:py:mod:`anomalib.models.patchcore.anomaly_map`
===============================================

.. py:module:: anomalib.models.patchcore.anomaly_map

.. autoapi-nested-parse::

   Anomaly Map Generator for the PatchCore model implementation.



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

Classes
~~~~~~~

.. autoapisummary::

   anomalib.models.patchcore.anomaly_map.AnomalyMapGenerator




.. py:class:: AnomalyMapGenerator(input_size: Union[omegaconf.ListConfig, Tuple], sigma: int = 4)

   Bases: :py:obj:`torch.nn.Module`

   Generate Anomaly Heatmap.

   .. py:method:: compute_anomaly_map(patch_scores: torch.Tensor, feature_map_shape: torch.Size) -> torch.Tensor

      Pixel Level Anomaly Heatmap.

      :param patch_scores: Patch-level anomaly scores
      :type patch_scores: torch.Tensor
      :param feature_map_shape: 2-D feature map shape (width, height)
      :type feature_map_shape: torch.Size

      :returns: Map of the pixel-level anomaly scores
      :rtype: torch.Tensor


   .. py:method:: compute_anomaly_score(patch_scores: torch.Tensor) -> torch.Tensor
      :staticmethod:

      Compute Image-Level Anomaly Score.

      :param patch_scores: Patch-level anomaly scores
      :type patch_scores: torch.Tensor

      :returns: Image-level anomaly scores
      :rtype: torch.Tensor


   .. py:method:: forward(**kwargs: torch.Tensor) -> Tuple[torch.Tensor, torch.Tensor]

      Returns anomaly_map and anomaly_score.

      Expects `patch_scores` keyword to be passed explicitly
      Expects `feature_map_shape` keyword to be passed explicitly

      Example
      >>> anomaly_map_generator = AnomalyMapGenerator(input_size=input_size)
      >>> map, score = anomaly_map_generator(patch_scores=numpy_array, feature_map_shape=feature_map_shape)

      :raises ValueError: If `patch_scores` key is not found

      :returns: anomaly_map, anomaly_score
      :rtype: Tuple[torch.Tensor, torch.Tensor]