:orphan:

:py:mod:`traccuracy.matchers._iou`
==================================

.. py:module:: traccuracy.matchers._iou


Module Contents
---------------

Classes
~~~~~~~

.. autoapisummary::

   traccuracy.matchers._iou.IOUMatcher



Functions
~~~~~~~~~

.. autoapisummary::

   traccuracy.matchers._iou.match_iou



.. py:function:: match_iou(gt: traccuracy._tracking_graph.TrackingGraph, pred: traccuracy._tracking_graph.TrackingGraph, threshold: float = 0.6, one_to_one: bool = False) -> list[tuple[collections.abc.Hashable, collections.abc.Hashable]]

   Identifies pairs of cells between gt and pred that have iou > threshold

   This can return more than one match for any node
   Assumes that within a frame, each object has a unique segmentation label
   and that the label is recorded on each node using label_key

   :param gt: Tracking data object containing graph and segmentations
   :type gt: traccuracy.TrackingGraph
   :param pred: Tracking data object containing graph and segmentations
   :type pred: traccuracy.TrackingGraph
   :param threshold: Minimum IoU for matching cells. Defaults to 0.6.
   :type threshold: float, optional
   :param one_to_one: If True, forces the mapping to be one-to-one by running
                      linear assignment on the thresholded iou array. Default False.
   :type one_to_one: optional, bool

   :returns: list of tuples where each tuple contains a gt node and pred node
   :rtype: list[(gt_node, pred_node)]

   :raises ValueError: gt and pred must be a TrackingData object
   :raises ValueError: GT and pred segmentations must be the same shape


.. py:class:: IOUMatcher(iou_threshold: float = 0.6, one_to_one: bool = False)




   Constructs a mapping between gt and pred nodes using the IoU of the segmentations

   Lower values for iou_threshold will be more permissive of imperfect matches

   :param iou_threshold: Minimum IoU value to assign a match. Defaults to 0.6.
   :type iou_threshold: float, optional
   :param one_to_one: If True, forces the mapping to be one-to-one by running
                      linear assignment on the thresholded iou array. Default False.
   :type one_to_one: optional, bool

   .. py:property:: info
      :type: dict[str, Any]

      Dictionary of Matcher name and any parameters

   .. py:method:: compute_mapping(gt_graph: traccuracy._tracking_graph.TrackingGraph, pred_graph: traccuracy._tracking_graph.TrackingGraph) -> traccuracy.matchers._matched.Matched

      Run the matching on a given set of gt and pred TrackingGraph and returns a Matched object
      with a new copy of each TrackingGraph

      :param gt_graph: Tracking graph object for the gt
      :type gt_graph: traccuracy.TrackingGraph
      :param pred_graph: Tracking graph object for the pred
      :type pred_graph: traccuracy.TrackingGraph

      :returns: Matched data object
      :rtype: matched (traccuracy.matchers.Matched)

      :raises ValueError: gt and pred must be a TrackingGraph object