:py:mod:`mipcandy.data.inspection` ================================== .. py:module:: mipcandy.data.inspection .. autodoc2-docstring:: mipcandy.data.inspection :allowtitles: Module Contents --------------- Classes ~~~~~~~ .. list-table:: :class: autosummary longtable :align: left * - :py:obj:`InspectionAnnotation ` - .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotation :summary: * - :py:obj:`InspectionAnnotations ` - * - :py:obj:`ROIDataset ` - * - :py:obj:`RandomROIDataset ` - Functions ~~~~~~~~~ .. list-table:: :class: autosummary longtable :align: left * - :py:obj:`format_bbox ` - .. autodoc2-docstring:: mipcandy.data.inspection.format_bbox :summary: * - :py:obj:`_lists_to_tuples ` - .. autodoc2-docstring:: mipcandy.data.inspection._lists_to_tuples :summary: * - :py:obj:`_str_indices_to_int_indices ` - .. autodoc2-docstring:: mipcandy.data.inspection._str_indices_to_int_indices :summary: * - :py:obj:`parse_inspection_annotation ` - .. autodoc2-docstring:: mipcandy.data.inspection.parse_inspection_annotation :summary: * - :py:obj:`load_inspection_annotations ` - .. autodoc2-docstring:: mipcandy.data.inspection.load_inspection_annotations :summary: * - :py:obj:`bbox_from_indices ` - .. autodoc2-docstring:: mipcandy.data.inspection.bbox_from_indices :summary: * - :py:obj:`inspect ` - .. autodoc2-docstring:: mipcandy.data.inspection.inspect :summary: * - :py:obj:`crop_and_pad ` - .. autodoc2-docstring:: mipcandy.data.inspection.crop_and_pad :summary: API ~~~ .. py:function:: format_bbox(bbox: typing.Sequence[int]) -> tuple[int, int, int, int] | tuple[int, int, int, int, int, int] :canonical: mipcandy.data.inspection.format_bbox .. autodoc2-docstring:: mipcandy.data.inspection.format_bbox .. py:class:: InspectionAnnotation :canonical: mipcandy.data.inspection.InspectionAnnotation Bases: :py:obj:`object` .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotation .. py:attribute:: shape :canonical: mipcandy.data.inspection.InspectionAnnotation.shape :type: mipcandy.types.AmbiguousShape :value: None .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotation.shape .. py:attribute:: foreground_bbox :canonical: mipcandy.data.inspection.InspectionAnnotation.foreground_bbox :type: tuple[int, int, int, int] | tuple[int, int, int, int, int, int] :value: None .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotation.foreground_bbox .. py:attribute:: class_ids :canonical: mipcandy.data.inspection.InspectionAnnotation.class_ids :type: tuple[int, ...] :value: None .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotation.class_ids .. py:attribute:: class_counts :canonical: mipcandy.data.inspection.InspectionAnnotation.class_counts :type: dict[int, int] :value: None .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotation.class_counts .. py:attribute:: class_bboxes :canonical: mipcandy.data.inspection.InspectionAnnotation.class_bboxes :type: dict[int, tuple[int, int, int, int] | tuple[int, int, int, int, int, int]] :value: None .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotation.class_bboxes .. py:attribute:: class_locations :canonical: mipcandy.data.inspection.InspectionAnnotation.class_locations :type: dict[int, tuple[tuple[int, int] | tuple[int, int, int], ...]] :value: None .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotation.class_locations .. py:attribute:: spacing :canonical: mipcandy.data.inspection.InspectionAnnotation.spacing :type: mipcandy.types.Shape | None :value: None .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotation.spacing .. py:method:: foreground_shape() -> mipcandy.types.Shape :canonical: mipcandy.data.inspection.InspectionAnnotation.foreground_shape .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotation.foreground_shape .. py:method:: center_of_foreground() -> tuple[int, int] | tuple[int, int, int] :canonical: mipcandy.data.inspection.InspectionAnnotation.center_of_foreground .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotation.center_of_foreground .. py:class:: InspectionAnnotations(dataset: mipcandy.data.dataset.SupervisedDataset, background: int, intensity_stats: tuple[float, float, float, float], *annotations: mipcandy.data.inspection.InspectionAnnotation) :canonical: mipcandy.data.inspection.InspectionAnnotations Bases: :py:obj:`typing.Sequence`\ [\ :py:obj:`mipcandy.data.inspection.InspectionAnnotation`\ ] .. py:method:: dataset() -> mipcandy.data.dataset.SupervisedDataset :canonical: mipcandy.data.inspection.InspectionAnnotations.dataset .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotations.dataset .. py:method:: background() -> int :canonical: mipcandy.data.inspection.InspectionAnnotations.background .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotations.background .. py:method:: intensity_stats() -> tuple[float, float, float, float] :canonical: mipcandy.data.inspection.InspectionAnnotations.intensity_stats .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotations.intensity_stats .. py:method:: annotations() -> tuple[mipcandy.data.inspection.InspectionAnnotation, ...] :canonical: mipcandy.data.inspection.InspectionAnnotations.annotations .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotations.annotations .. py:method:: __getitem__(item: int) -> mipcandy.data.inspection.InspectionAnnotation :canonical: mipcandy.data.inspection.InspectionAnnotations.__getitem__ .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotations.__getitem__ .. py:method:: __len__() -> int :canonical: mipcandy.data.inspection.InspectionAnnotations.__len__ .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotations.__len__ .. py:method:: save(path: str | os.PathLike[str]) -> None :canonical: mipcandy.data.inspection.InspectionAnnotations.save .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotations.save .. py:method:: _get_shapes(get_shape: typing.Callable[[mipcandy.data.inspection.InspectionAnnotation], mipcandy.types.AmbiguousShape]) -> tuple[tuple[int, ...] | None, tuple[int, ...], tuple[int, ...]] :canonical: mipcandy.data.inspection.InspectionAnnotations._get_shapes .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotations._get_shapes .. py:method:: shapes() -> tuple[tuple[int, ...] | None, tuple[int, ...], tuple[int, ...]] :canonical: mipcandy.data.inspection.InspectionAnnotations.shapes .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotations.shapes .. py:method:: statistical_shape(*, percentile: float = 0.95) -> mipcandy.types.Shape :canonical: mipcandy.data.inspection.InspectionAnnotations.statistical_shape .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotations.statistical_shape .. py:method:: foreground_shapes() -> tuple[tuple[int, ...] | None, tuple[int, ...], tuple[int, ...]] :canonical: mipcandy.data.inspection.InspectionAnnotations.foreground_shapes .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotations.foreground_shapes .. py:method:: statistical_foreground_shape(*, percentile: float = 0.95) -> mipcandy.types.Shape :canonical: mipcandy.data.inspection.InspectionAnnotations.statistical_foreground_shape .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotations.statistical_foreground_shape .. py:method:: crop_foreground(i: int, *, expand_ratio: float = 1) -> tuple[torch.Tensor, torch.Tensor] :canonical: mipcandy.data.inspection.InspectionAnnotations.crop_foreground .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotations.crop_foreground .. py:method:: foreground_heatmap() -> torch.Tensor :canonical: mipcandy.data.inspection.InspectionAnnotations.foreground_heatmap .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotations.foreground_heatmap .. py:method:: center_of_foregrounds() -> tuple[int, int] | tuple[int, int, int] :canonical: mipcandy.data.inspection.InspectionAnnotations.center_of_foregrounds .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotations.center_of_foregrounds .. py:method:: center_of_foregrounds_offsets() -> tuple[int, int] | tuple[int, int, int] :canonical: mipcandy.data.inspection.InspectionAnnotations.center_of_foregrounds_offsets .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotations.center_of_foregrounds_offsets .. py:method:: set_roi_shape(roi_shape: mipcandy.types.Shape | None) -> None :canonical: mipcandy.data.inspection.InspectionAnnotations.set_roi_shape .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotations.set_roi_shape .. py:method:: roi_shape(*, clamp: bool = True, percentile: float = 0.95) -> mipcandy.types.Shape :canonical: mipcandy.data.inspection.InspectionAnnotations.roi_shape .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotations.roi_shape .. py:method:: roi(i: int, *, clamp: bool = True, percentile: float = 0.95) -> tuple[int, int, int, int] | tuple[int, int, int, int, int, int] :canonical: mipcandy.data.inspection.InspectionAnnotations.roi .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotations.roi .. py:method:: crop_roi(i: int, *, clamp: bool = True, percentile: float = 0.95) -> tuple[torch.Tensor, torch.Tensor] :canonical: mipcandy.data.inspection.InspectionAnnotations.crop_roi .. autodoc2-docstring:: mipcandy.data.inspection.InspectionAnnotations.crop_roi .. py:function:: _lists_to_tuples(pairs: typing.Sequence[tuple[str, typing.Any]]) -> dict[str, typing.Any] :canonical: mipcandy.data.inspection._lists_to_tuples .. autodoc2-docstring:: mipcandy.data.inspection._lists_to_tuples .. py:function:: _str_indices_to_int_indices(obj: dict[str, typing.Any]) -> dict[int, typing.Any] :canonical: mipcandy.data.inspection._str_indices_to_int_indices .. autodoc2-docstring:: mipcandy.data.inspection._str_indices_to_int_indices .. py:function:: parse_inspection_annotation(obj: dict[str, typing.Any]) -> mipcandy.data.inspection.InspectionAnnotation :canonical: mipcandy.data.inspection.parse_inspection_annotation .. autodoc2-docstring:: mipcandy.data.inspection.parse_inspection_annotation .. py:function:: load_inspection_annotations(path: str | os.PathLike[str], dataset: mipcandy.data.dataset.SupervisedDataset) -> mipcandy.data.inspection.InspectionAnnotations :canonical: mipcandy.data.inspection.load_inspection_annotations .. autodoc2-docstring:: mipcandy.data.inspection.load_inspection_annotations .. py:function:: bbox_from_indices(indices: torch.Tensor, num_dim: typing.Literal[2, 3]) -> tuple[int, int, int, int] :canonical: mipcandy.data.inspection.bbox_from_indices .. autodoc2-docstring:: mipcandy.data.inspection.bbox_from_indices .. py:function:: inspect(dataset: mipcandy.data.dataset.SupervisedDataset, *, background: int = 0, max_samples: int = 10000, console: rich.console.Console = Console()) -> mipcandy.data.inspection.InspectionAnnotations :canonical: mipcandy.data.inspection.inspect .. autodoc2-docstring:: mipcandy.data.inspection.inspect .. py:class:: ROIDataset(annotations: mipcandy.data.inspection.InspectionAnnotations, *, clamp: bool = True, percentile: float = 0.95) :canonical: mipcandy.data.inspection.ROIDataset Bases: :py:obj:`mipcandy.data.dataset.SupervisedDataset`\ [\ :py:obj:`list`\ [\ :py:obj:`int`\ ]\ ] .. py:method:: construct_new(images: list[typing.Any], labels: list[typing.Any]) -> typing.Self :canonical: mipcandy.data.inspection.ROIDataset.construct_new .. autodoc2-docstring:: mipcandy.data.inspection.ROIDataset.construct_new .. py:method:: load_image(idx: int) -> torch.Tensor :canonical: mipcandy.data.inspection.ROIDataset.load_image :abstractmethod: .. autodoc2-docstring:: mipcandy.data.inspection.ROIDataset.load_image .. py:method:: load_label(idx: int) -> torch.Tensor :canonical: mipcandy.data.inspection.ROIDataset.load_label :abstractmethod: .. autodoc2-docstring:: mipcandy.data.inspection.ROIDataset.load_label .. py:method:: load(idx: int) -> tuple[torch.Tensor, torch.Tensor] :canonical: mipcandy.data.inspection.ROIDataset.load .. py:function:: crop_and_pad(x: torch.Tensor, bbox_lbs: list[int], bbox_ubs: list[int], *, pad_value: int | float = 0) -> torch.Tensor :canonical: mipcandy.data.inspection.crop_and_pad .. autodoc2-docstring:: mipcandy.data.inspection.crop_and_pad .. py:class:: RandomROIDataset(annotations: mipcandy.data.inspection.InspectionAnnotations, batch_size: int, *, num_patches_per_case: int = 1, oversample_rate: float = 0.33, clamp: bool = False, percentile: float = 0.5, min_factor: int = 16) :canonical: mipcandy.data.inspection.RandomROIDataset Bases: :py:obj:`mipcandy.data.inspection.ROIDataset` .. py:method:: convert_idx(idx: int) -> int :canonical: mipcandy.data.inspection.RandomROIDataset.convert_idx .. autodoc2-docstring:: mipcandy.data.inspection.RandomROIDataset.convert_idx .. py:method:: roi_shape(*, roi_shape: mipcandy.types.Shape | None = None) -> None | mipcandy.types.Shape :canonical: mipcandy.data.inspection.RandomROIDataset.roi_shape .. autodoc2-docstring:: mipcandy.data.inspection.RandomROIDataset.roi_shape .. py:method:: construct_new(images: list[typing.Any], labels: list[typing.Any]) -> typing.Self :canonical: mipcandy.data.inspection.RandomROIDataset.construct_new .. autodoc2-docstring:: mipcandy.data.inspection.RandomROIDataset.construct_new .. py:method:: random_roi(idx: int, force_foreground: bool) -> tuple[list[int], list[int]] :canonical: mipcandy.data.inspection.RandomROIDataset.random_roi .. autodoc2-docstring:: mipcandy.data.inspection.RandomROIDataset.random_roi .. py:method:: oversample_foreground(idx: int) -> bool :canonical: mipcandy.data.inspection.RandomROIDataset.oversample_foreground .. autodoc2-docstring:: mipcandy.data.inspection.RandomROIDataset.oversample_foreground .. py:method:: load(idx: int) -> tuple[torch.Tensor, torch.Tensor] :canonical: mipcandy.data.inspection.RandomROIDataset.load