module

assignment.lap

module unitrack.assignment.lap

CUDA LAP solvers backed by :mod:`torchmatch.assignment`.

Three backends (classical/Munkres, tree/Lawler, hybrid/deprecated) plus a batched entry point. Prefer :func:`unitrack.assignment.auto_assignment` / :class:`unitrack.assignment.AutoLAP` for new code; reach for these backends only when you need explicit CUDA solver control.

Members

type

  • BackendLAP solver backend selection.
  • LAPSolves the linear assignment problem via torchmatch CUDA solvers.

function

  • lap_assignmentSolve the linear assignment problem via a torchmatch CUDA solver.
  • lap_batch_assignmentSolve a batch of linear assignment problems via torchmatch CUDA solvers.

Source: unitrack/assignment/lap/__init__.py:1