Args: img: channel first array, must have shape: (num_channels, H[, W, ..., ]). mode: {``"nearest"``, ``"nearest-exact"``, ``"linear"``, ``"bilinear"``, ``"bicubic"``, ``"trilinear"``, ``"area"``} The interpolation mode. Defaults to ``
(
self,
img: torch.Tensor,
mode: str | None = None,
padding_mode: str | None = None,
align_corners: bool | None = None,
dtype: DtypeLike | torch.dtype = None,
lazy: bool | None = None,
)
| 1101 | self.kwargs = kwargs |
| 1102 | |
| 1103 | def __call__( |
| 1104 | self, |
| 1105 | img: torch.Tensor, |
| 1106 | mode: str | None = None, |
| 1107 | padding_mode: str | None = None, |
| 1108 | align_corners: bool | None = None, |
| 1109 | dtype: DtypeLike | torch.dtype = None, |
| 1110 | lazy: bool | None = None, |
| 1111 | ) -> torch.Tensor: |
| 1112 | """ |
| 1113 | Args: |
| 1114 | img: channel first array, must have shape: (num_channels, H[, W, ..., ]). |
| 1115 | mode: {``"nearest"``, ``"nearest-exact"``, ``"linear"``, |
| 1116 | ``"bilinear"``, ``"bicubic"``, ``"trilinear"``, ``"area"``} |
| 1117 | The interpolation mode. Defaults to ``self.mode``. |
| 1118 | See also: https://pytorch.org/docs/stable/generated/torch.nn.functional.interpolate.html |
| 1119 | padding_mode: available modes for numpy array:{``"constant"``, ``"edge"``, ``"linear_ramp"``, ``"maximum"``, |
| 1120 | ``"mean"``, ``"median"``, ``"minimum"``, ``"reflect"``, ``"symmetric"``, ``"wrap"``, ``"empty"``} |
| 1121 | available modes for PyTorch Tensor: {``"constant"``, ``"reflect"``, ``"replicate"``, ``"circular"``}. |
| 1122 | One of the listed string values or a user supplied function. Defaults to ``"edge"``. |
| 1123 | The mode to pad data after zooming. |
| 1124 | See also: https://numpy.org/doc/1.18/reference/generated/numpy.pad.html |
| 1125 | https://pytorch.org/docs/stable/generated/torch.nn.functional.pad.html |
| 1126 | align_corners: This only has an effect when mode is |
| 1127 | 'linear', 'bilinear', 'bicubic' or 'trilinear'. Defaults to ``self.align_corners``. |
| 1128 | See also: https://pytorch.org/docs/stable/generated/torch.nn.functional.interpolate.html |
| 1129 | dtype: data type for resampling computation. Defaults to ``self.dtype``. |
| 1130 | If None, use the data type of input data. |
| 1131 | lazy: a flag to indicate whether this transform should execute lazily or not |
| 1132 | during this call. Setting this to False or True overrides the ``lazy`` flag set |
| 1133 | during initialization for this call. Defaults to None. |
| 1134 | """ |
| 1135 | img = convert_to_tensor(img, track_meta=get_track_meta()) |
| 1136 | _zoom = ensure_tuple_rep(self.zoom, img.ndim - 1) # match the spatial image dim |
| 1137 | _mode = self.mode if mode is None else mode |
| 1138 | _padding_mode = padding_mode or self.padding_mode |
| 1139 | _align_corners = self.align_corners if align_corners is None else align_corners |
| 1140 | _dtype = get_equivalent_dtype(dtype or self.dtype or img.dtype, torch.Tensor) |
| 1141 | lazy_ = self.lazy if lazy is None else lazy |
| 1142 | return zoom( # type: ignore |
| 1143 | img, |
| 1144 | _zoom, |
| 1145 | self.keep_size, |
| 1146 | _mode, |
| 1147 | _padding_mode, |
| 1148 | _align_corners, |
| 1149 | _dtype, |
| 1150 | lazy=lazy_, |
| 1151 | transform_info=self.get_transform_info(), |
| 1152 | **self.kwargs, |
| 1153 | ) |
| 1154 | |
| 1155 | def inverse(self, data: torch.Tensor) -> torch.Tensor: |
| 1156 | transform = self.pop_transform(data) |
nothing calls this directly
no test coverage detected