Method __call__

monai/transforms/croppad/array.py:132–166

Args: img: data to be transformed, assuming `img` is channel-first and padding doesn't apply to the channel dim. to_pad: the amount to be padded in each dimension [(low_H, high_H), (low_W, high_W), ...]. default to `self.to_pad`. mode: ava

(  # type: ignore[override]
        self,
        img: torch.Tensor,
        to_pad: tuple[tuple[int, int]] | None = None,
        mode: str | None = None,
        lazy: bool | None = None,
        **kwargs,
    )

        raise NotImplementedError(f"subclass {self.__class__.__name__} must implement this method.")

    def __call__(  # type: ignore[override]
        self,
        img: torch.Tensor,
        to_pad: tuple[tuple[int, int]] | None = None,
        mode: str | None = None,
        lazy: bool | None = None,
        **kwargs,
    ) -> torch.Tensor:
        """
        Args:
            img: data to be transformed, assuming `img` is channel-first and padding doesn't apply to the channel dim.
            to_pad: the amount to be padded in each dimension [(low_H, high_H), (low_W, high_W), ...].
                default to `self.to_pad`.
            mode: available modes: (Numpy) {``"constant"``, ``"edge"``, ``"linear_ramp"``, ``"maximum"``,
                ``"mean"``, ``"median"``, ``"minimum"``, ``"reflect"``, ``"symmetric"``, ``"wrap"``, ``"empty"``}
                (PyTorch) {``"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
            lazy: a flag to override the lazy behaviour for this call, if set. Defaults to None.
            kwargs: other arguments for the `np.pad` or `torch.pad` function.
                note that `np.pad` treats channel dimension as the first dimension.

        """
        to_pad_ = self.to_pad if to_pad is None else to_pad
        if to_pad_ is None:
            spatial_shape = img.peek_pending_shape() if isinstance(img, MetaTensor) else img.shape[1:]
            to_pad_ = self.compute_pad_width(spatial_shape)
        mode_ = self.mode if mode is None else mode
        kwargs_ = dict(self.kwargs)
        kwargs_.update(kwargs)

        img_t = convert_to_tensor(data=img, track_meta=get_track_meta())
        lazy_ = self.lazy if lazy is None else lazy
        return pad_func(img_t, to_pad_, self.get_transform_info(), mode_, lazy_, **kwargs_)

    def inverse(self, data: MetaTensor) -> MetaTensor:
        transform = self.pop_transform(data)