Class ResizeWithPadOrCrop

monai/transforms/croppad/array.py:1397–1502

Resize an image to a target spatial size by either centrally cropping the image or padding it evenly with a user-specified mode. When the dimension is smaller than the target size, do symmetric padding along that dim. When the dimension is larger than the target size, do central cro

class ResizeWithPadOrCrop(InvertibleTransform, LazyTransform):
    """
    Resize an image to a target spatial size by either centrally cropping the image or
    padding it evenly with a user-specified mode.
    When the dimension is smaller than the target size, do symmetric padding along that dim.
    When the dimension is larger than the target size, do central cropping along that dim.

    This transform is capable of lazy execution. See the :ref:`Lazy Resampling topic<lazy_resampling>`
    for more information.

    Args:
        spatial_size: the spatial size of output data after padding or crop.
            If has non-positive values, the corresponding size of input image will be used (no padding).
        method: {``"symmetric"``, ``"end"``}
            Pad image symmetrically on every side or only pad at the end sides. Defaults to ``"symmetric"``.
        mode: available modes for numpy array:{``"constant"``, ``"edge"``, ``"linear_ramp"``, ``"maximum"``,
            ``"mean"``, ``"median"``, ``"minimum"``, ``"reflect"``, ``"symmetric"``, ``"wrap"``, ``"empty"``}
            available modes for PyTorch Tensor: {``"constant"``, ``"reflect"``, ``"replicate"``, ``"circular"``}.
            One of the listed string values or a user supplied function. Defaults to ``"constant"``.
            See also: https://numpy.org/doc/1.18/reference/generated/numpy.pad.html
            https://pytorch.org/docs/stable/generated/torch.nn.functional.pad.html
        pad_kwargs: other arguments for the `np.pad` or `torch.pad` function.
            note that `np.pad` treats channel dimension as the first dimension.
        lazy: a flag to indicate whether this transform should execute lazily or not. Defaults to False.

    """

    backend = list(set(SpatialPad.backend) & set(CenterSpatialCrop.backend))

    def __init__(
        self,
        spatial_size: Sequence[int] | int,
        method: str = Method.SYMMETRIC,
        mode: str = PytorchPadMode.CONSTANT,
        lazy: bool = False,
        **pad_kwargs,
    ):
        LazyTransform.__init__(self, lazy)
        self.padder = SpatialPad(spatial_size=spatial_size, method=method, mode=mode, lazy=lazy, **pad_kwargs)
        self.cropper = CenterSpatialCrop(roi_size=spatial_size, lazy=lazy)

    @LazyTransform.lazy.setter  # type: ignore
    def lazy(self, val: bool):
        self.padder.lazy = val
        self.cropper.lazy = val
        self._lazy = val

    def __call__(  # type: ignore[override]
        self, img: torch.Tensor, mode: str | None = None, lazy: bool | None = None, **pad_kwargs
    ) -> torch.Tensor:
        """
        Args:
            img: data to pad or crop, assuming `img` is channel-first and
                padding or cropping doesn&#x27;t apply to the channel dim.
            mode: available modes for numpy array:{``"constant"``, ``"edge"``, ``"linear_ramp"``, ``"maximum"``,
                ``"mean"``, ``"median"``, ``"minimum"``, ``"reflect"``, ``"symmetric"``, ``"wrap"``, ``"empty"``}
                available modes for PyTorch Tensor: {``"constant"``, ``"reflect"``, ``"replicate"``, ``"circular"``}.
                One of the listed string values or a user supplied function. Defaults to ``"constant"``.