# Copyright 2020 - 2021 MONAI Consortium
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
# http://www.apache.org/licenses/LICENSE-2.0
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""
A collection of "vanilla" transforms for spatial operations
https://github.com/Project-MONAI/MONAI/wiki/MONAI_Design
"""
import warnings
from typing import Any, List, Optional, Sequence, Tuple, Union
import numpy as np
import torch
from monai.config import USE_COMPILED, DtypeLike
from monai.data.utils import compute_shape_offset, to_affine_nd, zoom_affine
from monai.networks.layers import AffineTransform, GaussianFilter, grid_pull
from monai.transforms.croppad.array import CenterSpatialCrop
from monai.transforms.transform import Randomizable, RandomizableTransform, Transform
from monai.transforms.utils import (
create_control_grid,
create_grid,
create_rotate,
create_scale,
create_shear,
create_translate,
map_spatial_axes,
)
from monai.utils import (
GridSampleMode,
GridSamplePadMode,
InterpolateMode,
NumpyPadMode,
ensure_tuple,
ensure_tuple_rep,
ensure_tuple_size,
fall_back_tuple,
issequenceiterable,
optional_import,
)
nib, _ = optional_import("nibabel")
__all__ = [
"Spacing",
"Orientation",
"Flip",
"Resize",
"Rotate",
"Zoom",
"Rotate90",
"RandRotate90",
"RandRotate",
"RandFlip",
"RandAxisFlip",
"RandZoom",
"AffineGrid",
"RandAffineGrid",
"RandDeformGrid",
"Resample",
"Affine",
"RandAffine",
"Rand2DElastic",
"Rand3DElastic",
]
RandRange = Optional[Union[Sequence[Union[Tuple[float, float], float]], float]]
[docs]class Spacing(Transform):
"""
Resample input image into the specified `pixdim`.
"""
def __init__(
self,
pixdim: Union[Sequence[float], float],
diagonal: bool = False,
mode: Union[GridSampleMode, str] = GridSampleMode.BILINEAR,
padding_mode: Union[GridSamplePadMode, str] = GridSamplePadMode.BORDER,
align_corners: bool = False,
dtype: DtypeLike = np.float64,
) -> None:
"""
Args:
pixdim: output voxel spacing.
diagonal: whether to resample the input to have a diagonal affine matrix.
If True, the input data is resampled to the following affine::
np.diag((pixdim_0, pixdim_1, ..., pixdim_n, 1))
This effectively resets the volume to the world coordinate system (RAS+ in nibabel).
The original orientation, rotation, shearing are not preserved.
If False, this transform preserves the axes orientation, orthogonal rotation and
translation components from the original affine. This option will not flip/swap axes
of the original data.
mode: {``"bilinear"``, ``"nearest"``}
Interpolation mode to calculate output values. Defaults to ``"bilinear"``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
padding_mode: {``"zeros"``, ``"border"``, ``"reflection"``}
Padding mode for outside grid values. Defaults to ``"border"``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
align_corners: Geometrically, we consider the pixels of the input as squares rather than points.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
dtype: data type for resampling computation. Defaults to ``np.float64`` for best precision.
If None, use the data type of input data. To be compatible with other modules,
the output data type is always ``np.float32``.
"""
self.pixdim = np.array(ensure_tuple(pixdim), dtype=np.float64)
self.diagonal = diagonal
self.mode: GridSampleMode = GridSampleMode(mode)
self.padding_mode: GridSamplePadMode = GridSamplePadMode(padding_mode)
self.align_corners = align_corners
self.dtype = dtype
[docs] def __call__(
self,
data_array: np.ndarray,
affine: Optional[np.ndarray] = None,
mode: Optional[Union[GridSampleMode, str]] = None,
padding_mode: Optional[Union[GridSamplePadMode, str]] = None,
align_corners: Optional[bool] = None,
dtype: DtypeLike = None,
output_spatial_shape: Optional[np.ndarray] = None,
) -> Tuple[np.ndarray, np.ndarray, np.ndarray]:
"""
Args:
data_array: in shape (num_channels, H[, W, ...]).
affine (matrix): (N+1)x(N+1) original affine matrix for spatially ND `data_array`. Defaults to identity.
mode: {``"bilinear"``, ``"nearest"``}
Interpolation mode to calculate output values. Defaults to ``self.mode``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
padding_mode: {``"zeros"``, ``"border"``, ``"reflection"``}
Padding mode for outside grid values. Defaults to ``self.padding_mode``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
align_corners: Geometrically, we consider the pixels of the input as squares rather than points.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
dtype: data type for resampling computation. Defaults to ``self.dtype``.
If None, use the data type of input data. To be compatible with other modules,
the output data type is always ``np.float32``.
output_spatial_shape: specify the shape of the output data_array. This is typically useful for
the inverse of `Spacingd` where sometimes we could not compute the exact shape due to the quantization
error with the affines.
Raises:
ValueError: When ``data_array`` has no spatial dimensions.
ValueError: When ``pixdim`` is nonpositive.
Returns:
data_array (resampled into `self.pixdim`), original affine, current affine.
"""
_dtype = dtype or self.dtype or data_array.dtype
sr = data_array.ndim - 1
if sr <= 0:
raise ValueError("data_array must have at least one spatial dimension.")
if affine is None:
# default to identity
affine = np.eye(sr + 1, dtype=np.float64)
affine_ = np.eye(sr + 1, dtype=np.float64)
else:
affine_ = to_affine_nd(sr, affine)
out_d = self.pixdim[:sr]
if out_d.size < sr:
out_d = np.append(out_d, [1.0] * (out_d.size - sr))
if np.any(out_d <= 0):
raise ValueError(f"pixdim must be positive, got {out_d}.")
# compute output affine, shape and offset
new_affine = zoom_affine(affine_, out_d, diagonal=self.diagonal)
output_shape, offset = compute_shape_offset(data_array.shape[1:], affine_, new_affine)
new_affine[:sr, -1] = offset[:sr]
transform = np.linalg.inv(affine_) @ new_affine
# adapt to the actual rank
transform = to_affine_nd(sr, transform)
# no resampling if it's identity transform
if np.allclose(transform, np.diag(np.ones(len(transform))), atol=1e-3):
output_data = data_array.copy().astype(np.float32)
new_affine = to_affine_nd(affine, new_affine)
return output_data, affine, new_affine
# resample
affine_xform = AffineTransform(
normalized=False,
mode=mode or self.mode,
padding_mode=padding_mode or self.padding_mode,
align_corners=self.align_corners if align_corners is None else align_corners,
reverse_indexing=True,
)
output_data = affine_xform(
# AffineTransform requires a batch dim
torch.as_tensor(np.ascontiguousarray(data_array).astype(_dtype)).unsqueeze(0),
torch.as_tensor(np.ascontiguousarray(transform).astype(_dtype)),
spatial_size=output_shape if output_spatial_shape is None else output_spatial_shape,
)
output_data = np.asarray(output_data.squeeze(0).detach().cpu().numpy(), dtype=np.float32) # type: ignore
new_affine = to_affine_nd(affine, new_affine)
return output_data, affine, new_affine
[docs]class Orientation(Transform):
"""
Change the input image's orientation into the specified based on `axcodes`.
"""
def __init__(
self,
axcodes: Optional[str] = None,
as_closest_canonical: bool = False,
labels: Optional[Sequence[Tuple[str, str]]] = tuple(zip("LPI", "RAS")),
) -> None:
"""
Args:
axcodes: N elements sequence for spatial ND input's orientation.
e.g. axcodes='RAS' represents 3D orientation:
(Left, Right), (Posterior, Anterior), (Inferior, Superior).
default orientation labels options are: 'L' and 'R' for the first dimension,
'P' and 'A' for the second, 'I' and 'S' for the third.
as_closest_canonical: if True, load the image as closest to canonical axis format.
labels: optional, None or sequence of (2,) sequences
(2,) sequences are labels for (beginning, end) of output axis.
Defaults to ``(('L', 'R'), ('P', 'A'), ('I', 'S'))``.
Raises:
ValueError: When ``axcodes=None`` and ``as_closest_canonical=True``. Incompatible values.
See Also: `nibabel.orientations.ornt2axcodes`.
"""
if axcodes is None and not as_closest_canonical:
raise ValueError("Incompatible values: axcodes=None and as_closest_canonical=True.")
if axcodes is not None and as_closest_canonical:
warnings.warn("using as_closest_canonical=True, axcodes ignored.")
self.axcodes = axcodes
self.as_closest_canonical = as_closest_canonical
self.labels = labels
[docs] def __call__(
self, data_array: np.ndarray, affine: Optional[np.ndarray] = None
) -> Tuple[np.ndarray, np.ndarray, np.ndarray]:
"""
original orientation of `data_array` is defined by `affine`.
Args:
data_array: in shape (num_channels, H[, W, ...]).
affine (matrix): (N+1)x(N+1) original affine matrix for spatially ND `data_array`. Defaults to identity.
Raises:
ValueError: When ``data_array`` has no spatial dimensions.
ValueError: When ``axcodes`` spatiality differs from ``data_array``.
Returns:
data_array (reoriented in `self.axcodes`), original axcodes, current axcodes.
"""
sr = data_array.ndim - 1
if sr <= 0:
raise ValueError("data_array must have at least one spatial dimension.")
if affine is None:
affine = np.eye(sr + 1, dtype=np.float64)
affine_ = np.eye(sr + 1, dtype=np.float64)
else:
affine_ = to_affine_nd(sr, affine)
src = nib.io_orientation(affine_)
if self.as_closest_canonical:
spatial_ornt = src
else:
if self.axcodes is None:
raise AssertionError
dst = nib.orientations.axcodes2ornt(self.axcodes[:sr], labels=self.labels)
if len(dst) < sr:
raise ValueError(
f"axcodes must match data_array spatially, got axcodes={len(self.axcodes)}D data_array={sr}D"
)
spatial_ornt = nib.orientations.ornt_transform(src, dst)
ornt = spatial_ornt.copy()
ornt[:, 0] += 1 # skip channel dim
ornt = np.concatenate([np.array([[0, 1]]), ornt])
shape = data_array.shape[1:]
data_array = np.ascontiguousarray(nib.orientations.apply_orientation(data_array, ornt))
new_affine = affine_ @ nib.orientations.inv_ornt_aff(spatial_ornt, shape)
new_affine = to_affine_nd(affine, new_affine)
return data_array, affine, new_affine
[docs]class Flip(Transform):
"""
Reverses the order of elements along the given spatial axis. Preserves shape.
Uses ``np.flip`` in practice. See numpy.flip for additional details:
https://docs.scipy.org/doc/numpy/reference/generated/numpy.flip.html.
Args:
spatial_axis: spatial axes along which to flip over. Default is None.
The default `axis=None` will flip over all of the axes of the input array.
If axis is negative it counts from the last to the first axis.
If axis is a tuple of ints, flipping is performed on all of the axes
specified in the tuple.
"""
def __init__(self, spatial_axis: Optional[Union[Sequence[int], int]] = None) -> None:
self.spatial_axis = spatial_axis
[docs] def __call__(self, img: np.ndarray) -> np.ndarray:
"""
Args:
img: channel first array, must have shape: (num_channels, H[, W, ..., ]),
"""
result: np.ndarray = np.flip(img, map_spatial_axes(img.ndim, self.spatial_axis))
return result.astype(img.dtype)
[docs]class Resize(Transform):
"""
Resize the input image to given spatial size (with scaling, not cropping/padding).
Implemented using :py:class:`torch.nn.functional.interpolate`.
Args:
spatial_size: expected shape of spatial dimensions after resize operation.
if the components of the `spatial_size` are non-positive values, the transform will use the
corresponding components of img size. For example, `spatial_size=(32, -1)` will be adapted
to `(32, 64)` if the second spatial dimension size of img is `64`.
mode: {``"nearest"``, ``"linear"``, ``"bilinear"``, ``"bicubic"``, ``"trilinear"``, ``"area"``}
The interpolation mode. Defaults to ``"area"``.
See also: https://pytorch.org/docs/stable/nn.functional.html#interpolate
align_corners: This only has an effect when mode is
'linear', 'bilinear', 'bicubic' or 'trilinear'. Default: None.
See also: https://pytorch.org/docs/stable/nn.functional.html#interpolate
"""
def __init__(
self,
spatial_size: Union[Sequence[int], int],
mode: Union[InterpolateMode, str] = InterpolateMode.AREA,
align_corners: Optional[bool] = None,
) -> None:
self.spatial_size = ensure_tuple(spatial_size)
self.mode: InterpolateMode = InterpolateMode(mode)
self.align_corners = align_corners
[docs] def __call__(
self,
img: np.ndarray,
mode: Optional[Union[InterpolateMode, str]] = None,
align_corners: Optional[bool] = None,
) -> np.ndarray:
"""
Args:
img: channel first array, must have shape: (num_channels, H[, W, ..., ]).
mode: {``"nearest"``, ``"linear"``, ``"bilinear"``, ``"bicubic"``, ``"trilinear"``, ``"area"``}
The interpolation mode. Defaults to ``self.mode``.
See also: https://pytorch.org/docs/stable/nn.functional.html#interpolate
align_corners: This only has an effect when mode is
'linear', 'bilinear', 'bicubic' or 'trilinear'. Defaults to ``self.align_corners``.
See also: https://pytorch.org/docs/stable/nn.functional.html#interpolate
Raises:
ValueError: When ``self.spatial_size`` length is less than ``img`` spatial dimensions.
"""
input_ndim = img.ndim - 1 # spatial ndim
output_ndim = len(self.spatial_size)
if output_ndim > input_ndim:
input_shape = ensure_tuple_size(img.shape, output_ndim + 1, 1)
img = img.reshape(input_shape)
elif output_ndim < input_ndim:
raise ValueError(
"len(spatial_size) must be greater or equal to img spatial dimensions, "
f"got spatial_size={output_ndim} img={input_ndim}."
)
spatial_size = fall_back_tuple(self.spatial_size, img.shape[1:])
resized = torch.nn.functional.interpolate( # type: ignore
input=torch.as_tensor(np.ascontiguousarray(img), dtype=torch.float).unsqueeze(0),
size=spatial_size,
mode=self.mode.value if mode is None else InterpolateMode(mode).value,
align_corners=self.align_corners if align_corners is None else align_corners,
)
resized = resized.squeeze(0).detach().cpu().numpy()
return np.asarray(resized)
[docs]class Rotate(Transform):
"""
Rotates an input image by given angle using :py:class:`monai.networks.layers.AffineTransform`.
Args:
angle: Rotation angle(s) in radians. should a float for 2D, three floats for 3D.
keep_size: If it is True, the output shape is kept the same as the input.
If it is False, the output shape is adapted so that the
input array is contained completely in the output. Default is True.
mode: {``"bilinear"``, ``"nearest"``}
Interpolation mode to calculate output values. Defaults to ``"bilinear"``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
padding_mode: {``"zeros"``, ``"border"``, ``"reflection"``}
Padding mode for outside grid values. Defaults to ``"border"``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
align_corners: Defaults to False.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
dtype: data type for resampling computation. Defaults to ``np.float64`` for best precision.
If None, use the data type of input data. To be compatible with other modules,
the output data type is always ``np.float32``.
"""
def __init__(
self,
angle: Union[Sequence[float], float],
keep_size: bool = True,
mode: Union[GridSampleMode, str] = GridSampleMode.BILINEAR,
padding_mode: Union[GridSamplePadMode, str] = GridSamplePadMode.BORDER,
align_corners: bool = False,
dtype: DtypeLike = np.float64,
) -> None:
self.angle = angle
self.keep_size = keep_size
self.mode: GridSampleMode = GridSampleMode(mode)
self.padding_mode: GridSamplePadMode = GridSamplePadMode(padding_mode)
self.align_corners = align_corners
self.dtype = dtype
self.rotation_matrix: Optional[np.ndarray] = None
[docs] def __call__(
self,
img: np.ndarray,
mode: Optional[Union[GridSampleMode, str]] = None,
padding_mode: Optional[Union[GridSamplePadMode, str]] = None,
align_corners: Optional[bool] = None,
dtype: DtypeLike = None,
) -> np.ndarray:
"""
Args:
img: channel first array, must have shape: [chns, H, W] or [chns, H, W, D].
mode: {``"bilinear"``, ``"nearest"``}
Interpolation mode to calculate output values. Defaults to ``self.mode``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
padding_mode: {``"zeros"``, ``"border"``, ``"reflection"``}
Padding mode for outside grid values. Defaults to ``self.padding_mode``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
align_corners: Defaults to ``self.align_corners``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
align_corners: Defaults to ``self.align_corners``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
dtype: data type for resampling computation. Defaults to ``self.dtype``.
If None, use the data type of input data. To be compatible with other modules,
the output data type is always ``np.float32``.
Raises:
ValueError: When ``img`` spatially is not one of [2D, 3D].
"""
_dtype = dtype or self.dtype or img.dtype
im_shape = np.asarray(img.shape[1:]) # spatial dimensions
input_ndim = len(im_shape)
if input_ndim not in (2, 3):
raise ValueError(f"Unsupported img dimension: {input_ndim}, available options are [2, 3].")
_angle = ensure_tuple_rep(self.angle, 1 if input_ndim == 2 else 3)
transform = create_rotate(input_ndim, _angle)
shift = create_translate(input_ndim, (im_shape - 1) / 2)
if self.keep_size:
output_shape = im_shape
else:
corners = np.asarray(np.meshgrid(*[(0, dim) for dim in im_shape], indexing="ij")).reshape(
(len(im_shape), -1)
)
corners = transform[:-1, :-1] @ corners
output_shape = np.asarray(corners.ptp(axis=1) + 0.5, dtype=int)
shift_1 = create_translate(input_ndim, -(output_shape - 1) / 2)
transform = shift @ transform @ shift_1
xform = AffineTransform(
normalized=False,
mode=mode or self.mode,
padding_mode=padding_mode or self.padding_mode,
align_corners=self.align_corners if align_corners is None else align_corners,
reverse_indexing=True,
)
output = xform(
torch.as_tensor(np.ascontiguousarray(img).astype(_dtype)).unsqueeze(0),
torch.as_tensor(np.ascontiguousarray(transform).astype(_dtype)),
spatial_size=output_shape,
)
self.rotation_matrix = transform
return np.asarray(output.squeeze(0).detach().cpu().numpy(), dtype=np.float32)
[docs] def get_rotation_matrix(self) -> Optional[np.ndarray]:
"""Get the most recently applied rotation matrix"""
return self.rotation_matrix
[docs]class Zoom(Transform):
"""
Zooms an ND image using :py:class:`torch.nn.functional.interpolate`.
For details, please see https://pytorch.org/docs/stable/nn.functional.html#interpolate.
Different from :py:class:`monai.transforms.resize`, this transform takes scaling factors
as input, and provides an option of preserving the input spatial size.
Args:
zoom: The zoom factor along the spatial axes.
If a float, zoom is the same for each spatial axis.
If a sequence, zoom should contain one value for each spatial axis.
mode: {``"nearest"``, ``"linear"``, ``"bilinear"``, ``"bicubic"``, ``"trilinear"``, ``"area"``}
The interpolation mode. Defaults to ``"area"``.
See also: https://pytorch.org/docs/stable/nn.functional.html#interpolate
padding_mode: {``"constant"``, ``"edge``", ``"linear_ramp``", ``"maximum``", ``"mean``", `"median``",
``"minimum``", `"reflect``", ``"symmetric``", ``"wrap``", ``"empty``", ``"<function>``"}
The mode to pad data after zooming.
See also: https://numpy.org/doc/stable/reference/generated/numpy.pad.html
align_corners: This only has an effect when mode is
'linear', 'bilinear', 'bicubic' or 'trilinear'. Default: None.
See also: https://pytorch.org/docs/stable/nn.functional.html#interpolate
keep_size: Should keep original size (padding/slicing if needed), default is True.
"""
def __init__(
self,
zoom: Union[Sequence[float], float],
mode: Union[InterpolateMode, str] = InterpolateMode.AREA,
padding_mode: Union[NumpyPadMode, str] = NumpyPadMode.EDGE,
align_corners: Optional[bool] = None,
keep_size: bool = True,
) -> None:
self.zoom = zoom
self.mode: InterpolateMode = InterpolateMode(mode)
self.padding_mode: NumpyPadMode = NumpyPadMode(padding_mode)
self.align_corners = align_corners
self.keep_size = keep_size
[docs] def __call__(
self,
img: np.ndarray,
mode: Optional[Union[InterpolateMode, str]] = None,
padding_mode: Optional[Union[NumpyPadMode, str]] = None,
align_corners: Optional[bool] = None,
):
"""
Args:
img: channel first array, must have shape: (num_channels, H[, W, ..., ]).
mode: {``"nearest"``, ``"linear"``, ``"bilinear"``, ``"bicubic"``, ``"trilinear"``, ``"area"``}
The interpolation mode. Defaults to ``self.mode``.
See also: https://pytorch.org/docs/stable/nn.functional.html#interpolate
padding_mode: {``"constant"``, ``"edge``", ``"linear_ramp``", ``"maximum``", ``"mean``", `"median``",
``"minimum``", `"reflect``", ``"symmetric``", ``"wrap``", ``"empty``", ``"<function>``"}
The mode to pad data after zooming, default to ``self.padding_mode``.
See also: https://numpy.org/doc/stable/reference/generated/numpy.pad.html
align_corners: This only has an effect when mode is
'linear', 'bilinear', 'bicubic' or 'trilinear'. Defaults to ``self.align_corners``.
See also: https://pytorch.org/docs/stable/nn.functional.html#interpolate
"""
_zoom = ensure_tuple_rep(self.zoom, img.ndim - 1) # match the spatial image dim
zoomed = torch.nn.functional.interpolate( # type: ignore
recompute_scale_factor=True,
input=torch.as_tensor(np.ascontiguousarray(img), dtype=torch.float).unsqueeze(0),
scale_factor=list(_zoom),
mode=self.mode.value if mode is None else InterpolateMode(mode).value,
align_corners=self.align_corners if align_corners is None else align_corners,
)
zoomed = zoomed.squeeze(0).detach().cpu().numpy()
if not self.keep_size or np.allclose(img.shape, zoomed.shape):
return zoomed
pad_vec = [[0, 0]] * len(img.shape)
slice_vec = [slice(None)] * len(img.shape)
for idx, (od, zd) in enumerate(zip(img.shape, zoomed.shape)):
diff = od - zd
half = abs(diff) // 2
if diff > 0: # need padding
pad_vec[idx] = [half, diff - half]
elif diff < 0: # need slicing
slice_vec[idx] = slice(half, half + od)
padding_mode = self.padding_mode if padding_mode is None else NumpyPadMode(padding_mode)
zoomed = np.pad(zoomed, pad_vec, mode=padding_mode.value)
return zoomed[tuple(slice_vec)]
[docs]class Rotate90(Transform):
"""
Rotate an array by 90 degrees in the plane specified by `axes`.
See np.rot90 for additional details:
https://numpy.org/doc/stable/reference/generated/numpy.rot90.html.
"""
def __init__(self, k: int = 1, spatial_axes: Tuple[int, int] = (0, 1)) -> None:
"""
Args:
k: number of times to rotate by 90 degrees.
spatial_axes: 2 int numbers, defines the plane to rotate with 2 spatial axes.
Default: (0, 1), this is the first two axis in spatial dimensions.
If axis is negative it counts from the last to the first axis.
"""
self.k = k
spatial_axes_: Tuple[int, int] = ensure_tuple(spatial_axes) # type: ignore
if len(spatial_axes_) != 2:
raise ValueError("spatial_axes must be 2 int numbers to indicate the axes to rotate 90 degrees.")
self.spatial_axes = spatial_axes_
[docs] def __call__(self, img: np.ndarray) -> np.ndarray:
"""
Args:
img: channel first array, must have shape: (num_channels, H[, W, ..., ]),
"""
result: np.ndarray = np.rot90(img, self.k, map_spatial_axes(img.ndim, self.spatial_axes))
return result.astype(img.dtype)
[docs]class RandRotate90(RandomizableTransform):
"""
With probability `prob`, input arrays are rotated by 90 degrees
in the plane specified by `spatial_axes`.
"""
def __init__(self, prob: float = 0.1, max_k: int = 3, spatial_axes: Tuple[int, int] = (0, 1)) -> None:
"""
Args:
prob: probability of rotating.
(Default 0.1, with 10% probability it returns a rotated array)
max_k: number of rotations will be sampled from `np.random.randint(max_k) + 1`, (Default 3).
spatial_axes: 2 int numbers, defines the plane to rotate with 2 spatial axes.
Default: (0, 1), this is the first two axis in spatial dimensions.
"""
RandomizableTransform.__init__(self, prob)
self.max_k = max_k
self.spatial_axes = spatial_axes
self._rand_k = 0
[docs] def randomize(self, data: Optional[Any] = None) -> None:
self._rand_k = self.R.randint(self.max_k) + 1
super().randomize(None)
[docs] def __call__(self, img: np.ndarray) -> np.ndarray:
"""
Args:
img: channel first array, must have shape: (num_channels, H[, W, ..., ]),
"""
self.randomize()
if not self._do_transform:
return img
rotator = Rotate90(self._rand_k, self.spatial_axes)
return rotator(img)
[docs]class RandRotate(RandomizableTransform):
"""
Randomly rotate the input arrays.
Args:
range_x: Range of rotation angle in radians in the plane defined by the first and second axes.
If single number, angle is uniformly sampled from (-range_x, range_x).
range_y: Range of rotation angle in radians in the plane defined by the first and third axes.
If single number, angle is uniformly sampled from (-range_y, range_y).
range_z: Range of rotation angle in radians in the plane defined by the second and third axes.
If single number, angle is uniformly sampled from (-range_z, range_z).
prob: Probability of rotation.
keep_size: If it is False, the output shape is adapted so that the
input array is contained completely in the output.
If it is True, the output shape is the same as the input. Default is True.
mode: {``"bilinear"``, ``"nearest"``}
Interpolation mode to calculate output values. Defaults to ``"bilinear"``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
padding_mode: {``"zeros"``, ``"border"``, ``"reflection"``}
Padding mode for outside grid values. Defaults to ``"border"``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
align_corners: Defaults to False.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
dtype: data type for resampling computation. Defaults to ``np.float64`` for best precision.
If None, use the data type of input data. To be compatible with other modules,
the output data type is always ``np.float32``.
"""
def __init__(
self,
range_x: Union[Tuple[float, float], float] = 0.0,
range_y: Union[Tuple[float, float], float] = 0.0,
range_z: Union[Tuple[float, float], float] = 0.0,
prob: float = 0.1,
keep_size: bool = True,
mode: Union[GridSampleMode, str] = GridSampleMode.BILINEAR,
padding_mode: Union[GridSamplePadMode, str] = GridSamplePadMode.BORDER,
align_corners: bool = False,
dtype: DtypeLike = np.float64,
) -> None:
RandomizableTransform.__init__(self, prob)
self.range_x = ensure_tuple(range_x)
if len(self.range_x) == 1:
self.range_x = tuple(sorted([-self.range_x[0], self.range_x[0]]))
self.range_y = ensure_tuple(range_y)
if len(self.range_y) == 1:
self.range_y = tuple(sorted([-self.range_y[0], self.range_y[0]]))
self.range_z = ensure_tuple(range_z)
if len(self.range_z) == 1:
self.range_z = tuple(sorted([-self.range_z[0], self.range_z[0]]))
self.keep_size = keep_size
self.mode: GridSampleMode = GridSampleMode(mode)
self.padding_mode: GridSamplePadMode = GridSamplePadMode(padding_mode)
self.align_corners = align_corners
self.dtype = dtype
self.x = 0.0
self.y = 0.0
self.z = 0.0
[docs] def randomize(self, data: Optional[Any] = None) -> None:
super().randomize(None)
self.x = self.R.uniform(low=self.range_x[0], high=self.range_x[1])
self.y = self.R.uniform(low=self.range_y[0], high=self.range_y[1])
self.z = self.R.uniform(low=self.range_z[0], high=self.range_z[1])
[docs] def __call__(
self,
img: np.ndarray,
mode: Optional[Union[GridSampleMode, str]] = None,
padding_mode: Optional[Union[GridSamplePadMode, str]] = None,
align_corners: Optional[bool] = None,
dtype: DtypeLike = None,
) -> np.ndarray:
"""
Args:
img: channel first array, must have shape 2D: (nchannels, H, W), or 3D: (nchannels, H, W, D).
mode: {``"bilinear"``, ``"nearest"``}
Interpolation mode to calculate output values. Defaults to ``self.mode``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
padding_mode: {``"zeros"``, ``"border"``, ``"reflection"``}
Padding mode for outside grid values. Defaults to ``self.padding_mode``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
align_corners: Defaults to ``self.align_corners``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
dtype: data type for resampling computation. Defaults to ``self.dtype``.
If None, use the data type of input data. To be compatible with other modules,
the output data type is always ``np.float32``.
"""
self.randomize()
if not self._do_transform:
return img
rotator = Rotate(
angle=self.x if img.ndim == 3 else (self.x, self.y, self.z),
keep_size=self.keep_size,
mode=mode or self.mode,
padding_mode=padding_mode or self.padding_mode,
align_corners=self.align_corners if align_corners is None else align_corners,
dtype=dtype or self.dtype or img.dtype,
)
return np.array(rotator(img))
[docs]class RandFlip(RandomizableTransform):
"""
Randomly flips the image along axes. Preserves shape.
See numpy.flip for additional details.
https://docs.scipy.org/doc/numpy/reference/generated/numpy.flip.html
Args:
prob: Probability of flipping.
spatial_axis: Spatial axes along which to flip over. Default is None.
"""
def __init__(self, prob: float = 0.1, spatial_axis: Optional[Union[Sequence[int], int]] = None) -> None:
RandomizableTransform.__init__(self, prob)
self.flipper = Flip(spatial_axis=spatial_axis)
[docs] def __call__(self, img: np.ndarray) -> np.ndarray:
"""
Args:
img: channel first array, must have shape: (num_channels, H[, W, ..., ]),
"""
self.randomize(None)
if not self._do_transform:
return img
return self.flipper(img)
[docs]class RandAxisFlip(RandomizableTransform):
"""
Randomly select a spatial axis and flip along it.
See numpy.flip for additional details.
https://docs.scipy.org/doc/numpy/reference/generated/numpy.flip.html
Args:
prob: Probability of flipping.
"""
def __init__(self, prob: float = 0.1) -> None:
RandomizableTransform.__init__(self, prob)
self._axis: Optional[int] = None
[docs] def randomize(self, data: np.ndarray) -> None:
super().randomize(None)
self._axis = self.R.randint(data.ndim - 1)
[docs] def __call__(self, img: np.ndarray) -> np.ndarray:
"""
Args:
img: channel first array, must have shape: (num_channels, H[, W, ..., ]),
"""
self.randomize(data=img)
if not self._do_transform:
return img
flipper = Flip(spatial_axis=self._axis)
return flipper(img)
[docs]class RandZoom(RandomizableTransform):
"""
Randomly zooms input arrays with given probability within given zoom range.
Args:
prob: Probability of zooming.
min_zoom: Min zoom factor. Can be float or sequence same size as image.
If a float, select a random factor from `[min_zoom, max_zoom]` then apply to all spatial dims
to keep the original spatial shape ratio.
If a sequence, min_zoom should contain one value for each spatial axis.
If 2 values provided for 3D data, use the first value for both H & W dims to keep the same zoom ratio.
max_zoom: Max zoom factor. Can be float or sequence same size as image.
If a float, select a random factor from `[min_zoom, max_zoom]` then apply to all spatial dims
to keep the original spatial shape ratio.
If a sequence, max_zoom should contain one value for each spatial axis.
If 2 values provided for 3D data, use the first value for both H & W dims to keep the same zoom ratio.
mode: {``"nearest"``, ``"linear"``, ``"bilinear"``, ``"bicubic"``, ``"trilinear"``, ``"area"``}
The interpolation mode. Defaults to ``"area"``.
See also: https://pytorch.org/docs/stable/nn.functional.html#interpolate
padding_mode: {``"constant"``, ``"edge``", ``"linear_ramp``", ``"maximum``", ``"mean``", `"median``",
``"minimum``", `"reflect``", ``"symmetric``", ``"wrap``", ``"empty``", ``"<function>``"}
The mode to pad data after zooming.
See also: https://numpy.org/doc/stable/reference/generated/numpy.pad.html
align_corners: This only has an effect when mode is
'linear', 'bilinear', 'bicubic' or 'trilinear'. Default: None.
See also: https://pytorch.org/docs/stable/nn.functional.html#interpolate
keep_size: Should keep original size (pad if needed), default is True.
"""
def __init__(
self,
prob: float = 0.1,
min_zoom: Union[Sequence[float], float] = 0.9,
max_zoom: Union[Sequence[float], float] = 1.1,
mode: Union[InterpolateMode, str] = InterpolateMode.AREA,
padding_mode: Union[NumpyPadMode, str] = NumpyPadMode.EDGE,
align_corners: Optional[bool] = None,
keep_size: bool = True,
) -> None:
RandomizableTransform.__init__(self, prob)
self.min_zoom = ensure_tuple(min_zoom)
self.max_zoom = ensure_tuple(max_zoom)
if len(self.min_zoom) != len(self.max_zoom):
raise AssertionError("min_zoom and max_zoom must have same length.")
self.mode: InterpolateMode = InterpolateMode(mode)
self.padding_mode: NumpyPadMode = NumpyPadMode(padding_mode)
self.align_corners = align_corners
self.keep_size = keep_size
self._zoom: Sequence[float] = [1.0]
[docs] def randomize(self, data: Optional[Any] = None) -> None:
super().randomize(None)
self._zoom = [self.R.uniform(l, h) for l, h in zip(self.min_zoom, self.max_zoom)]
[docs] def __call__(
self,
img: np.ndarray,
mode: Optional[Union[InterpolateMode, str]] = None,
padding_mode: Optional[Union[NumpyPadMode, str]] = None,
align_corners: Optional[bool] = None,
) -> np.ndarray:
"""
Args:
img: channel first array, must have shape 2D: (nchannels, H, W), or 3D: (nchannels, H, W, D).
mode: {``"nearest"``, ``"linear"``, ``"bilinear"``, ``"bicubic"``, ``"trilinear"``, ``"area"``}
The interpolation mode. Defaults to ``self.mode``.
See also: https://pytorch.org/docs/stable/nn.functional.html#interpolate
padding_mode: {``"constant"``, ``"edge``", ``"linear_ramp``", ``"maximum``", ``"mean``", `"median``",
``"minimum``", `"reflect``", ``"symmetric``", ``"wrap``", ``"empty``", ``"<function>``"}
The mode to pad data after zooming, default to ``self.padding_mode``.
See also: https://numpy.org/doc/stable/reference/generated/numpy.pad.html
align_corners: This only has an effect when mode is
'linear', 'bilinear', 'bicubic' or 'trilinear'. Defaults to ``self.align_corners``.
See also: https://pytorch.org/docs/stable/nn.functional.html#interpolate
"""
# match the spatial image dim
self.randomize()
_dtype = np.float32
if not self._do_transform:
return img.astype(_dtype)
if len(self._zoom) == 1:
# to keep the spatial shape ratio, use same random zoom factor for all dims
self._zoom = ensure_tuple_rep(self._zoom[0], img.ndim - 1)
elif len(self._zoom) == 2 and img.ndim > 3:
# if 2 zoom factors provided for 3D data, use the first factor for H and W dims, second factor for D dim
self._zoom = ensure_tuple_rep(self._zoom[0], img.ndim - 2) + ensure_tuple(self._zoom[-1])
zoomer = Zoom(self._zoom, keep_size=self.keep_size)
return np.asarray(
zoomer(
img,
mode=mode or self.mode,
padding_mode=padding_mode or self.padding_mode,
align_corners=self.align_corners if align_corners is None else align_corners,
),
dtype=_dtype,
)
[docs]class AffineGrid(Transform):
"""
Affine transforms on the coordinates.
Args:
rotate_params: angle range in radians. rotate_params[0] with be used to generate the 1st rotation
parameter from `uniform[-rotate_params[0], rotate_params[0])`. Similarly, `rotate_params[1]` and
`rotate_params[2]` are used in 3D affine for the range of 2nd and 3rd axes.
shear_params: shear_params[0] with be used to generate the 1st shearing parameter from
`uniform[-shear_params[0], shear_params[0])`. Similarly, `shear_params[1]` to
`shear_params[N]` controls the range of the uniform distribution used to generate the 2nd to
N-th parameter.
translate_params : translate_params[0] with be used to generate the 1st shift parameter from
`uniform[-translate_params[0], translate_params[0])`. Similarly, `translate_params[1]`
to `translate_params[N]` controls the range of the uniform distribution used to generate
the 2nd to N-th parameter.
scale_params: scale_params[0] with be used to generate the 1st scaling factor from
`uniform[-scale_params[0], scale_params[0]) + 1.0`. Similarly, `scale_params[1]` to
`scale_params[N]` controls the range of the uniform distribution used to generate the 2nd to
N-th parameter.
as_tensor_output: whether to output tensor instead of numpy array.
defaults to True.
device: device to store the output grid data.
affine: If applied, ignore the params (`rotate_params`, etc.) and use the
supplied matrix. Should be square with each side = num of image spatial
dimensions + 1.
"""
def __init__(
self,
rotate_params: Optional[Union[Sequence[float], float]] = None,
shear_params: Optional[Union[Sequence[float], float]] = None,
translate_params: Optional[Union[Sequence[float], float]] = None,
scale_params: Optional[Union[Sequence[float], float]] = None,
as_tensor_output: bool = True,
device: Optional[torch.device] = None,
affine: Optional[Union[np.ndarray, torch.Tensor]] = None,
) -> None:
self.rotate_params = rotate_params
self.shear_params = shear_params
self.translate_params = translate_params
self.scale_params = scale_params
self.as_tensor_output = as_tensor_output
self.device = device
self.affine = affine
[docs] def __call__(
self,
spatial_size: Optional[Sequence[int]] = None,
grid: Optional[Union[np.ndarray, torch.Tensor]] = None,
) -> Tuple[Union[np.ndarray, torch.Tensor], Union[np.ndarray, torch.Tensor]]:
"""
Args:
spatial_size: output grid size.
grid: grid to be transformed. Shape must be (3, H, W) for 2D or (4, H, W, D) for 3D.
Raises:
ValueError: When ``grid=None`` and ``spatial_size=None``. Incompatible values.
"""
if grid is None:
if spatial_size is not None:
grid = create_grid(spatial_size)
else:
raise ValueError("Incompatible values: grid=None and spatial_size=None.")
if self.affine is None:
spatial_dims = len(grid.shape) - 1
affine = np.eye(spatial_dims + 1)
if self.rotate_params:
affine = affine @ create_rotate(spatial_dims, self.rotate_params)
if self.shear_params:
affine = affine @ create_shear(spatial_dims, self.shear_params)
if self.translate_params:
affine = affine @ create_translate(spatial_dims, self.translate_params)
if self.scale_params:
affine = affine @ create_scale(spatial_dims, self.scale_params)
else:
affine = self.affine
if isinstance(affine, np.ndarray):
affine = torch.as_tensor(np.ascontiguousarray(affine))
grid = torch.tensor(grid) if not isinstance(grid, torch.Tensor) else grid.detach().clone()
if self.device:
affine = affine.to(self.device)
grid = grid.to(self.device)
grid = (affine.float() @ grid.reshape((grid.shape[0], -1)).float()).reshape([-1] + list(grid.shape[1:]))
if grid is None or not isinstance(grid, torch.Tensor):
raise ValueError("Unknown grid.")
return grid if self.as_tensor_output else np.asarray(grid.cpu().numpy()), affine
[docs]class RandAffineGrid(Randomizable):
"""
Generate randomised affine grid.
"""
def __init__(
self,
rotate_range: RandRange = None,
shear_range: RandRange = None,
translate_range: RandRange = None,
scale_range: RandRange = None,
as_tensor_output: bool = True,
device: Optional[torch.device] = None,
) -> None:
"""
Args:
rotate_range: angle range in radians. If element `i` is iterable, then
`uniform[-rotate_range[i][0], rotate_range[i][1])` will be used to generate the rotation parameter
for the ith dimension. If not, `uniform[-rotate_range[i], rotate_range[i])` will be used. This can
be altered on a per-dimension basis. E.g., `((0,3), 1, ...)`: for dim0, rotation will be in range
`[0, 3]`, and for dim1 `[-1, 1]` will be used. Setting a single value will use `[-x, x]` for dim0
and nothing for the remaining dimensions.
shear_range: shear_range with format matching `rotate_range`.
translate_range: translate_range with format matching `rotate_range`.
scale_range: scaling_range with format matching `rotate_range`. A value of 1.0 is added to the result.
This allows 0 to correspond to no change (i.e., a scaling of 1).
as_tensor_output: whether to output tensor instead of numpy array.
defaults to True.
device: device to store the output grid data.
See also:
- :py:meth:`monai.transforms.utils.create_rotate`
- :py:meth:`monai.transforms.utils.create_shear`
- :py:meth:`monai.transforms.utils.create_translate`
- :py:meth:`monai.transforms.utils.create_scale`
"""
self.rotate_range = ensure_tuple(rotate_range)
self.shear_range = ensure_tuple(shear_range)
self.translate_range = ensure_tuple(translate_range)
self.scale_range = ensure_tuple(scale_range)
self.rotate_params: Optional[List[float]] = None
self.shear_params: Optional[List[float]] = None
self.translate_params: Optional[List[float]] = None
self.scale_params: Optional[List[float]] = None
self.as_tensor_output = as_tensor_output
self.device = device
self.affine: Optional[Union[np.ndarray, torch.Tensor]] = None
def _get_rand_param(self, param_range, add_scalar: float = 0.0):
out_param = []
for f in param_range:
if issequenceiterable(f):
if len(f) != 2:
raise ValueError("If giving range as [min,max], should only have two elements per dim.")
out_param.append(self.R.uniform(f[0], f[1]) + add_scalar)
elif f is not None:
out_param.append(self.R.uniform(-f, f) + add_scalar)
return out_param
[docs] def randomize(self, data: Optional[Any] = None) -> None:
self.rotate_params = self._get_rand_param(self.rotate_range)
self.shear_params = self._get_rand_param(self.shear_range)
self.translate_params = self._get_rand_param(self.translate_range)
self.scale_params = self._get_rand_param(self.scale_range, 1.0)
[docs] def __call__(
self,
spatial_size: Optional[Sequence[int]] = None,
grid: Optional[Union[np.ndarray, torch.Tensor]] = None,
) -> Union[np.ndarray, torch.Tensor]:
"""
Args:
spatial_size: output grid size.
grid: grid to be transformed. Shape must be (3, H, W) for 2D or (4, H, W, D) for 3D.
Returns:
a 2D (3xHxW) or 3D (4xHxWxD) grid.
"""
self.randomize()
affine_grid = AffineGrid(
rotate_params=self.rotate_params,
shear_params=self.shear_params,
translate_params=self.translate_params,
scale_params=self.scale_params,
as_tensor_output=self.as_tensor_output,
device=self.device,
)
grid, self.affine = affine_grid(spatial_size, grid)
return grid
[docs]class Resample(Transform):
def __init__(
self,
mode: Union[GridSampleMode, str] = GridSampleMode.BILINEAR,
padding_mode: Union[GridSamplePadMode, str] = GridSamplePadMode.BORDER,
as_tensor_output: bool = False,
device: Optional[torch.device] = None,
) -> None:
"""
computes output image using values from `img`, locations from `grid` using pytorch.
supports spatially 2D or 3D (num_channels, H, W[, D]).
Args:
mode: {``"bilinear"``, ``"nearest"``}
Interpolation mode to calculate output values. Defaults to ``"bilinear"``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
padding_mode: {``"zeros"``, ``"border"``, ``"reflection"``}
Padding mode for outside grid values. Defaults to ``"border"``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
as_tensor_output: whether to return a torch tensor. Defaults to False.
device: device on which the tensor will be allocated.
"""
self.mode: GridSampleMode = GridSampleMode(mode)
self.padding_mode: GridSamplePadMode = GridSamplePadMode(padding_mode)
self.as_tensor_output = as_tensor_output
self.device = device
[docs] def __call__(
self,
img: Union[np.ndarray, torch.Tensor],
grid: Optional[Union[np.ndarray, torch.Tensor]] = None,
mode: Optional[Union[GridSampleMode, str]] = None,
padding_mode: Optional[Union[GridSamplePadMode, str]] = None,
) -> Union[np.ndarray, torch.Tensor]:
"""
Args:
img: shape must be (num_channels, H, W[, D]).
grid: shape must be (3, H, W) for 2D or (4, H, W, D) for 3D.
mode: {``"bilinear"``, ``"nearest"``}
Interpolation mode to calculate output values. Defaults to ``self.mode``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
padding_mode: {``"zeros"``, ``"border"``, ``"reflection"``}
Padding mode for outside grid values. Defaults to ``self.padding_mode``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
"""
if not isinstance(img, torch.Tensor):
img = torch.as_tensor(np.ascontiguousarray(img))
if grid is None:
raise AssertionError("Error, grid argument must be supplied as an ndarray or tensor ")
grid = torch.tensor(grid) if not isinstance(grid, torch.Tensor) else grid.detach().clone()
if self.device:
img = img.to(self.device)
grid = grid.to(self.device)
if USE_COMPILED:
for i, dim in enumerate(img.shape[1:]):
grid[i] += (dim - 1.0) / 2.0
grid = grid[:-1] / grid[-1:]
grid = grid.permute(list(range(grid.ndimension()))[1:] + [0])
_padding_mode = self.padding_mode.value if padding_mode is None else GridSamplePadMode(padding_mode).value
if _padding_mode == "zeros":
bound = 7
elif _padding_mode == "border":
bound = 0
else:
bound = 1
_interp_mode = self.mode.value if mode is None else GridSampleMode(mode).value
out = grid_pull(
img.unsqueeze(0).float(),
grid.unsqueeze(0).float(),
bound=bound,
extrapolate=True,
interpolation=1 if _interp_mode == "bilinear" else _interp_mode,
)[0]
else:
for i, dim in enumerate(img.shape[1:]):
grid[i] = 2.0 * grid[i] / (dim - 1.0)
grid = grid[:-1] / grid[-1:]
index_ordering: List[int] = list(range(img.ndimension() - 2, -1, -1))
grid = grid[index_ordering]
grid = grid.permute(list(range(grid.ndimension()))[1:] + [0])
out = torch.nn.functional.grid_sample(
img.unsqueeze(0).float(),
grid.unsqueeze(0).float(),
mode=self.mode.value if mode is None else GridSampleMode(mode).value,
padding_mode=self.padding_mode.value if padding_mode is None else GridSamplePadMode(padding_mode).value,
align_corners=True,
)[0]
if self.as_tensor_output:
return torch.as_tensor(out)
return np.asarray(out.cpu().numpy())
[docs]class Affine(Transform):
"""
Transform ``img`` given the affine parameters.
"""
def __init__(
self,
rotate_params: Optional[Union[Sequence[float], float]] = None,
shear_params: Optional[Union[Sequence[float], float]] = None,
translate_params: Optional[Union[Sequence[float], float]] = None,
scale_params: Optional[Union[Sequence[float], float]] = None,
spatial_size: Optional[Union[Sequence[int], int]] = None,
mode: Union[GridSampleMode, str] = GridSampleMode.BILINEAR,
padding_mode: Union[GridSamplePadMode, str] = GridSamplePadMode.REFLECTION,
as_tensor_output: bool = False,
device: Optional[torch.device] = None,
) -> None:
"""
The affine transformations are applied in rotate, shear, translate, scale order.
Args:
rotate_params: a rotation angle in radians, a scalar for 2D image, a tuple of 3 floats for 3D.
Defaults to no rotation.
shear_params: a tuple of 2 floats for 2D, a tuple of 6 floats for 3D. Defaults to no shearing.
translate_params: a tuple of 2 floats for 2D, a tuple of 3 floats for 3D. Translation is in
pixel/voxel relative to the center of the input image. Defaults to no translation.
scale_params: a tuple of 2 floats for 2D, a tuple of 3 floats for 3D. Defaults to no scaling.
spatial_size: output image spatial size.
if `spatial_size` and `self.spatial_size` are not defined, or smaller than 1,
the transform will use the spatial size of `img`.
if the components of the `spatial_size` are non-positive values, the transform will use the
corresponding components of img size. For example, `spatial_size=(32, -1)` will be adapted
to `(32, 64)` if the second spatial dimension size of img is `64`.
mode: {``"bilinear"``, ``"nearest"``}
Interpolation mode to calculate output values. Defaults to ``"bilinear"``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
padding_mode: {``"zeros"``, ``"border"``, ``"reflection"``}
Padding mode for outside grid values. Defaults to ``"reflection"``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
as_tensor_output: the computation is implemented using pytorch tensors, this option specifies
whether to convert it back to numpy arrays.
device: device on which the tensor will be allocated.
"""
self.affine_grid = AffineGrid(
rotate_params=rotate_params,
shear_params=shear_params,
translate_params=translate_params,
scale_params=scale_params,
as_tensor_output=True,
device=device,
)
self.resampler = Resample(as_tensor_output=as_tensor_output, device=device)
self.spatial_size = spatial_size
self.mode: GridSampleMode = GridSampleMode(mode)
self.padding_mode: GridSamplePadMode = GridSamplePadMode(padding_mode)
[docs] def __call__(
self,
img: Union[np.ndarray, torch.Tensor],
spatial_size: Optional[Union[Sequence[int], int]] = None,
mode: Optional[Union[GridSampleMode, str]] = None,
padding_mode: Optional[Union[GridSamplePadMode, str]] = None,
) -> Tuple[Union[np.ndarray, torch.Tensor], Union[np.ndarray, torch.Tensor]]:
"""
Args:
img: shape must be (num_channels, H, W[, D]),
spatial_size: output image spatial size.
if `spatial_size` and `self.spatial_size` are not defined, or smaller than 1,
the transform will use the spatial size of `img`.
if `img` has two spatial dimensions, `spatial_size` should have 2 elements [h, w].
if `img` has three spatial dimensions, `spatial_size` should have 3 elements [h, w, d].
mode: {``"bilinear"``, ``"nearest"``}
Interpolation mode to calculate output values. Defaults to ``self.mode``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
padding_mode: {``"zeros"``, ``"border"``, ``"reflection"``}
Padding mode for outside grid values. Defaults to ``self.padding_mode``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
"""
sp_size = fall_back_tuple(spatial_size or self.spatial_size, img.shape[1:])
grid, affine = self.affine_grid(spatial_size=sp_size)
return (
self.resampler(img=img, grid=grid, mode=mode or self.mode, padding_mode=padding_mode or self.padding_mode),
affine,
)
[docs]class RandAffine(RandomizableTransform):
"""
Random affine transform.
"""
def __init__(
self,
prob: float = 0.1,
rotate_range: RandRange = None,
shear_range: RandRange = None,
translate_range: RandRange = None,
scale_range: RandRange = None,
spatial_size: Optional[Union[Sequence[int], int]] = None,
mode: Union[GridSampleMode, str] = GridSampleMode.BILINEAR,
padding_mode: Union[GridSamplePadMode, str] = GridSamplePadMode.REFLECTION,
cache_grid: bool = False,
as_tensor_output: bool = True,
device: Optional[torch.device] = None,
) -> None:
"""
Args:
prob: probability of returning a randomized affine grid.
defaults to 0.1, with 10% chance returns a randomized grid.
rotate_range: angle range in radians. If element `i` is iterable, then
`uniform[-rotate_range[i][0], rotate_range[i][1])` will be used to generate the rotation parameter
for the ith dimension. If not, `uniform[-rotate_range[i], rotate_range[i])` will be used. This can
be altered on a per-dimension basis. E.g., `((0,3), 1, ...)`: for dim0, rotation will be in range
`[0, 3]`, and for dim1 `[-1, 1]` will be used. Setting a single value will use `[-x, x]` for dim0
and nothing for the remaining dimensions.
shear_range: shear_range with format matching `rotate_range`.
translate_range: translate_range with format matching `rotate_range`.
scale_range: scaling_range with format matching `rotate_range`. A value of 1.0 is added to the result.
This allows 0 to correspond to no change (i.e., a scaling of 1).
spatial_size: output image spatial size.
if `spatial_size` and `self.spatial_size` are not defined, or smaller than 1,
the transform will use the spatial size of `img`.
if the components of the `spatial_size` are non-positive values, the transform will use the
corresponding components of img size. For example, `spatial_size=(32, -1)` will be adapted
to `(32, 64)` if the second spatial dimension size of img is `64`.
mode: {``"bilinear"``, ``"nearest"``}
Interpolation mode to calculate output values. Defaults to ``"bilinear"``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
padding_mode: {``"zeros"``, ``"border"``, ``"reflection"``}
Padding mode for outside grid values. Defaults to ``"reflection"``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
cache_grid: whether to cache the identity sampling grid.
If the spatial size is not dynamically defined by input image, enabling this option could
accelerate the transform.
as_tensor_output: the computation is implemented using pytorch tensors, this option specifies
whether to convert it back to numpy arrays.
device: device on which the tensor will be allocated.
See also:
- :py:class:`RandAffineGrid` for the random affine parameters configurations.
- :py:class:`Affine` for the affine transformation parameters configurations.
"""
RandomizableTransform.__init__(self, prob)
self.rand_affine_grid = RandAffineGrid(
rotate_range=rotate_range,
shear_range=shear_range,
translate_range=translate_range,
scale_range=scale_range,
as_tensor_output=True,
device=device,
)
self.resampler = Resample(as_tensor_output=as_tensor_output, device=device)
self.spatial_size = spatial_size
self.cache_grid = cache_grid
self._cached_grid = self._init_identity_cache()
self.mode: GridSampleMode = GridSampleMode(mode)
self.padding_mode: GridSamplePadMode = GridSamplePadMode(padding_mode)
def _init_identity_cache(self):
"""
Create cache of the identity grid if cache_grid=True and spatial_size is known.
"""
if self.spatial_size is None:
if self.cache_grid:
warnings.warn(
"cache_grid=True is not compatible with the dynamic spatial_size, please specify 'spatial_size'."
)
return None
_sp_size = ensure_tuple(self.spatial_size)
_ndim = len(_sp_size)
if _sp_size != fall_back_tuple(_sp_size, [1] * _ndim) or _sp_size != fall_back_tuple(_sp_size, [2] * _ndim):
# dynamic shape because it falls back to different outcomes
if self.cache_grid:
warnings.warn(
"cache_grid=True is not compatible with the dynamic spatial_size "
f"'spatial_size={self.spatial_size}', please specify 'spatial_size'."
)
return None
return torch.tensor(create_grid(spatial_size=_sp_size)).to(self.rand_affine_grid.device)
[docs] def get_identity_grid(self, spatial_size: Sequence[int]):
"""
Return a cached or new identity grid depends on the availability.
Args:
spatial_size: non-dynamic spatial size
"""
ndim = len(spatial_size)
if spatial_size != fall_back_tuple(spatial_size, [1] * ndim) or spatial_size != fall_back_tuple(
spatial_size, [2] * ndim
):
raise RuntimeError(f"spatial_size should not be dynamic, got {spatial_size}.")
return create_grid(spatial_size=spatial_size) if self._cached_grid is None else self._cached_grid
[docs] def set_random_state(
self, seed: Optional[int] = None, state: Optional[np.random.RandomState] = None
) -> "RandAffine":
self.rand_affine_grid.set_random_state(seed, state)
super().set_random_state(seed, state)
return self
[docs] def randomize(self, data: Optional[Any] = None) -> None:
super().randomize(None)
self.rand_affine_grid.randomize()
[docs] def __call__(
self,
img: Union[np.ndarray, torch.Tensor],
spatial_size: Optional[Union[Sequence[int], int]] = None,
mode: Optional[Union[GridSampleMode, str]] = None,
padding_mode: Optional[Union[GridSamplePadMode, str]] = None,
) -> Union[np.ndarray, torch.Tensor]:
"""
Args:
img: shape must be (num_channels, H, W[, D]),
spatial_size: output image spatial size.
if `spatial_size` and `self.spatial_size` are not defined, or smaller than 1,
the transform will use the spatial size of `img`.
if `img` has two spatial dimensions, `spatial_size` should have 2 elements [h, w].
if `img` has three spatial dimensions, `spatial_size` should have 3 elements [h, w, d].
mode: {``"bilinear"``, ``"nearest"``}
Interpolation mode to calculate output values. Defaults to ``self.mode``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
padding_mode: {``"zeros"``, ``"border"``, ``"reflection"``}
Padding mode for outside grid values. Defaults to ``self.padding_mode``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
"""
self.randomize()
# if not doing transform and spatial size doesn't change, nothing to do
# except convert to float and convert numpy/torch
sp_size = fall_back_tuple(spatial_size or self.spatial_size, img.shape[1:])
do_resampling = self._do_transform or (sp_size != ensure_tuple(img.shape[1:]))
if not do_resampling:
img = img.float() if isinstance(img, torch.Tensor) else img.astype("float32")
return torch.Tensor(img) if self.resampler.as_tensor_output else np.array(img)
grid = self.get_identity_grid(sp_size)
if self._do_transform:
grid = self.rand_affine_grid(grid=grid)
return self.resampler(
img=img, grid=grid, mode=mode or self.mode, padding_mode=padding_mode or self.padding_mode
)
[docs]class Rand2DElastic(RandomizableTransform):
"""
Random elastic deformation and affine in 2D
"""
def __init__(
self,
spacing: Union[Tuple[float, float], float],
magnitude_range: Tuple[float, float],
prob: float = 0.1,
rotate_range: RandRange = None,
shear_range: RandRange = None,
translate_range: RandRange = None,
scale_range: RandRange = None,
spatial_size: Optional[Union[Tuple[int, int], int]] = None,
mode: Union[GridSampleMode, str] = GridSampleMode.BILINEAR,
padding_mode: Union[GridSamplePadMode, str] = GridSamplePadMode.REFLECTION,
as_tensor_output: bool = False,
device: Optional[torch.device] = None,
) -> None:
"""
Args:
spacing : distance in between the control points.
magnitude_range: the random offsets will be generated from ``uniform[magnitude[0], magnitude[1])``.
prob: probability of returning a randomized elastic transform.
defaults to 0.1, with 10% chance returns a randomized elastic transform,
otherwise returns a ``spatial_size`` centered area extracted from the input image.
rotate_range: angle range in radians. If element `i` is iterable, then
`uniform[-rotate_range[i][0], rotate_range[i][1])` will be used to generate the rotation parameter
for the ith dimension. If not, `uniform[-rotate_range[i], rotate_range[i])` will be used. This can
be altered on a per-dimension basis. E.g., `((0,3), 1, ...)`: for dim0, rotation will be in range
`[0, 3]`, and for dim1 `[-1, 1]` will be used. Setting a single value will use `[-x, x]` for dim0
and nothing for the remaining dimensions.
shear_range: shear_range with format matching `rotate_range`.
translate_range: translate_range with format matching `rotate_range`.
scale_range: scaling_range with format matching `rotate_range`. A value of 1.0 is added to the result.
This allows 0 to correspond to no change (i.e., a scaling of 1).
spatial_size: specifying output image spatial size [h, w].
if `spatial_size` and `self.spatial_size` are not defined, or smaller than 1,
the transform will use the spatial size of `img`.
if the components of the `spatial_size` are non-positive values, the transform will use the
corresponding components of img size. For example, `spatial_size=(32, -1)` will be adapted
to `(32, 64)` if the second spatial dimension size of img is `64`.
mode: {``"bilinear"``, ``"nearest"``}
Interpolation mode to calculate output values. Defaults to ``"bilinear"``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
padding_mode: {``"zeros"``, ``"border"``, ``"reflection"``}
Padding mode for outside grid values. Defaults to ``"reflection"``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
as_tensor_output: the computation is implemented using pytorch tensors, this option specifies
whether to convert it back to numpy arrays.
device: device on which the tensor will be allocated.
See also:
- :py:class:`RandAffineGrid` for the random affine parameters configurations.
- :py:class:`Affine` for the affine transformation parameters configurations.
"""
RandomizableTransform.__init__(self, prob)
self.deform_grid = RandDeformGrid(
spacing=spacing, magnitude_range=magnitude_range, as_tensor_output=True, device=device
)
self.rand_affine_grid = RandAffineGrid(
rotate_range=rotate_range,
shear_range=shear_range,
translate_range=translate_range,
scale_range=scale_range,
as_tensor_output=True,
device=device,
)
self.resampler = Resample(as_tensor_output=as_tensor_output, device=device)
self.spatial_size = spatial_size
self.mode: GridSampleMode = GridSampleMode(mode)
self.padding_mode: GridSamplePadMode = GridSamplePadMode(padding_mode)
[docs] def set_random_state(
self, seed: Optional[int] = None, state: Optional[np.random.RandomState] = None
) -> "Rand2DElastic":
self.deform_grid.set_random_state(seed, state)
self.rand_affine_grid.set_random_state(seed, state)
super().set_random_state(seed, state)
return self
[docs] def randomize(self, spatial_size: Sequence[int]) -> None:
super().randomize(None)
self.deform_grid.randomize(spatial_size)
self.rand_affine_grid.randomize()
[docs] def __call__(
self,
img: Union[np.ndarray, torch.Tensor],
spatial_size: Optional[Union[Tuple[int, int], int]] = None,
mode: Optional[Union[GridSampleMode, str]] = None,
padding_mode: Optional[Union[GridSamplePadMode, str]] = None,
) -> Union[np.ndarray, torch.Tensor]:
"""
Args:
img: shape must be (num_channels, H, W),
spatial_size: specifying output image spatial size [h, w].
if `spatial_size` and `self.spatial_size` are not defined, or smaller than 1,
the transform will use the spatial size of `img`.
mode: {``"bilinear"``, ``"nearest"``}
Interpolation mode to calculate output values. Defaults to ``self.mode``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
padding_mode: {``"zeros"``, ``"border"``, ``"reflection"``}
Padding mode for outside grid values. Defaults to ``self.padding_mode``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
"""
sp_size = fall_back_tuple(spatial_size or self.spatial_size, img.shape[1:])
self.randomize(spatial_size=sp_size)
if self._do_transform:
grid = self.deform_grid(spatial_size=sp_size)
grid = self.rand_affine_grid(grid=grid)
grid = torch.nn.functional.interpolate( # type: ignore
recompute_scale_factor=True,
input=torch.as_tensor(grid).unsqueeze(0),
scale_factor=list(ensure_tuple(self.deform_grid.spacing)),
mode=InterpolateMode.BICUBIC.value,
align_corners=False,
)
grid = CenterSpatialCrop(roi_size=sp_size)(grid[0])
else:
grid = create_grid(spatial_size=sp_size)
return self.resampler(img, grid, mode=mode or self.mode, padding_mode=padding_mode or self.padding_mode)
[docs]class Rand3DElastic(RandomizableTransform):
"""
Random elastic deformation and affine in 3D
"""
def __init__(
self,
sigma_range: Tuple[float, float],
magnitude_range: Tuple[float, float],
prob: float = 0.1,
rotate_range: RandRange = None,
shear_range: RandRange = None,
translate_range: RandRange = None,
scale_range: RandRange = None,
spatial_size: Optional[Union[Tuple[int, int, int], int]] = None,
mode: Union[GridSampleMode, str] = GridSampleMode.BILINEAR,
padding_mode: Union[GridSamplePadMode, str] = GridSamplePadMode.REFLECTION,
as_tensor_output: bool = False,
device: Optional[torch.device] = None,
) -> None:
"""
Args:
sigma_range: a Gaussian kernel with standard deviation sampled from
``uniform[sigma_range[0], sigma_range[1])`` will be used to smooth the random offset grid.
magnitude_range: the random offsets on the grid will be generated from
``uniform[magnitude[0], magnitude[1])``.
prob: probability of returning a randomized elastic transform.
defaults to 0.1, with 10% chance returns a randomized elastic transform,
otherwise returns a ``spatial_size`` centered area extracted from the input image.
rotate_range: angle range in radians. If element `i` is iterable, then
`uniform[-rotate_range[i][0], rotate_range[i][1])` will be used to generate the rotation parameter
for the ith dimension. If not, `uniform[-rotate_range[i], rotate_range[i])` will be used. This can
be altered on a per-dimension basis. E.g., `((0,3), 1, ...)`: for dim0, rotation will be in range
`[0, 3]`, and for dim1 `[-1, 1]` will be used. Setting a single value will use `[-x, x]` for dim0
and nothing for the remaining dimensions.
shear_range: shear_range with format matching `rotate_range`.
translate_range: translate_range with format matching `rotate_range`.
scale_range: scaling_range with format matching `rotate_range`. A value of 1.0 is added to the result.
This allows 0 to correspond to no change (i.e., a scaling of 1).
spatial_size: specifying output image spatial size [h, w, d].
if `spatial_size` and `self.spatial_size` are not defined, or smaller than 1,
the transform will use the spatial size of `img`.
if the components of the `spatial_size` are non-positive values, the transform will use the
corresponding components of img size. For example, `spatial_size=(32, 32, -1)` will be adapted
to `(32, 32, 64)` if the third spatial dimension size of img is `64`.
mode: {``"bilinear"``, ``"nearest"``}
Interpolation mode to calculate output values. Defaults to ``"bilinear"``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
padding_mode: {``"zeros"``, ``"border"``, ``"reflection"``}
Padding mode for outside grid values. Defaults to ``"reflection"``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
as_tensor_output: the computation is implemented using pytorch tensors, this option specifies
whether to convert it back to numpy arrays.
device: device on which the tensor will be allocated.
See also:
- :py:class:`RandAffineGrid` for the random affine parameters configurations.
- :py:class:`Affine` for the affine transformation parameters configurations.
"""
RandomizableTransform.__init__(self, prob)
self.rand_affine_grid = RandAffineGrid(rotate_range, shear_range, translate_range, scale_range, True, device)
self.resampler = Resample(as_tensor_output=as_tensor_output, device=device)
self.sigma_range = sigma_range
self.magnitude_range = magnitude_range
self.spatial_size = spatial_size
self.mode: GridSampleMode = GridSampleMode(mode)
self.padding_mode: GridSamplePadMode = GridSamplePadMode(padding_mode)
self.device = device
self.rand_offset = None
self.magnitude = 1.0
self.sigma = 1.0
[docs] def set_random_state(
self, seed: Optional[int] = None, state: Optional[np.random.RandomState] = None
) -> "Rand3DElastic":
self.rand_affine_grid.set_random_state(seed, state)
super().set_random_state(seed, state)
return self
[docs] def randomize(self, grid_size: Sequence[int]) -> None:
super().randomize(None)
if self._do_transform:
self.rand_offset = self.R.uniform(-1.0, 1.0, [3] + list(grid_size)).astype(np.float32)
self.magnitude = self.R.uniform(self.magnitude_range[0], self.magnitude_range[1])
self.sigma = self.R.uniform(self.sigma_range[0], self.sigma_range[1])
self.rand_affine_grid.randomize()
[docs] def __call__(
self,
img: Union[np.ndarray, torch.Tensor],
spatial_size: Optional[Union[Tuple[int, int, int], int]] = None,
mode: Optional[Union[GridSampleMode, str]] = None,
padding_mode: Optional[Union[GridSamplePadMode, str]] = None,
) -> Union[np.ndarray, torch.Tensor]:
"""
Args:
img: shape must be (num_channels, H, W, D),
spatial_size: specifying spatial 3D output image spatial size [h, w, d].
if `spatial_size` and `self.spatial_size` are not defined, or smaller than 1,
the transform will use the spatial size of `img`.
mode: {``"bilinear"``, ``"nearest"``}
Interpolation mode to calculate output values. Defaults to ``self.mode``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
padding_mode: {``"zeros"``, ``"border"``, ``"reflection"``}
Padding mode for outside grid values. Defaults to ``self.padding_mode``.
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
"""
sp_size = fall_back_tuple(spatial_size or self.spatial_size, img.shape[1:])
self.randomize(grid_size=sp_size)
grid = create_grid(spatial_size=sp_size)
if self._do_transform:
if self.rand_offset is None:
raise AssertionError
grid = torch.as_tensor(np.ascontiguousarray(grid), device=self.device)
gaussian = GaussianFilter(3, self.sigma, 3.0).to(device=self.device)
offset = torch.as_tensor(self.rand_offset, device=self.device).unsqueeze(0)
grid[:3] += gaussian(offset)[0] * self.magnitude
grid = self.rand_affine_grid(grid=grid)
return self.resampler(img, grid, mode=mode or self.mode, padding_mode=padding_mode or self.padding_mode)