hole gets a chamfer-wide angled lip, kept on the wall (it reads as a bevel of the wall surface around the opening); the sill is just the deep reveal tunnel. The footprint must be pre-expanded by chamfer (see face_expand_margin) so the inner opening lands at the originally-intended size.

Parameters:

• geometry (pf.ProcNode[pf.MeshObject])

• thickness (pf.ProcNode[float] | float)

• blocker_thickness (pf.ProcNode[float] | float) -- (default: 0.1)

• uv_winding_sign (pf.ProcNode[float] | float) -- (default: 1.0)

• delete_facecap (pf.ProcNode[bool] | bool) -- (default: True)

• chamfer (pf.ProcNode[float] | float) -- (default: 0.006)

render_metadata

infinigen2.util.render_metadata.time_step(times, name) → Iterator[None]

Parameters:

• times (dict[str, float])

• name (str)

infinigen2.util.render_metadata.write_render_metadata(...) → dict

Parameters:

• output (str | Path)

• seed (int)

• times (dict[str, float])

• exports (dict[ExportType, list[Path]])

• build_keys (set[str])

• render_keys (set[str])

• n_frames (int)

scene_cleanup

infinigen2.util.scene_cleanup.cleanup_except(keep) → list[str]

Delete every bpy.data.objects entry not in keep.

Placement helpers (e.g. repeat_attempts) leave failed-placement objects in the scene. They are excluded from the returned all_objects lists but still live in the blend and would otherwise be rendered. Pass the union of objects/cameras/lights you want kept and this removes the rest.

Parameters:

keep (Iterable[Object])

Module contents