Skip to content

GitLab

Unverified Commit 7b4681a6 authored by Nicolas Hug's avatar Nicolas Hug Committed by GitHub
Browse files

Remove BETA status for v2 transforms (#8111)

parent 0e2a5ae7
Hide whitespace changes
Inline Side-by-side
Showing with 86 additions and 214 deletions
docs/source/beta_status.py
View file @ 7b4681a6
......@@ -12,18 +12,8 @@ class BetaStatus(Directive):
return [self.node("", nodes.paragraph("", "", nodes.Text(text)))]
class V2BetaStatus(BetaStatus):
text = (
"The {api_name} is in Beta stage, and while we do not expect disruptive breaking changes, "
"some APIs may slightly change according to user feedback. Please submit any feedback you may have "
"in this issue: https://github.com/pytorch/vision/issues/6753."
)
node = nodes.note
def setup(app):
app.add_directive("betastatus", BetaStatus)
app.add_directive("v2betastatus", V2BetaStatus)
return {
"version": "0.1",
"parallel_read_safe": True,
......
docs/source/transforms.rst
View file @ 7b4681a6
......@@ -126,13 +126,6 @@ you're already using tranforms from ``torchvision.transforms``, all you need to
do to is to update the import to ``torchvision.transforms.v2``. In terms of
output, there might be negligible differences due to implementation differences.
.. note::
The v2 transforms are still BETA, but at this point we do not expect
disruptive changes to be made to their public APIs. We're planning to make
them fully stable in version 0.17. Please submit any feedback you may have
`here <https://github.com/pytorch/vision/issues/6753>`_.
.. _transforms_perf:
Performance considerations
......
torchvision/transforms/v2/_augment.py
View file @ 7b4681a6
......@@ -15,9 +15,7 @@ from ._utils import _parse_labels_getter, has_any, is_pure_tensor, query_chw, qu
class RandomErasing(_RandomApplyTransform):
"""[BETA] Randomly select a rectangle region in the input image or video and erase its pixels.
.. v2betastatus:: RandomErasing transform
"""Randomly select a rectangle region in the input image or video and erase its pixels.
This transform does not support PIL Image.
'Random Erasing Data Augmentation' by Zhong et al. See https://arxiv.org/abs/1708.04896
......@@ -207,9 +205,7 @@ class _BaseMixUpCutMix(Transform):
class MixUp(_BaseMixUpCutMix):
"""[BETA] Apply MixUp to the provided batch of images and labels.
.. v2betastatus:: MixUp transform
"""Apply MixUp to the provided batch of images and labels.
Paper: `mixup: Beyond Empirical Risk Minimization <https://arxiv.org/abs/1710.09412>`_.
......@@ -256,9 +252,7 @@ class MixUp(_BaseMixUpCutMix):
class CutMix(_BaseMixUpCutMix):
"""[BETA] Apply CutMix to the provided batch of images and labels.
.. v2betastatus:: CutMix transform
"""Apply CutMix to the provided batch of images and labels.
Paper: `CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features
<https://arxiv.org/abs/1905.04899>`_.
......
torchvision/transforms/v2/_auto_augment.py
View file @ 7b4681a6
......@@ -174,11 +174,9 @@ class _AutoAugmentBase(Transform):
class AutoAugment(_AutoAugmentBase):
r"""[BETA] AutoAugment data augmentation method based on
r"""AutoAugment data augmentation method based on
`"AutoAugment: Learning Augmentation Strategies from Data" <https://arxiv.org/pdf/1805.09501.pdf>`_.
.. v2betastatus:: AutoAugment transform
This transformation works on images and videos only.
If the input is :class:`torch.Tensor`, it should be of type ``torch.uint8``, and it is expected
......@@ -350,12 +348,10 @@ class AutoAugment(_AutoAugmentBase):
class RandAugment(_AutoAugmentBase):
r"""[BETA] RandAugment data augmentation method based on
r"""RandAugment data augmentation method based on
`"RandAugment: Practical automated data augmentation with a reduced search space"
<https://arxiv.org/abs/1909.13719>`_.
.. v2betastatus:: RandAugment transform
This transformation works on images and videos only.
If the input is :class:`torch.Tensor`, it should be of type ``torch.uint8``, and it is expected
......@@ -434,11 +430,9 @@ class RandAugment(_AutoAugmentBase):
class TrivialAugmentWide(_AutoAugmentBase):
r"""[BETA] Dataset-independent data-augmentation with TrivialAugment Wide, as described in
r"""Dataset-independent data-augmentation with TrivialAugment Wide, as described in
`"TrivialAugment: Tuning-free Yet State-of-the-Art Data Augmentation" <https://arxiv.org/abs/2103.10158>`_.
.. v2betastatus:: TrivialAugmentWide transform
This transformation works on images and videos only.
If the input is :class:`torch.Tensor`, it should be of type ``torch.uint8``, and it is expected
......@@ -505,11 +499,9 @@ class TrivialAugmentWide(_AutoAugmentBase):
class AugMix(_AutoAugmentBase):
r"""[BETA] AugMix data augmentation method based on
r"""AugMix data augmentation method based on
`"AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty" <https://arxiv.org/abs/1912.02781>`_.
.. v2betastatus:: AugMix transform
This transformation works on images and videos only.
If the input is :class:`torch.Tensor`, it should be of type ``torch.uint8``, and it is expected
......
torchvision/transforms/v2/_color.py
View file @ 7b4681a6
......@@ -10,9 +10,7 @@ from ._utils import query_chw
class Grayscale(Transform):
"""[BETA] Convert images or videos to grayscale.
.. v2betastatus:: Grayscale transform
"""Convert images or videos to grayscale.
If the input is a :class:`torch.Tensor`, it is expected
to have [..., 3 or 1, H, W] shape, where ... means an arbitrary number of leading dimensions
......@@ -32,9 +30,7 @@ class Grayscale(Transform):
class RandomGrayscale(_RandomApplyTransform):
"""[BETA] Randomly convert image or videos to grayscale with a probability of p (default 0.1).
.. v2betastatus:: RandomGrayscale transform
"""Randomly convert image or videos to grayscale with a probability of p (default 0.1).
If the input is a :class:`torch.Tensor`, it is expected to have [..., 3 or 1, H, W] shape,
where ... means an arbitrary number of leading dimensions
......@@ -59,9 +55,7 @@ class RandomGrayscale(_RandomApplyTransform):
class ColorJitter(Transform):
"""[BETA] Randomly change the brightness, contrast, saturation and hue of an image or video.
.. v2betastatus:: ColorJitter transform
"""Randomly change the brightness, contrast, saturation and hue of an image or video.
If the input is a :class:`torch.Tensor`, it is expected
to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.
......@@ -163,10 +157,7 @@ class ColorJitter(Transform):
class RandomChannelPermutation(Transform):
"""[BETA] Randomly permute the channels of an image or video
.. v2betastatus:: RandomChannelPermutation transform
"""
"""Randomly permute the channels of an image or video"""
def _get_params(self, flat_inputs: List[Any]) -> Dict[str, Any]:
num_channels, *_ = query_chw(flat_inputs)
......@@ -177,11 +168,9 @@ class RandomChannelPermutation(Transform):
class RandomPhotometricDistort(Transform):
"""[BETA] Randomly distorts the image or video as used in `SSD: Single Shot
"""Randomly distorts the image or video as used in `SSD: Single Shot
MultiBox Detector <https://arxiv.org/abs/1512.02325>`_.
.. v2betastatus:: RandomPhotometricDistort transform
This transform relies on :class:`~torchvision.transforms.v2.ColorJitter`
under the hood to adjust the contrast, saturation, hue, brightness, and also
randomly permutes channels.
......@@ -249,9 +238,7 @@ class RandomPhotometricDistort(Transform):
class RandomEqualize(_RandomApplyTransform):
"""[BETA] Equalize the histogram of the given image or video with a given probability.
.. v2betastatus:: RandomEqualize transform
"""Equalize the histogram of the given image or video with a given probability.
If the input is a :class:`torch.Tensor`, it is expected
to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.
......@@ -268,9 +255,7 @@ class RandomEqualize(_RandomApplyTransform):
class RandomInvert(_RandomApplyTransform):
"""[BETA] Inverts the colors of the given image or video with a given probability.
.. v2betastatus:: RandomInvert transform
"""Inverts the colors of the given image or video with a given probability.
If img is a Tensor, it is expected to be in [..., 1 or 3, H, W] format,
where ... means it can have an arbitrary number of leading dimensions.
......@@ -287,11 +272,9 @@ class RandomInvert(_RandomApplyTransform):
class RandomPosterize(_RandomApplyTransform):
"""[BETA] Posterize the image or video with a given probability by reducing the
"""Posterize the image or video with a given probability by reducing the
number of bits for each color channel.
.. v2betastatus:: RandomPosterize transform
If the input is a :class:`torch.Tensor`, it should be of type torch.uint8,
and it is expected to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.
If img is PIL Image, it is expected to be in mode "L" or "RGB".
......@@ -312,11 +295,9 @@ class RandomPosterize(_RandomApplyTransform):
class RandomSolarize(_RandomApplyTransform):
"""[BETA] Solarize the image or video with a given probability by inverting all pixel
"""Solarize the image or video with a given probability by inverting all pixel
values above a threshold.
.. v2betastatus:: RandomSolarize transform
If img is a Tensor, it is expected to be in [..., 1 or 3, H, W] format,
where ... means it can have an arbitrary number of leading dimensions.
If img is PIL Image, it is expected to be in mode "L" or "RGB".
......@@ -342,9 +323,7 @@ class RandomSolarize(_RandomApplyTransform):
class RandomAutocontrast(_RandomApplyTransform):
"""[BETA] Autocontrast the pixels of the given image or video with a given probability.
.. v2betastatus:: RandomAutocontrast transform
"""Autocontrast the pixels of the given image or video with a given probability.
If the input is a :class:`torch.Tensor`, it is expected
to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.
......@@ -361,9 +340,7 @@ class RandomAutocontrast(_RandomApplyTransform):
class RandomAdjustSharpness(_RandomApplyTransform):
"""[BETA] Adjust the sharpness of the image or video with a given probability.
.. v2betastatus:: RandomAdjustSharpness transform
"""Adjust the sharpness of the image or video with a given probability.
If the input is a :class:`torch.Tensor`,
it is expected to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.
......
torchvision/transforms/v2/_container.py
View file @ 7b4681a6
......@@ -8,9 +8,7 @@ from torchvision.transforms.v2 import Transform
class Compose(Transform):
"""[BETA] Composes several transforms together.
.. v2betastatus:: Compose transform
"""Composes several transforms together.
This transform does not support torchscript.
Please, see the note below.
......@@ -62,9 +60,7 @@ class Compose(Transform):
class RandomApply(Transform):
"""[BETA] Apply randomly a list of transformations with a given probability.
.. v2betastatus:: RandomApply transform
"""Apply randomly a list of transformations with a given probability.
.. note::
In order to script the transformation, please use ``torch.nn.ModuleList`` as input instead of list/tuple of
......@@ -118,9 +114,7 @@ class RandomApply(Transform):
class RandomChoice(Transform):
"""[BETA] Apply single transformation randomly picked from a list.
.. v2betastatus:: RandomChoice transform
"""Apply single transformation randomly picked from a list.
This transform does not support torchscript.
......@@ -157,9 +151,7 @@ class RandomChoice(Transform):
class RandomOrder(Transform):
"""[BETA] Apply a list of transformations in a random order.
.. v2betastatus:: RandomOrder transform
"""Apply a list of transformations in a random order.
This transform does not support torchscript.
......
torchvision/transforms/v2/_deprecated.py
View file @ 7b4681a6
......@@ -10,12 +10,10 @@ from torchvision.transforms.v2 import Transform
class ToTensor(Transform):
"""[BETA] [DEPRECATED] Use ``v2.Compose([v2.ToImage(), v2.ToDtype(torch.float32, scale=True)])`` instead.
"""[DEPRECATED] Use ``v2.Compose([v2.ToImage(), v2.ToDtype(torch.float32, scale=True)])`` instead.
Convert a PIL Image or ndarray to tensor and scale the values accordingly.
.. v2betastatus:: ToTensor transform
.. warning::
:class:`v2.ToTensor` is deprecated and will be removed in a future release.
Please use instead ``v2.Compose([v2.ToImage(), v2.ToDtype(torch.float32, scale=True)])``.
......
torchvision/transforms/v2/_geometry.py
View file @ 7b4681a6
......@@ -31,9 +31,7 @@ from ._utils import (
class RandomHorizontalFlip(_RandomApplyTransform):
"""[BETA] Horizontally flip the input with a given probability.
.. v2betastatus:: RandomHorizontalFlip transform
"""Horizontally flip the input with a given probability.
If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,
:class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)
......@@ -51,9 +49,7 @@ class RandomHorizontalFlip(_RandomApplyTransform):
class RandomVerticalFlip(_RandomApplyTransform):
"""[BETA] Vertically flip the input with a given probability.
.. v2betastatus:: RandomVerticalFlip transform
"""Vertically flip the input with a given probability.
If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,
:class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)
......@@ -71,9 +67,7 @@ class RandomVerticalFlip(_RandomApplyTransform):
class Resize(Transform):
"""[BETA] Resize the input to the given size.
.. v2betastatus:: Resize transform
"""Resize the input to the given size.
If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,
:class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)
......@@ -159,9 +153,7 @@ class Resize(Transform):
class CenterCrop(Transform):
"""[BETA] Crop the input at the center.
.. v2betastatus:: CenterCrop transform
"""Crop the input at the center.
If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,
:class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)
......@@ -187,9 +179,7 @@ class CenterCrop(Transform):
class RandomResizedCrop(Transform):
"""[BETA] Crop a random portion of the input and resize it to a given size.
.. v2betastatus:: RandomResizedCrop transform
"""Crop a random portion of the input and resize it to a given size.
If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,
:class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)
......@@ -310,9 +300,7 @@ class RandomResizedCrop(Transform):
class FiveCrop(Transform):
"""[BETA] Crop the image or video into four corners and the central crop.
.. v2betastatus:: FiveCrop transform
"""Crop the image or video into four corners and the central crop.
If the input is a :class:`torch.Tensor` or a :class:`~torchvision.tv_tensors.Image` or a
:class:`~torchvision.tv_tensors.Video` it can have arbitrary number of leading batch dimensions.
......@@ -371,11 +359,9 @@ class FiveCrop(Transform):
class TenCrop(Transform):
"""[BETA] Crop the image or video into four corners and the central crop plus the flipped version of
"""Crop the image or video into four corners and the central crop plus the flipped version of
these (horizontal flipping is used by default).
.. v2betastatus:: TenCrop transform
If the input is a :class:`torch.Tensor` or a :class:`~torchvision.tv_tensors.Image` or a
:class:`~torchvision.tv_tensors.Video` it can have arbitrary number of leading batch dimensions.
For example, the image can have ``[..., C, H, W]`` shape.
......@@ -418,9 +404,7 @@ class TenCrop(Transform):
class Pad(Transform):
"""[BETA] Pad the input on all sides with the given "pad" value.
.. v2betastatus:: Pad transform
"""Pad the input on all sides with the given "pad" value.
If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,
:class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)
......@@ -492,11 +476,9 @@ class Pad(Transform):
class RandomZoomOut(_RandomApplyTransform):
"""[BETA] "Zoom out" transformation from
""" "Zoom out" transformation from
`"SSD: Single Shot MultiBox Detector" <https://arxiv.org/abs/1512.02325>`_.
.. v2betastatus:: RandomZoomOut transform
This transformation randomly pads images, videos, bounding boxes and masks creating a zoom out effect.
Output spatial size is randomly sampled from original size up to a maximum size configured
with ``side_range`` parameter:
......@@ -562,9 +544,7 @@ class RandomZoomOut(_RandomApplyTransform):
class RandomRotation(Transform):
"""[BETA] Rotate the input by angle.
.. v2betastatus:: RandomRotation transform
"""Rotate the input by angle.
If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,
:class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)
......@@ -643,9 +623,7 @@ class RandomRotation(Transform):
class RandomAffine(Transform):
"""[BETA] Random affine transformation the input keeping center invariant.
.. v2betastatus:: RandomAffine transform
"""Random affine transformation the input keeping center invariant.
If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,
:class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)
......@@ -765,9 +743,7 @@ class RandomAffine(Transform):
class RandomCrop(Transform):
"""[BETA] Crop the input at a random location.
.. v2betastatus:: RandomCrop transform
"""Crop the input at a random location.
If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,
:class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)
......@@ -922,9 +898,7 @@ class RandomCrop(Transform):
class RandomPerspective(_RandomApplyTransform):
"""[BETA] Perform a random perspective transformation of the input with a given probability.
.. v2betastatus:: RandomPerspective transform
"""Perform a random perspective transformation of the input with a given probability.
If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,
:class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)
......@@ -1009,9 +983,7 @@ class RandomPerspective(_RandomApplyTransform):
class ElasticTransform(Transform):
"""[BETA] Transform the input with elastic transformations.
.. v2betastatus:: RandomPerspective transform
"""Transform the input with elastic transformations.
If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,
:class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)
......@@ -1101,11 +1073,9 @@ class ElasticTransform(Transform):
class RandomIoUCrop(Transform):
"""[BETA] Random IoU crop transformation from
"""Random IoU crop transformation from
`"SSD: Single Shot MultiBox Detector" <https://arxiv.org/abs/1512.02325>`_.
.. v2betastatus:: RandomIoUCrop transform
This transformation requires an image or video data and ``tv_tensors.BoundingBoxes`` in the input.
.. warning::
......@@ -1229,11 +1199,9 @@ class RandomIoUCrop(Transform):
class ScaleJitter(Transform):
"""[BETA] Perform Large Scale Jitter on the input according to
"""Perform Large Scale Jitter on the input according to
`"Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentation" <https://arxiv.org/abs/2012.07177>`_.
.. v2betastatus:: ScaleJitter transform
If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,
:class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)
it can have arbitrary number of leading batch dimensions. For example,
......@@ -1298,9 +1266,7 @@ class ScaleJitter(Transform):
class RandomShortestSize(Transform):
"""[BETA] Randomly resize the input.
.. v2betastatus:: RandomShortestSize transform
"""Randomly resize the input.
If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,
:class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)
......@@ -1368,9 +1334,7 @@ class RandomShortestSize(Transform):
class RandomResize(Transform):
"""[BETA] Randomly resize the input.
.. v2betastatus:: RandomResize transform
"""Randomly resize the input.
This transformation can be used together with ``RandomCrop`` as data augmentations to train
models on image segmentation task.
......
torchvision/transforms/v2/_meta.py
View file @ 7b4681a6
......@@ -5,9 +5,7 @@ from torchvision.transforms.v2 import functional as F, Transform
class ConvertBoundingBoxFormat(Transform):
"""[BETA] Convert bounding box coordinates to the given ``format``, eg from "CXCYWH" to "XYXY".
.. v2betastatus:: ConvertBoundingBoxFormat transform
"""Convert bounding box coordinates to the given ``format``, eg from "CXCYWH" to "XYXY".
Args:
format (str or tv_tensors.BoundingBoxFormat): output bounding box format.
......@@ -28,12 +26,10 @@ class ConvertBoundingBoxFormat(Transform):
class ClampBoundingBoxes(Transform):
"""[BETA] Clamp bounding boxes to their corresponding image dimensions.
"""Clamp bounding boxes to their corresponding image dimensions.
The clamping is done according to the bounding boxes' ``canvas_size`` meta-data.
.. v2betastatus:: ClampBoundingBoxes transform
"""
_transformed_types = (tv_tensors.BoundingBoxes,)
......
torchvision/transforms/v2/_misc.py
View file @ 7b4681a6
......@@ -19,9 +19,7 @@ class Identity(Transform):
class Lambda(Transform):
"""[BETA] Apply a user-defined function as a transform.
.. v2betastatus:: Lambda transform
"""Apply a user-defined function as a transform.
This transform does not support torchscript.
......@@ -52,9 +50,7 @@ class Lambda(Transform):
class LinearTransformation(Transform):
"""[BETA] Transform a tensor image or video with a square transformation matrix and a mean_vector computed offline.
.. v2betastatus:: LinearTransformation transform
"""Transform a tensor image or video with a square transformation matrix and a mean_vector computed offline.
This transform does not support PIL Image.
Given transformation_matrix and mean_vector, will flatten the torch.*Tensor and
......@@ -135,9 +131,7 @@ class LinearTransformation(Transform):
class Normalize(Transform):
"""[BETA] Normalize a tensor image or video with mean and standard deviation.
.. v2betastatus:: Normalize transform
"""Normalize a tensor image or video with mean and standard deviation.
This transform does not support PIL Image.
Given mean: ``(mean[1],...,mean[n])`` and std: ``(std[1],..,std[n])`` for ``n``
......@@ -172,9 +166,7 @@ class Normalize(Transform):
class GaussianBlur(Transform):
"""[BETA] Blurs image with randomly chosen Gaussian blur.
.. v2betastatus:: GausssianBlur transform
"""Blurs image with randomly chosen Gaussian blur.
If the input is a Tensor, it is expected
to have [..., C, H, W] shape, where ... means an arbitrary number of leading dimensions.
......@@ -212,9 +204,7 @@ class GaussianBlur(Transform):
class ToDtype(Transform):
"""[BETA] Converts the input to a specific dtype, optionally scaling the values for images or videos.
.. v2betastatus:: ToDtype transform
"""Converts the input to a specific dtype, optionally scaling the values for images or videos.
.. note::
``ToDtype(dtype, scale=True)`` is the recommended replacement for ``ConvertImageDtype(dtype)``.
......@@ -286,12 +276,10 @@ class ToDtype(Transform):
class ConvertImageDtype(Transform):
"""[BETA] [DEPRECATED] Use ``v2.ToDtype(dtype, scale=True)`` instead.
"""[DEPRECATED] Use ``v2.ToDtype(dtype, scale=True)`` instead.
Convert input image to the given ``dtype`` and scale the values accordingly.
.. v2betastatus:: ConvertImageDtype transform
.. warning::
Consider using ``ToDtype(dtype, scale=True)`` instead. See :class:`~torchvision.transforms.v2.ToDtype`.
......@@ -323,9 +311,7 @@ class ConvertImageDtype(Transform):
class SanitizeBoundingBoxes(Transform):
"""[BETA] Remove degenerate/invalid bounding boxes and their corresponding labels and masks.
.. v2betastatus:: SanitizeBoundingBoxes transform
"""Remove degenerate/invalid bounding boxes and their corresponding labels and masks.
This transform removes bounding boxes and their associated labels/masks that:
......
torchvision/transforms/v2/_temporal.py
View file @ 7b4681a6
......@@ -5,9 +5,7 @@ from torchvision.transforms.v2 import functional as F, Transform
class UniformTemporalSubsample(Transform):
"""[BETA] Uniformly subsample ``num_samples`` indices from the temporal dimension of the video.
.. v2betastatus:: UniformTemporalSubsample transform
"""Uniformly subsample ``num_samples`` indices from the temporal dimension of the video.
Videos are expected to be of shape ``[..., T, C, H, W]`` where ``T`` denotes the temporal dimension.
......
torchvision/transforms/v2/_type_conversion.py
View file @ 7b4681a6
......@@ -11,9 +11,7 @@ from torchvision.transforms.v2._utils import is_pure_tensor
class PILToTensor(Transform):
"""[BETA] Convert a PIL Image to a tensor of the same type - this does not scale values.
.. v2betastatus:: PILToTensor transform
"""Convert a PIL Image to a tensor of the same type - this does not scale values.
This transform does not support torchscript.
......@@ -27,11 +25,9 @@ class PILToTensor(Transform):
class ToImage(Transform):
"""[BETA] Convert a tensor, ndarray, or PIL Image to :class:`~torchvision.tv_tensors.Image`
"""Convert a tensor, ndarray, or PIL Image to :class:`~torchvision.tv_tensors.Image`
; this does not scale values.
.. v2betastatus:: ToImage transform
This transform does not support torchscript.
"""
......@@ -44,9 +40,7 @@ class ToImage(Transform):
class ToPILImage(Transform):
"""[BETA] Convert a tensor or an ndarray to PIL Image
.. v2betastatus:: ToPILImage transform
"""Convert a tensor or an ndarray to PIL Image
This transform does not support torchscript.
......@@ -79,9 +73,7 @@ class ToPILImage(Transform):
class ToPureTensor(Transform):
"""[BETA] Convert all TVTensors to pure tensors, removing associated metadata (if any).
.. v2betastatus:: ToPureTensor transform
"""Convert all TVTensors to pure tensors, removing associated metadata (if any).
This doesn't scale or change the values, only the type.
"""
......
torchvision/transforms/v2/functional/_augment.py
View file @ 7b4681a6
......@@ -17,7 +17,7 @@ def erase(
v: torch.Tensor,
inplace: bool = False,
) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.v2.RandomErase` for details."""
"""See :class:`~torchvision.transforms.v2.RandomErase` for details."""
if torch.jit.is_scripting():
return erase_image(inpt, i=i, j=j, h=h, w=w, v=v, inplace=inplace)
......
torchvision/transforms/v2/functional/_color.py
View file @ 7b4681a6
......@@ -15,7 +15,7 @@ from ._utils import _get_kernel, _register_kernel_internal
def rgb_to_grayscale(inpt: torch.Tensor, num_output_channels: int = 1) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.v2.Grayscale` for details."""
"""See :class:`~torchvision.transforms.v2.Grayscale` for details."""
if torch.jit.is_scripting():
return rgb_to_grayscale_image(inpt, num_output_channels=num_output_channels)
......@@ -147,7 +147,7 @@ def adjust_saturation_video(video: torch.Tensor, saturation_factor: float) -> to
def adjust_contrast(inpt: torch.Tensor, contrast_factor: float) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.RandomAutocontrast`"""
"""See :class:`~torchvision.transforms.RandomAutocontrast`"""
if torch.jit.is_scripting():
return adjust_contrast_image(inpt, contrast_factor=contrast_factor)
......@@ -186,7 +186,7 @@ def adjust_contrast_video(video: torch.Tensor, contrast_factor: float) -> torch.
def adjust_sharpness(inpt: torch.Tensor, sharpness_factor: float) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.RandomAdjustSharpness`"""
"""See :class:`~torchvision.transforms.RandomAdjustSharpness`"""
if torch.jit.is_scripting():
return adjust_sharpness_image(inpt, sharpness_factor=sharpness_factor)
......@@ -417,7 +417,7 @@ def adjust_gamma_video(video: torch.Tensor, gamma: float, gain: float = 1) -> to
def posterize(inpt: torch.Tensor, bits: int) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.v2.RandomPosterize` for details."""
"""See :class:`~torchvision.transforms.v2.RandomPosterize` for details."""
if torch.jit.is_scripting():
return posterize_image(inpt, bits=bits)
......@@ -451,7 +451,7 @@ def posterize_video(video: torch.Tensor, bits: int) -> torch.Tensor:
def solarize(inpt: torch.Tensor, threshold: float) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.v2.RandomSolarize` for details."""
"""See :class:`~torchvision.transforms.v2.RandomSolarize` for details."""
if torch.jit.is_scripting():
return solarize_image(inpt, threshold=threshold)
......@@ -479,7 +479,7 @@ def solarize_video(video: torch.Tensor, threshold: float) -> torch.Tensor:
def autocontrast(inpt: torch.Tensor) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.v2.RandomAutocontrast` for details."""
"""See :class:`~torchvision.transforms.v2.RandomAutocontrast` for details."""
if torch.jit.is_scripting():
return autocontrast_image(inpt)
......@@ -529,7 +529,7 @@ def autocontrast_video(video: torch.Tensor) -> torch.Tensor:
def equalize(inpt: torch.Tensor) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.v2.RandomEqualize` for details."""
"""See :class:`~torchvision.transforms.v2.RandomEqualize` for details."""
if torch.jit.is_scripting():
return equalize_image(inpt)
......@@ -619,7 +619,7 @@ def equalize_video(video: torch.Tensor) -> torch.Tensor:
def invert(inpt: torch.Tensor) -> torch.Tensor:
"""[BETA] See :func:`~torchvision.transforms.v2.RandomInvert`."""
"""See :func:`~torchvision.transforms.v2.RandomInvert`."""
if torch.jit.is_scripting():
return invert_image(inpt)
......
torchvision/transforms/v2/functional/_deprecated.py
View file @ 7b4681a6
......@@ -8,7 +8,7 @@ from torchvision.transforms import functional as _F
@torch.jit.unused
def to_tensor(inpt: Any) -> torch.Tensor:
"""[BETA] [DEPREACTED] Use to_image() and to_dtype() instead."""
"""[DEPREACTED] Use to_image() and to_dtype() instead."""
warnings.warn(
"The function `to_tensor(...)` is deprecated and will be removed in a future release. "
"Instead, please use `to_image(...)` followed by `to_dtype(..., dtype=torch.float32, scale=True)`."
......
torchvision/transforms/v2/functional/_geometry.py
View file @ 7b4681a6
......@@ -39,7 +39,7 @@ def _check_interpolation(interpolation: Union[InterpolationMode, int]) -> Interp
def horizontal_flip(inpt: torch.Tensor) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.v2.RandomHorizontalFlip` for details."""
"""See :class:`~torchvision.transforms.v2.RandomHorizontalFlip` for details."""
if torch.jit.is_scripting():
return horizontal_flip_image(inpt)
......@@ -96,7 +96,7 @@ def horizontal_flip_video(video: torch.Tensor) -> torch.Tensor:
def vertical_flip(inpt: torch.Tensor) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.v2.RandomVerticalFlip` for details."""
"""See :class:`~torchvision.transforms.v2.RandomVerticalFlip` for details."""
if torch.jit.is_scripting():
return vertical_flip_image(inpt)
......@@ -178,7 +178,7 @@ def resize(
max_size: Optional[int] = None,
antialias: Optional[bool] = True,
) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.v2.Resize` for details."""
"""See :class:`~torchvision.transforms.v2.Resize` for details."""
if torch.jit.is_scripting():
return resize_image(inpt, size=size, interpolation=interpolation, max_size=max_size, antialias=antialias)
......@@ -373,7 +373,7 @@ def affine(
fill: _FillTypeJIT = None,
center: Optional[List[float]] = None,
) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.v2.RandomAffine` for details."""
"""See :class:`~torchvision.transforms.v2.RandomAffine` for details."""
if torch.jit.is_scripting():
return affine_image(
inpt,
......@@ -939,7 +939,7 @@ def rotate(
center: Optional[List[float]] = None,
fill: _FillTypeJIT = None,
) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.v2.RandomRotation` for details."""
"""See :class:`~torchvision.transforms.v2.RandomRotation` for details."""
if torch.jit.is_scripting():
return rotate_image(inpt, angle=angle, interpolation=interpolation, expand=expand, fill=fill, center=center)
......@@ -1094,7 +1094,7 @@ def pad(
fill: Optional[Union[int, float, List[float]]] = None,
padding_mode: str = "constant",
) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.v2.Pad` for details."""
"""See :class:`~torchvision.transforms.v2.Pad` for details."""
if torch.jit.is_scripting():
return pad_image(inpt, padding=padding, fill=fill, padding_mode=padding_mode)
......@@ -1314,7 +1314,7 @@ def pad_video(
def crop(inpt: torch.Tensor, top: int, left: int, height: int, width: int) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.v2.RandomCrop` for details."""
"""See :class:`~torchvision.transforms.v2.RandomCrop` for details."""
if torch.jit.is_scripting():
return crop_image(inpt, top=top, left=left, height=height, width=width)
......@@ -1408,7 +1408,7 @@ def perspective(
fill: _FillTypeJIT = None,
coefficients: Optional[List[float]] = None,
) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.v2.RandomPerspective` for details."""
"""See :class:`~torchvision.transforms.v2.RandomPerspective` for details."""
if torch.jit.is_scripting():
return perspective_image(
inpt,
......@@ -1696,7 +1696,7 @@ def elastic(
interpolation: Union[InterpolationMode, int] = InterpolationMode.BILINEAR,
fill: _FillTypeJIT = None,
) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.v2.ElasticTransform` for details."""
"""See :class:`~torchvision.transforms.v2.ElasticTransform` for details."""
if torch.jit.is_scripting():
return elastic_image(inpt, displacement=displacement, interpolation=interpolation, fill=fill)
......@@ -1880,7 +1880,7 @@ def elastic_video(
def center_crop(inpt: torch.Tensor, output_size: List[int]) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.v2.RandomCrop` for details."""
"""See :class:`~torchvision.transforms.v2.RandomCrop` for details."""
if torch.jit.is_scripting():
return center_crop_image(inpt, output_size=output_size)
......@@ -2009,7 +2009,7 @@ def resized_crop(
interpolation: Union[InterpolationMode, int] = InterpolationMode.BILINEAR,
antialias: Optional[bool] = True,
) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.v2.RandomResizedCrop` for details."""
"""See :class:`~torchvision.transforms.v2.RandomResizedCrop` for details."""
if torch.jit.is_scripting():
return resized_crop_image(
inpt,
......@@ -2154,7 +2154,7 @@ def resized_crop_video(
def five_crop(
inpt: torch.Tensor, size: List[int]
) -> Tuple[torch.Tensor, torch.Tensor, torch.Tensor, torch.Tensor, torch.Tensor]:
"""[BETA] See :class:`~torchvision.transforms.v2.FiveCrop` for details."""
"""See :class:`~torchvision.transforms.v2.FiveCrop` for details."""
if torch.jit.is_scripting():
return five_crop_image(inpt, size=size)
......@@ -2238,7 +2238,7 @@ def ten_crop(
torch.Tensor,
torch.Tensor,
]:
"""[BETA] See :class:`~torchvision.transforms.v2.TenCrop` for details."""
"""See :class:`~torchvision.transforms.v2.TenCrop` for details."""
if torch.jit.is_scripting():
return ten_crop_image(inpt, size=size, vertical_flip=vertical_flip)
......
torchvision/transforms/v2/functional/_meta.py
View file @ 7b4681a6
......@@ -203,7 +203,7 @@ def convert_bounding_box_format(
new_format: Optional[BoundingBoxFormat] = None,
inplace: bool = False,
) -> torch.Tensor:
"""[BETA] See :func:`~torchvision.transforms.v2.ConvertBoundingBoxFormat` for details."""
"""See :func:`~torchvision.transforms.v2.ConvertBoundingBoxFormat` for details."""
# This being a kernel / functional hybrid, we need an option to pass `old_format` explicitly for pure tensor
# inputs as well as extract it from `tv_tensors.BoundingBoxes` inputs. However, putting a default value on
# `old_format` means we also need to put one on `new_format` to have syntactically correct Python. Here we mimic the
......@@ -254,7 +254,7 @@ def clamp_bounding_boxes(
format: Optional[BoundingBoxFormat] = None,
canvas_size: Optional[Tuple[int, int]] = None,
) -> torch.Tensor:
"""[BETA] See :func:`~torchvision.transforms.v2.ClampBoundingBoxes` for details."""
"""See :func:`~torchvision.transforms.v2.ClampBoundingBoxes` for details."""
if not torch.jit.is_scripting():
_log_api_usage_once(clamp_bounding_boxes)
......
torchvision/transforms/v2/functional/_misc.py
View file @ 7b4681a6
......@@ -20,7 +20,7 @@ def normalize(
std: List[float],
inplace: bool = False,
) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.v2.Normalize` for details."""
"""See :class:`~torchvision.transforms.v2.Normalize` for details."""
if torch.jit.is_scripting():
return normalize_image(inpt, mean=mean, std=std, inplace=inplace)
......@@ -71,7 +71,7 @@ def normalize_video(video: torch.Tensor, mean: List[float], std: List[float], in
def gaussian_blur(inpt: torch.Tensor, kernel_size: List[int], sigma: Optional[List[float]] = None) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.v2.GaussianBlur` for details."""
"""See :class:`~torchvision.transforms.v2.GaussianBlur` for details."""
if torch.jit.is_scripting():
return gaussian_blur_image(inpt, kernel_size=kernel_size, sigma=sigma)
......@@ -180,7 +180,7 @@ def gaussian_blur_video(
def to_dtype(inpt: torch.Tensor, dtype: torch.dtype = torch.float, scale: bool = False) -> torch.Tensor:
"""[BETA] See :func:`~torchvision.transforms.v2.ToDtype` for details."""
"""See :func:`~torchvision.transforms.v2.ToDtype` for details."""
if torch.jit.is_scripting():
return to_dtype_image(inpt, dtype=dtype, scale=scale)
......@@ -261,7 +261,7 @@ def to_dtype_image(image: torch.Tensor, dtype: torch.dtype = torch.float, scale:
# We encourage users to use to_dtype() instead but we keep this for BC
def convert_image_dtype(image: torch.Tensor, dtype: torch.dtype = torch.float32) -> torch.Tensor:
"""[BETA] [DEPRECATED] Use to_dtype() instead."""
"""[DEPRECATED] Use to_dtype() instead."""
return to_dtype_image(image, dtype=dtype, scale=True)
......
torchvision/transforms/v2/functional/_temporal.py
View file @ 7b4681a6
......@@ -8,7 +8,7 @@ from ._utils import _get_kernel, _register_kernel_internal
def uniform_temporal_subsample(inpt: torch.Tensor, num_samples: int) -> torch.Tensor:
"""[BETA] See :class:`~torchvision.transforms.v2.UniformTemporalSubsample` for details."""
"""See :class:`~torchvision.transforms.v2.UniformTemporalSubsample` for details."""
if torch.jit.is_scripting():
return uniform_temporal_subsample_video(inpt, num_samples=num_samples)
......
torchvision/transforms/v2/functional/_type_conversion.py
View file @ 7b4681a6
......@@ -9,7 +9,7 @@ from torchvision.transforms import functional as _F
@torch.jit.unused
def to_image(inpt: Union[torch.Tensor, PIL.Image.Image, np.ndarray]) -> tv_tensors.Image:
"""[BETA] See :class:`~torchvision.transforms.v2.ToImage` for details."""
"""See :class:`~torchvision.transforms.v2.ToImage` for details."""
if isinstance(inpt, np.ndarray):
output = torch.from_numpy(inpt).permute((2, 0, 1)).contiguous()
elif isinstance(inpt, PIL.Image.Image):
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment