loki2.inference.inference_memory
================================

.. py:module:: loki2.inference.inference_memory

.. autoapi-nested-parse::

   Loki2 Inference Pipeline for Whole Slide Images (WSI) in Memory.

   This module provides an in-memory inference pipeline for CellViT models,
   including functionalities for processing WSIs, cleaning GeoJSON outputs,
   and handling various output formats.

   Patches dataset requirements:
   - Patch-Size must be 1024, with overlap of 64







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

.. py:function:: clean_geojson(input_fp: str, output_fp: str, *, minify: bool = True, gzip_if_ext: bool = True, drop_empty_props: bool = True) -> None

   Clean and optimize a GeoJSON file.

   Read a GeoJSON (FeatureCollection or list of Features),
   remove null/empty geometries, make valid via buffer(0),
   cast coordinates to int, optionally strip empty properties,
   and write a minified (and optionally gzipped) output.

   Geometry/topology is preserved except for make-valid (buffer(0)), and
   coordinates are integer-cast exactly as before.

   :param input_fp: Path to input GeoJSON file.
   :param output_fp: Path to output GeoJSON file.
   :param minify: Whether to minify JSON output. Defaults to True.
   :param gzip_if_ext: Whether to gzip if output has .gz extension. Defaults to True.
   :param drop_empty_props: Whether to drop empty properties. Defaults to True.

   :raises ValueError: If GeoJSON format is unexpected.


.. py:class:: CellViTInferenceMemory(model_path: Union[pathlib.Path, str], gpu: int, outdir: Union[pathlib.Path, str], classifier_path: Union[pathlib.Path, str] = None, binary: bool = False, batch_size: int = 8, patch_size: int = 1024, overlap: int = 64, geojson: bool = False, graph: bool = False, compression: bool = False, geojson_gz: bool = False, enforce_mixed_precision: bool = False)

   Bases: :py:obj:`loki2.inference.inference_disk.CellViTInference`


   CellViT Inference class for in-memory processing of WSIs.

   This class extends `CellViTInference` to handle WSI processing entirely in memory,
   which is suitable for smaller WSIs or when disk I/O is a bottleneck.
   It integrates with Ray for parallel processing of patches.

   :param model_path: Path to model checkpoint.
   :param gpu: CUDA GPU ID to use.
   :param outdir: Output directory to store results.
   :param classifier_path: Path to a classifier (.pth) to exchange UniSeg classification
                           results with a new scheme. Defaults to None.
   :param binary: If just a binary detection/segmentation should be performed.
                  Cannot be used with classifier. Defaults to False.
   :param batch_size: Batch-size for inference. Defaults to 8.
   :param patch_size: Size of patches to extract. Defaults to 1024.
   :param overlap: Overlap between patches in pixels. Defaults to 64.
   :param geojson: If a geojson export should be performed. Defaults to False.
   :param graph: If a graph export should be performed. Defaults to False.
   :param compression: If a snappy compression should be performed. Defaults to False.
   :param geojson_gz: If a compressed geojson.gz export should be performed with cleaned
                      geometry. Defaults to False.
   :param enforce_mixed_precision: Using PyTorch autocasting with dtype float16 to speed
                                   up inference. Can be used to enforce amp inference even for networks trained
                                   without amp. Otherwise, the network setting is used. Defaults to False.


   .. py:attribute:: outdir


   .. py:attribute:: geojson_gz


   .. py:method:: process_wsi(wsi_path: Union[pathlib.Path, str], wsi_properties: dict = {}, resolution: float = 0.25, apply_prefilter: bool = True, filter_patches: bool = False, **kwargs) -> None

      Process a whole slide image with CellViT.

      :param wsi_path: Path to the whole slide image.
      :type wsi_path: Union[Path, str]
      :param wsi_properties: Optional WSI properties,
                             Allowed keys are 'slide_mpp' and 'magnification'. Defaults to {}.
      :type wsi_properties: dict, optional
      :param resolution: Target resolution. Defaults to 0.25.
      :type resolution: float, optional
      :param apply_prefilter: Prefilter. Defaults to True.
      :type apply_prefilter: bool, optional
      :param filter_patches: Filter patches after processing. Defaults to False.
      :type filter_patches: bool, optional