forked from mindspore-Ecosystem/mindspore
Update doc for Unify Transform API - stage 2
This commit is contained in:
luoyang
parent
5c565a0520
commit
076d141913
|
|
@ -0,0 +1,20 @@
|
|||
mindspore.dataset.transforms.Compose
|
||||
====================================
|
||||
|
||||
.. py:class:: mindspore.dataset.transforms.Compose(transforms)
|
||||
|
||||
将多个数据增强算子组合使用。
|
||||
|
||||
.. note::
|
||||
Compose可以将 `mindspore.dataset.transforms` / `mindspore.dataset.vision` 等模块中的数据增强算子以及用户自定义的Python可调用对象
|
||||
合并成单个数据增强。对于用户定义的Python可调用对象，要求其返回值是numpy.ndarray类型。
|
||||
|
||||
**参数：**
|
||||
|
||||
- **transforms** (list) - 一个数据增强的列表。
|
||||
|
||||
**异常：**
|
||||
|
||||
- **TypeError** - 参数 `transforms` 类型不为list。
|
||||
- **ValueError** - 参数 `transforms` 是空的list。
|
||||
- **TypeError** - 参数 `transforms` 的元素不是Python的可调用对象或audio/text/transforms/vision模块中的数据增强方法。
|
||||
|
|
@ -1,7 +1,7 @@
|
|||
mindspore.dataset.transforms.c_transforms.Concatenate
|
||||
=====================================================
|
||||
mindspore.dataset.transforms.Concatenate
|
||||
========================================
|
||||
|
||||
.. py:class:: mindspore.dataset.transforms.c_transforms.Concatenate(axis=0, prepend=None, append=None)
|
||||
.. py:class:: mindspore.dataset.transforms.Concatenate(axis=0, prepend=None, append=None)
|
||||
|
||||
在Tensor的某一个轴上进行元素拼接。
|
||||
|
||||
|
|
@ -1,7 +1,7 @@
|
|||
mindspore.dataset.transforms.c_transforms.Duplicate
|
||||
===================================================
|
||||
mindspore.dataset.transforms.Duplicate
|
||||
======================================
|
||||
|
||||
.. py:class:: mindspore.dataset.transforms.c_transforms.Duplicate()
|
||||
.. py:class:: mindspore.dataset.transforms.Duplicate()
|
||||
|
||||
将输入的数据列复制得到新的数据列，每次仅可以输入1个数据列进行复制。
|
||||
|
||||
|
|
@ -1,7 +1,7 @@
|
|||
mindspore.dataset.transforms.c_transforms.Fill
|
||||
==============================================
|
||||
mindspore.dataset.transforms.Fill
|
||||
=================================
|
||||
|
||||
.. py:class:: mindspore.dataset.transforms.c_transforms.Fill(fill_value)
|
||||
.. py:class:: mindspore.dataset.transforms.Fill(fill_value)
|
||||
|
||||
将Tensor的所有元素都赋值为指定的值，输出Tensor将与输入Tensor具有与具有相同的shape和数据类型。
|
||||
|
||||
|
|
@ -1,7 +1,7 @@
|
|||
mindspore.dataset.transforms.c_transforms.Mask
|
||||
==============================================
|
||||
mindspore.dataset.transforms.Mask
|
||||
=================================
|
||||
|
||||
.. py:class:: mindspore.dataset.transforms.c_transforms.Mask(operator, constant, dtype=mstype.bool_)
|
||||
.. py:class:: mindspore.dataset.transforms.Mask(operator, constant, dtype=mstype.bool_)
|
||||
|
||||
用给条件判断输入Tensor的内容，并返回一个掩码Tensor。Tensor中任何符合条件的元素都将被标记为True，否则为False。
|
||||
|
||||
|
|
@ -1,9 +1,9 @@
|
|||
mindspore.dataset.transforms.py_transforms.OneHotOp
|
||||
===================================================
|
||||
mindspore.dataset.transforms.OneHot
|
||||
===================================
|
||||
|
||||
.. py:class:: mindspore.dataset.transforms.py_transforms.OneHotOp(num_classes, smoothing_rate=0.0)
|
||||
.. py:class:: mindspore.dataset.transforms.OneHot(num_classes, smoothing_rate=0.0)
|
||||
|
||||
将Tensor进行OneHot编码，可以进一步对标签进行平滑处理。
|
||||
将Tensor进行OneHot编码。
|
||||
|
||||
**参数：**
|
||||
|
||||
|
|
@ -11,7 +11,9 @@ mindspore.dataset.transforms.py_transforms.OneHotOp
|
|||
- **smoothing_rate** (float，可选) - 标签平滑的系数，默认值：0.0。
|
||||
|
||||
**异常：**
|
||||
|
||||
|
||||
- **TypeError** - 参数 `num_classes` 类型不为int。
|
||||
- **TypeError** - 参数 `smoothing_rate` 类型不为float。
|
||||
- **ValueError** - 参数 `smoothing_rate` 取值范围不为[0.0, 1.0]。
|
||||
- **RuntimeError** - 输入Tensor的数据类型不为int。
|
||||
- **RuntimeError** - 参数Tensor的shape不是1-D。
|
||||
|
|
@ -1,7 +1,7 @@
|
|||
mindspore.dataset.transforms.c_transforms.PadEnd
|
||||
================================================
|
||||
mindspore.dataset.transforms.PadEnd
|
||||
===================================
|
||||
|
||||
.. py:class:: mindspore.dataset.transforms.c_transforms.PadEnd(pad_shape, pad_value=None)
|
||||
.. py:class:: mindspore.dataset.transforms.PadEnd(pad_shape, pad_value=None)
|
||||
|
||||
对输入Tensor进行填充，要求 `pad_shape` 与输入Tensor的维度保持一致。
|
||||
|
||||
|
|
@ -1,7 +1,7 @@
|
|||
mindspore.dataset.transforms.c_transforms.RandomApply
|
||||
=====================================================
|
||||
mindspore.dataset.transforms.RandomApply
|
||||
========================================
|
||||
|
||||
.. py:class:: mindspore.dataset.transforms.c_transforms.RandomApply(transforms, prob=0.5)
|
||||
.. py:class:: mindspore.dataset.transforms.RandomApply(transforms, prob=0.5)
|
||||
|
||||
指定一组数据增强处理及其被应用的概率，在运算时按概率随机应用其中的增强处理。
|
||||
|
||||
|
|
@ -14,6 +14,6 @@ mindspore.dataset.transforms.c_transforms.RandomApply
|
|||
|
||||
- **TypeError** - 参数 `transforms` 类型不为list。
|
||||
- **ValueError** - 参数 `transforms` 的长度为空。
|
||||
- **TypeError** - 参数 `transforms` 的元素不是Python可调用对象或c_transforms模块中的数据处理操作。
|
||||
- **TypeError** - 参数 `transforms` 的元素不是Python可调用对象或audio/text/transforms/vision模块中的数据处理操作。
|
||||
- **TypeError** - 参数 `prob` 的类型不为float。
|
||||
- **ValueError** - 参数 `prob` 的取值范围不为[0.0, 1.0]。
|
||||
|
|
@ -1,7 +1,7 @@
|
|||
mindspore.dataset.transforms.c_transforms.RandomChoice
|
||||
======================================================
|
||||
mindspore.dataset.transforms.RandomChoice
|
||||
=========================================
|
||||
|
||||
.. py:class:: mindspore.dataset.transforms.c_transforms.RandomChoice(transforms)
|
||||
.. py:class:: mindspore.dataset.transforms.RandomChoice(transforms)
|
||||
|
||||
在一组数据增强中随机选择部分增强处理进行应用。
|
||||
|
||||
|
|
@ -13,4 +13,4 @@ mindspore.dataset.transforms.c_transforms.RandomChoice
|
|||
|
||||
- **TypeError** - 参数 `transforms` 类型不为list。
|
||||
- **ValueError** - 参数 `transforms` 是空的list。
|
||||
- **TypeError** - 参数 `transforms` 的元素不是Python可调用对象或c_transforms模块中的数据处理操作。
|
||||
- **TypeError** - 参数 `transforms` 的元素不是Python可调用对象或audio/text/transforms/vision模块中的数据处理操作。
|
||||
|
|
@ -1,7 +1,7 @@
|
|||
mindspore.dataset.transforms.py_transforms.RandomOrder
|
||||
mindspore.dataset.transforms.RandomOrder
|
||||
======================================================
|
||||
|
||||
.. py:class:: mindspore.dataset.transforms.py_transforms.RandomOrder(transforms)
|
||||
.. py:class:: mindspore.dataset.transforms.RandomOrder(transforms)
|
||||
|
||||
给一个数据增强的列表，随机打乱数据增强处理的顺序。
|
||||
|
||||
|
|
@ -12,5 +12,5 @@ mindspore.dataset.transforms.py_transforms.RandomOrder
|
|||
**异常：**
|
||||
|
||||
- **TypeError** - 参数 `transforms` 类型不为list。
|
||||
- **TypeError** - 参数 `transforms` 的元素不是Python可调用对象或py_transforms模块中的数据处理操作。
|
||||
- **TypeError** - 参数 `transforms` 的元素不是Python可调用对象或audio/text/transforms/vision模块中的数据处理操作。
|
||||
- **ValueError** - 参数 `transforms` 是空的list。
|
||||
|
|
@ -1,7 +1,7 @@
|
|||
mindspore.dataset.transforms.c_transforms.Relational
|
||||
====================================================
|
||||
mindspore.dataset.transforms.Relational
|
||||
=======================================
|
||||
|
||||
.. py:class:: mindspore.dataset.transforms.c_transforms.Relational
|
||||
.. py:class:: mindspore.dataset.transforms.Relational
|
||||
|
||||
关系操作符，可以取值为Relational.EQ、Relational.NE、Relational.GT、Relational.GE、Relational.LT、Relational.LE。
|
||||
|
||||
|
|
@ -1,7 +1,7 @@
|
|||
mindspore.dataset.transforms.c_transforms.Slice
|
||||
===============================================
|
||||
mindspore.dataset.transforms.Slice
|
||||
==================================
|
||||
|
||||
.. py:class:: mindspore.dataset.transforms.c_transforms.Slice(*slices)
|
||||
.. py:class:: mindspore.dataset.transforms.Slice(*slices)
|
||||
|
||||
对Tensor进行切片操作，功能类似于NumPy的索引(目前只支持1D形状的Tensor)。
|
||||
|
||||
|
|
@ -1,7 +1,7 @@
|
|||
mindspore.dataset.transforms.c_transforms.TypeCast
|
||||
==================================================
|
||||
mindspore.dataset.transforms.TypeCast
|
||||
=====================================
|
||||
|
||||
.. py:class:: mindspore.dataset.transforms.c_transforms.TypeCast(data_type)
|
||||
.. py:class:: mindspore.dataset.transforms.TypeCast(data_type)
|
||||
|
||||
将输入的Tensor转换为指定的数据类型。
|
||||
|
||||
|
|
@ -1,7 +1,7 @@
|
|||
mindspore.dataset.transforms.c_transforms.Unique
|
||||
================================================
|
||||
mindspore.dataset.transforms.Unique
|
||||
===================================
|
||||
|
||||
.. py:class:: mindspore.dataset.transforms.c_transforms.Unique()
|
||||
.. py:class:: mindspore.dataset.transforms.Unique()
|
||||
|
||||
对输入张量进行唯一运算，每次只支持对一个数据列进行变换。
|
||||
|
||||
|
|
@ -1,16 +0,0 @@
|
|||
mindspore.dataset.transforms.c_transforms.Compose
|
||||
=================================================
|
||||
|
||||
.. py:class:: mindspore.dataset.transforms.c_transforms.Compose(transforms)
|
||||
|
||||
将多个数据增强算子组合使用。
|
||||
|
||||
**参数：**
|
||||
|
||||
- **transforms** (list) - 一个数据增强的列表。
|
||||
|
||||
**异常：**
|
||||
|
||||
- **TypeError** - 参数 `transforms` 类型不为list。
|
||||
- **ValueError** - 参数 `transforms` 是空的list。
|
||||
- **TypeError** - 参数 `transforms` 的元素不是Python可调用对象或c_transforms模块中的数据处理操作。
|
||||
|
|
@ -1,16 +0,0 @@
|
|||
mindspore.dataset.transforms.c_transforms.OneHot
|
||||
================================================
|
||||
|
||||
.. py:class:: mindspore.dataset.transforms.c_transforms.OneHot(num_classes)
|
||||
|
||||
将Tensor进行OneHot编码。
|
||||
|
||||
**参数：**
|
||||
|
||||
- **num_classes** (int) - 数据集的类别数，它应该大于数据集中最大的label编号。
|
||||
|
||||
**异常：**
|
||||
|
||||
- **TypeError** - 参数 `num_classes` 类型不为int。
|
||||
- **RuntimeError** - 输入Tensor的数据类型不为int。
|
||||
- **RuntimeError** - 参数Tensor的shape不是1-D。
|
||||
|
|
@ -1,33 +0,0 @@
|
|||
mindspore.dataset.transforms.py_transforms.Compose
|
||||
==================================================
|
||||
|
||||
.. py:class:: mindspore.dataset.transforms.py_transforms.Compose(transforms)
|
||||
|
||||
将多个数据增强算子组合使用。
|
||||
|
||||
.. note::
|
||||
Compose可以将 `mindspore.dataset.transforms.py_transforms` 模块中的数据增强算子以及用户自定义的Python可调用对象
|
||||
合并成单个数据增强。对于用户定义的Python可调用对象，要求其返回值是numpy.ndarray类型。有关如何使用，请参阅Compose的示例，或阅读
|
||||
:class:`mindspore.dataset.vision.py_transforms.FiveCrop` 的示例，学习如何与用户自定义Python可调用对象配合使用。
|
||||
|
||||
**参数：**
|
||||
|
||||
- **transforms** (list) - 一个数据增强的列表。
|
||||
|
||||
**异常：**
|
||||
|
||||
- **TypeError** - 参数 `transforms` 类型不为list。
|
||||
- **ValueError** - 参数 `transforms` 是空的list。
|
||||
- **TypeError** - 参数 `transforms` 的元素不是Python的可调用对象。
|
||||
|
||||
.. py:method:: reduce(operations)
|
||||
|
||||
使用Compose将指定数据增强操作列表中相邻的Python操作组合，以允许混用Python和C++操作。
|
||||
|
||||
**参数：**
|
||||
|
||||
- **operations** (list) - 数据增强的列表。
|
||||
|
||||
**返回：**
|
||||
|
||||
list，组合后的数据增强操作列表。
|
||||
|
|
@ -1,19 +0,0 @@
|
|||
mindspore.dataset.transforms.py_transforms.RandomApply
|
||||
======================================================
|
||||
|
||||
.. py:class:: mindspore.dataset.transforms.py_transforms.RandomApply(transforms, prob=0.5)
|
||||
|
||||
指定一组数据增强处理及其被应用的概率，在运算时按概率随机应用其中的增强处理。
|
||||
|
||||
**参数：**
|
||||
|
||||
- **transforms** (list) - 一个数据增强的列表。
|
||||
- **prob** (float, 可选) - 随机应用某个数据增强的概率，默认值：0.5。
|
||||
|
||||
**异常：**
|
||||
|
||||
- **TypeError** - 参数 `transforms` 类型不为list。
|
||||
- **ValueError** - 参数 `transforms` 为空的list。
|
||||
- **TypeError** - 参数 `transforms` 的元素不是Python可调用对象或py_transforms模块中的数据处理操作。
|
||||
- **TypeError** - 参数 `prob` 的类型不为float。
|
||||
- **ValueError** - 参数 `prob` 的取值范围不为[0.0, 1.0]。
|
||||
|
|
@ -1,16 +0,0 @@
|
|||
mindspore.dataset.transforms.py_transforms.RandomChoice
|
||||
=======================================================
|
||||
|
||||
.. py:class:: mindspore.dataset.transforms.py_transforms.RandomChoice(transforms)
|
||||
|
||||
在一组数据增强中随机选择部分增强处理进行应用。
|
||||
|
||||
**参数：**
|
||||
|
||||
- **transforms** (list) - 一个数据增强的列表。
|
||||
|
||||
**异常：**
|
||||
|
||||
- **TypeError** - 参数 `transforms` 类型不为list。
|
||||
- **TypeError** - 参数 `transforms` 的元素不是Python可调用对象或py_transforms模块中的数据处理操作。
|
||||
- **ValueError** - 参数 `transforms` 是空的list。
|
||||
|
|
@ -1,19 +1,19 @@
|
|||
mindspore.dataset.transforms
|
||||
============================
|
||||
|
||||
此模块用于通用数据增强，包括 `c_transforms` 和 `py_transforms` 两个子模块。
|
||||
|
||||
`c_transforms` 是一个高性能数据增强模块，基于C++实现。
|
||||
|
||||
而 `py_transforms` 提供了一种基于Python和NumPy的实现方式。
|
||||
此模块用于通用数据增强，其中一部分增强操作是用C++实现的，具有较好的高性能，另一部分是基于Python实现，使用了NumPy模块作为支持。
|
||||
|
||||
在API示例中，常用的模块导入方法如下：
|
||||
|
||||
.. code-block::
|
||||
|
||||
import mindspore.dataset as ds
|
||||
import mindspore.dataset.vision.c_transforms as c_vision
|
||||
import mindspore.dataset.vision.py_transforms as py_vision
|
||||
import mindspore.dataset.transforms as transforms
|
||||
|
||||
注意：旧的API导入方式已经过时且会逐步废弃，因此推荐使用上面的方式，但目前仍可按以下方式导入：
|
||||
|
||||
.. code-block::
|
||||
|
||||
from mindspore.dataset.transforms import c_transforms
|
||||
from mindspore.dataset.transforms import py_transforms
|
||||
|
||||
|
|
@ -22,38 +22,34 @@ mindspore.dataset.transforms
|
|||
- TensorOperation，所有C++实现的数据处理操作的基类。
|
||||
- PyTensorOperation，所有Python实现的数据处理操作的基类。
|
||||
|
||||
mindspore.dataset.transforms.c_transforms
|
||||
-----------------------------------------
|
||||
Transforms
|
||||
----------
|
||||
|
||||
.. mscnautosummary::
|
||||
:toctree: dataset_transforms
|
||||
:nosignatures:
|
||||
:template: classtemplate.rst
|
||||
|
||||
mindspore.dataset.transforms.c_transforms.Compose
|
||||
mindspore.dataset.transforms.c_transforms.Concatenate
|
||||
mindspore.dataset.transforms.c_transforms.Duplicate
|
||||
mindspore.dataset.transforms.c_transforms.Fill
|
||||
mindspore.dataset.transforms.c_transforms.Mask
|
||||
mindspore.dataset.transforms.c_transforms.OneHot
|
||||
mindspore.dataset.transforms.c_transforms.PadEnd
|
||||
mindspore.dataset.transforms.c_transforms.RandomApply
|
||||
mindspore.dataset.transforms.c_transforms.RandomChoice
|
||||
mindspore.dataset.transforms.c_transforms.Relational
|
||||
mindspore.dataset.transforms.c_transforms.Slice
|
||||
mindspore.dataset.transforms.c_transforms.TypeCast
|
||||
mindspore.dataset.transforms.c_transforms.Unique
|
||||
mindspore.dataset.transforms.Compose
|
||||
mindspore.dataset.transforms.Concatenate
|
||||
mindspore.dataset.transforms.Duplicate
|
||||
mindspore.dataset.transforms.Fill
|
||||
mindspore.dataset.transforms.Mask
|
||||
mindspore.dataset.transforms.OneHot
|
||||
mindspore.dataset.transforms.PadEnd
|
||||
mindspore.dataset.transforms.RandomApply
|
||||
mindspore.dataset.transforms.RandomChoice
|
||||
mindspore.dataset.transforms.RandomOrder
|
||||
mindspore.dataset.transforms.Slice
|
||||
mindspore.dataset.transforms.TypeCast
|
||||
mindspore.dataset.transforms.Unique
|
||||
|
||||
mindspore.dataset.transforms.py_transforms
|
||||
------------------------------------------
|
||||
Others
|
||||
------
|
||||
|
||||
.. mscnautosummary::
|
||||
:toctree: dataset_transforms
|
||||
:nosignatures:
|
||||
:template: classtemplate.rst
|
||||
|
||||
mindspore.dataset.transforms.py_transforms.Compose
|
||||
mindspore.dataset.transforms.py_transforms.OneHotOp
|
||||
mindspore.dataset.transforms.py_transforms.RandomApply
|
||||
mindspore.dataset.transforms.py_transforms.RandomChoice
|
||||
mindspore.dataset.transforms.py_transforms.RandomOrder
|
||||
mindspore.dataset.transforms.Relational
|
||||
|
|
|
|||
|
|
@ -1,12 +1,18 @@
|
|||
mindspore.dataset.vision
|
||||
===================================
|
||||
|
||||
此模块用于图像数据增强，包括 `c_transforms` 和 `py_transforms` 两个子模块。
|
||||
`c_transforms` 是使用 C++ OpenCv 开发的高性能图像增强模块。
|
||||
`py_transforms` 是使用 Python Pillow 开发的图像增强模块。
|
||||
此模块用于图像数据增强，其中有一部分增强是基于C++ OpenCV实现的，具有较好的性能，而另一部分是基于Python Pillow实现的。
|
||||
|
||||
API样例中常用的导入模块如下：
|
||||
|
||||
.. code-block::
|
||||
|
||||
import mindspore.dataset as ds
|
||||
import mindspore.dataset.vision as vision
|
||||
import mindspore.dataset.vision.utils as utils
|
||||
|
||||
注意：旧的API导入方式已经过时且会逐步废弃，因此推荐使用上面的方式，但目前仍可按以下方式导入：
|
||||
|
||||
.. code-block::
|
||||
|
||||
import mindspore.dataset.vision.c_transforms as c_vision
|
||||
|
|
@ -17,113 +23,90 @@ API样例中常用的导入模块如下：
|
|||
|
||||
- TensorOperation，所有C++实现的数据处理操作的基类。
|
||||
- PyTensorOperation，所有Python实现的数据处理操作的基类。
|
||||
- ImageTensorOperation，所有图像数据处理操作的基类，派生自TensorOperation。
|
||||
|
||||
mindspore.dataset.vision.c_transforms
|
||||
------------------------------------------------
|
||||
Transforms
|
||||
----------
|
||||
|
||||
.. mscnautosummary::
|
||||
:toctree: dataset_vision
|
||||
:nosignatures:
|
||||
:template: classtemplate.rst
|
||||
|
||||
mindspore.dataset.vision.c_transforms.AutoContrast
|
||||
mindspore.dataset.vision.c_transforms.BoundingBoxAugment
|
||||
mindspore.dataset.vision.c_transforms.CenterCrop
|
||||
mindspore.dataset.vision.c_transforms.ConvertColor
|
||||
mindspore.dataset.vision.c_transforms.Crop
|
||||
mindspore.dataset.vision.c_transforms.CutMixBatch
|
||||
mindspore.dataset.vision.c_transforms.CutOut
|
||||
mindspore.dataset.vision.c_transforms.Decode
|
||||
mindspore.dataset.vision.c_transforms.Equalize
|
||||
mindspore.dataset.vision.c_transforms.GaussianBlur
|
||||
mindspore.dataset.vision.c_transforms.HorizontalFlip
|
||||
mindspore.dataset.vision.c_transforms.HWC2CHW
|
||||
mindspore.dataset.vision.c_transforms.Invert
|
||||
mindspore.dataset.vision.c_transforms.MixUpBatch
|
||||
mindspore.dataset.vision.c_transforms.Normalize
|
||||
mindspore.dataset.vision.c_transforms.NormalizePad
|
||||
mindspore.dataset.vision.c_transforms.Pad
|
||||
mindspore.dataset.vision.c_transforms.RandomAffine
|
||||
mindspore.dataset.vision.c_transforms.RandomColor
|
||||
mindspore.dataset.vision.c_transforms.RandomColorAdjust
|
||||
mindspore.dataset.vision.c_transforms.RandomCrop
|
||||
mindspore.dataset.vision.c_transforms.RandomCropDecodeResize
|
||||
mindspore.dataset.vision.c_transforms.RandomCropWithBBox
|
||||
mindspore.dataset.vision.c_transforms.RandomHorizontalFlip
|
||||
mindspore.dataset.vision.c_transforms.RandomHorizontalFlipWithBBox
|
||||
mindspore.dataset.vision.c_transforms.RandomPosterize
|
||||
mindspore.dataset.vision.c_transforms.RandomResize
|
||||
mindspore.dataset.vision.c_transforms.RandomResizedCrop
|
||||
mindspore.dataset.vision.c_transforms.RandomResizedCropWithBBox
|
||||
mindspore.dataset.vision.c_transforms.RandomResizeWithBBox
|
||||
mindspore.dataset.vision.c_transforms.RandomRotation
|
||||
mindspore.dataset.vision.c_transforms.RandomSelectSubpolicy
|
||||
mindspore.dataset.vision.c_transforms.RandomSharpness
|
||||
mindspore.dataset.vision.c_transforms.RandomSolarize
|
||||
mindspore.dataset.vision.c_transforms.RandomVerticalFlip
|
||||
mindspore.dataset.vision.c_transforms.RandomVerticalFlipWithBBox
|
||||
mindspore.dataset.vision.c_transforms.Rescale
|
||||
mindspore.dataset.vision.c_transforms.Resize
|
||||
mindspore.dataset.vision.c_transforms.ResizeWithBBox
|
||||
mindspore.dataset.vision.c_transforms.Rotate
|
||||
mindspore.dataset.vision.c_transforms.SlicePatches
|
||||
mindspore.dataset.vision.c_transforms.SoftDvppDecodeRandomCropResizeJpeg
|
||||
mindspore.dataset.vision.c_transforms.SoftDvppDecodeResizeJpeg
|
||||
mindspore.dataset.vision.c_transforms.UniformAugment
|
||||
mindspore.dataset.vision.c_transforms.VerticalFlip
|
||||
mindspore.dataset.vision.AdjustGamma
|
||||
mindspore.dataset.vision.AutoAugment
|
||||
mindspore.dataset.vision.AutoContrast
|
||||
mindspore.dataset.vision.BoundingBoxAugment
|
||||
mindspore.dataset.vision.CenterCrop
|
||||
mindspore.dataset.vision.ConvertColor
|
||||
mindspore.dataset.vision.Crop
|
||||
mindspore.dataset.vision.CutMixBatch
|
||||
mindspore.dataset.vision.CutOut
|
||||
mindspore.dataset.vision.Decode
|
||||
mindspore.dataset.vision.Equalize
|
||||
mindspore.dataset.vision.FiveCrop
|
||||
mindspore.dataset.vision.GaussianBlur
|
||||
mindspore.dataset.vision.GrayScale
|
||||
mindspore.dataset.vision.HorizontalFlip
|
||||
mindspore.dataset.vision.HsvToRgb
|
||||
mindspore.dataset.vision.HWC2CHW
|
||||
mindspore.dataset.vision.Invert
|
||||
mindspore.dataset.vision.LinearTransformation
|
||||
mindspore.dataset.vision.MixUp
|
||||
mindspore.dataset.vision.MixUpBatch
|
||||
mindspore.dataset.vision.Normalize
|
||||
mindspore.dataset.vision.NormalizePad
|
||||
mindspore.dataset.vision.Pad
|
||||
mindspore.dataset.vision.PadToSize
|
||||
mindspore.dataset.vision.RandomAdjustSharpness
|
||||
mindspore.dataset.vision.RandomAffine
|
||||
mindspore.dataset.vision.RandomAutoContrast
|
||||
mindspore.dataset.vision.RandomColor
|
||||
mindspore.dataset.vision.RandomColorAdjust
|
||||
mindspore.dataset.vision.RandomCrop
|
||||
mindspore.dataset.vision.RandomCropDecodeResize
|
||||
mindspore.dataset.vision.RandomCropWithBBox
|
||||
mindspore.dataset.vision.RandomEqualize
|
||||
mindspore.dataset.vision.RandomErasing
|
||||
mindspore.dataset.vision.RandomGrayscale
|
||||
mindspore.dataset.vision.RandomHorizontalFlip
|
||||
mindspore.dataset.vision.RandomHorizontalFlipWithBBox
|
||||
mindspore.dataset.vision.RandomInvert
|
||||
mindspore.dataset.vision.RandomLighting
|
||||
mindspore.dataset.vision.RandomPerspective
|
||||
mindspore.dataset.vision.RandomPosterize
|
||||
mindspore.dataset.vision.RandomResizedCrop
|
||||
mindspore.dataset.vision.RandomResizedCropWithBBox
|
||||
mindspore.dataset.vision.RandomResize
|
||||
mindspore.dataset.vision.RandomResizeWithBBox
|
||||
mindspore.dataset.vision.RandomRotation
|
||||
mindspore.dataset.vision.RandomSelectSubpolicy
|
||||
mindspore.dataset.vision.RandomSharpness
|
||||
mindspore.dataset.vision.RandomSolarize
|
||||
mindspore.dataset.vision.RandomVerticalFlip
|
||||
mindspore.dataset.vision.RandomVerticalFlipWithBBox
|
||||
mindspore.dataset.vision.Rescale
|
||||
mindspore.dataset.vision.Resize
|
||||
mindspore.dataset.vision.ResizeWithBBox
|
||||
mindspore.dataset.vision.RgbToHsv
|
||||
mindspore.dataset.vision.Rotate
|
||||
mindspore.dataset.vision.SlicePatches
|
||||
mindspore.dataset.vision.TenCrop
|
||||
mindspore.dataset.vision.ToNumpy
|
||||
mindspore.dataset.vision.ToPIL
|
||||
mindspore.dataset.vision.ToTensor
|
||||
mindspore.dataset.vision.ToType
|
||||
mindspore.dataset.vision.UniformAugment
|
||||
mindspore.dataset.vision.VerticalFlip
|
||||
|
||||
mindspore.dataset.vision.py_transforms
|
||||
-------------------------------------------------
|
||||
|
||||
.. mscnautosummary::
|
||||
:toctree: dataset_vision
|
||||
:nosignatures:
|
||||
:template: classtemplate.rst
|
||||
|
||||
mindspore.dataset.vision.py_transforms.AutoContrast
|
||||
mindspore.dataset.vision.py_transforms.CenterCrop
|
||||
mindspore.dataset.vision.py_transforms.Cutout
|
||||
mindspore.dataset.vision.py_transforms.Decode
|
||||
mindspore.dataset.vision.py_transforms.Equalize
|
||||
mindspore.dataset.vision.py_transforms.FiveCrop
|
||||
mindspore.dataset.vision.py_transforms.Grayscale
|
||||
mindspore.dataset.vision.py_transforms.HsvToRgb
|
||||
mindspore.dataset.vision.py_transforms.HWC2CHW
|
||||
mindspore.dataset.vision.py_transforms.Invert
|
||||
mindspore.dataset.vision.py_transforms.LinearTransformation
|
||||
mindspore.dataset.vision.py_transforms.MixUp
|
||||
mindspore.dataset.vision.py_transforms.Normalize
|
||||
mindspore.dataset.vision.py_transforms.NormalizePad
|
||||
mindspore.dataset.vision.py_transforms.Pad
|
||||
mindspore.dataset.vision.py_transforms.RandomAffine
|
||||
mindspore.dataset.vision.py_transforms.RandomColor
|
||||
mindspore.dataset.vision.py_transforms.RandomColorAdjust
|
||||
mindspore.dataset.vision.py_transforms.RandomCrop
|
||||
mindspore.dataset.vision.py_transforms.RandomErasing
|
||||
mindspore.dataset.vision.py_transforms.RandomGrayscale
|
||||
mindspore.dataset.vision.py_transforms.RandomHorizontalFlip
|
||||
mindspore.dataset.vision.py_transforms.RandomPerspective
|
||||
mindspore.dataset.vision.py_transforms.RandomResizedCrop
|
||||
mindspore.dataset.vision.py_transforms.RandomRotation
|
||||
mindspore.dataset.vision.py_transforms.RandomSharpness
|
||||
mindspore.dataset.vision.py_transforms.RandomVerticalFlip
|
||||
mindspore.dataset.vision.py_transforms.Resize
|
||||
mindspore.dataset.vision.py_transforms.RgbToHsv
|
||||
mindspore.dataset.vision.py_transforms.TenCrop
|
||||
mindspore.dataset.vision.py_transforms.ToPIL
|
||||
mindspore.dataset.vision.py_transforms.ToTensor
|
||||
mindspore.dataset.vision.py_transforms.ToType
|
||||
mindspore.dataset.vision.py_transforms.UniformAugment
|
||||
|
||||
mindspore.dataset.vision.utils
|
||||
-------------------------------
|
||||
Others
|
||||
------
|
||||
|
||||
.. mscnautosummary::
|
||||
:toctree: dataset_vision
|
||||
:nosignatures:
|
||||
:template: classtemplate.rst
|
||||
|
||||
mindspore.dataset.vision.AutoAugmentPolicy
|
||||
mindspore.dataset.vision.Border
|
||||
mindspore.dataset.vision.ConvertMode
|
||||
mindspore.dataset.vision.ImageBatchFormat
|
||||
|
|
|
|||
|
|
@ -3,38 +3,34 @@ mindspore.dataset.transforms
|
|||
|
||||
.. automodule:: mindspore.dataset.transforms
|
||||
|
||||
mindspore.dataset.transforms.c_transforms
|
||||
-----------------------------------------
|
||||
Transforms
|
||||
----------
|
||||
|
||||
.. autosummary::
|
||||
:toctree: dataset_transforms
|
||||
:nosignatures:
|
||||
:template: classtemplate.rst
|
||||
|
||||
mindspore.dataset.transforms.c_transforms.Compose
|
||||
mindspore.dataset.transforms.c_transforms.Concatenate
|
||||
mindspore.dataset.transforms.c_transforms.Duplicate
|
||||
mindspore.dataset.transforms.c_transforms.Fill
|
||||
mindspore.dataset.transforms.c_transforms.Mask
|
||||
mindspore.dataset.transforms.c_transforms.OneHot
|
||||
mindspore.dataset.transforms.c_transforms.PadEnd
|
||||
mindspore.dataset.transforms.c_transforms.RandomApply
|
||||
mindspore.dataset.transforms.c_transforms.RandomChoice
|
||||
mindspore.dataset.transforms.c_transforms.Relational
|
||||
mindspore.dataset.transforms.c_transforms.Slice
|
||||
mindspore.dataset.transforms.c_transforms.TypeCast
|
||||
mindspore.dataset.transforms.c_transforms.Unique
|
||||
mindspore.dataset.transforms.Compose
|
||||
mindspore.dataset.transforms.Concatenate
|
||||
mindspore.dataset.transforms.Duplicate
|
||||
mindspore.dataset.transforms.Fill
|
||||
mindspore.dataset.transforms.Mask
|
||||
mindspore.dataset.transforms.OneHot
|
||||
mindspore.dataset.transforms.PadEnd
|
||||
mindspore.dataset.transforms.RandomApply
|
||||
mindspore.dataset.transforms.RandomChoice
|
||||
mindspore.dataset.transforms.RandomOrder
|
||||
mindspore.dataset.transforms.Slice
|
||||
mindspore.dataset.transforms.TypeCast
|
||||
mindspore.dataset.transforms.Unique
|
||||
|
||||
mindspore.dataset.transforms.py_transforms
|
||||
------------------------------------------
|
||||
Others
|
||||
------
|
||||
|
||||
.. autosummary::
|
||||
:toctree: dataset_transforms
|
||||
:nosignatures:
|
||||
:template: classtemplate.rst
|
||||
|
||||
mindspore.dataset.transforms.py_transforms.Compose
|
||||
mindspore.dataset.transforms.py_transforms.OneHotOp
|
||||
mindspore.dataset.transforms.py_transforms.RandomApply
|
||||
mindspore.dataset.transforms.py_transforms.RandomChoice
|
||||
mindspore.dataset.transforms.py_transforms.RandomOrder
|
||||
mindspore.dataset.transforms.Relational
|
||||
|
|
|
|||
|
|
@ -3,111 +3,89 @@ mindspore.dataset.vision
|
|||
|
||||
.. automodule:: mindspore.dataset.vision
|
||||
|
||||
mindspore.dataset.vision.c_transforms
|
||||
------------------------------------------------
|
||||
Transforms
|
||||
----------
|
||||
|
||||
.. autosummary::
|
||||
:toctree: dataset_vision
|
||||
:nosignatures:
|
||||
:template: classtemplate.rst
|
||||
|
||||
mindspore.dataset.vision.c_transforms.AutoContrast
|
||||
mindspore.dataset.vision.c_transforms.BoundingBoxAugment
|
||||
mindspore.dataset.vision.c_transforms.CenterCrop
|
||||
mindspore.dataset.vision.c_transforms.ConvertColor
|
||||
mindspore.dataset.vision.c_transforms.Crop
|
||||
mindspore.dataset.vision.c_transforms.CutMixBatch
|
||||
mindspore.dataset.vision.c_transforms.CutOut
|
||||
mindspore.dataset.vision.c_transforms.Decode
|
||||
mindspore.dataset.vision.c_transforms.Equalize
|
||||
mindspore.dataset.vision.c_transforms.GaussianBlur
|
||||
mindspore.dataset.vision.c_transforms.HorizontalFlip
|
||||
mindspore.dataset.vision.c_transforms.HWC2CHW
|
||||
mindspore.dataset.vision.c_transforms.Invert
|
||||
mindspore.dataset.vision.c_transforms.MixUpBatch
|
||||
mindspore.dataset.vision.c_transforms.Normalize
|
||||
mindspore.dataset.vision.c_transforms.NormalizePad
|
||||
mindspore.dataset.vision.c_transforms.Pad
|
||||
mindspore.dataset.vision.c_transforms.RandomAffine
|
||||
mindspore.dataset.vision.c_transforms.RandomColor
|
||||
mindspore.dataset.vision.c_transforms.RandomColorAdjust
|
||||
mindspore.dataset.vision.c_transforms.RandomCrop
|
||||
mindspore.dataset.vision.c_transforms.RandomCropDecodeResize
|
||||
mindspore.dataset.vision.c_transforms.RandomCropWithBBox
|
||||
mindspore.dataset.vision.c_transforms.RandomHorizontalFlip
|
||||
mindspore.dataset.vision.c_transforms.RandomHorizontalFlipWithBBox
|
||||
mindspore.dataset.vision.c_transforms.RandomPosterize
|
||||
mindspore.dataset.vision.c_transforms.RandomResize
|
||||
mindspore.dataset.vision.c_transforms.RandomResizedCrop
|
||||
mindspore.dataset.vision.c_transforms.RandomResizedCropWithBBox
|
||||
mindspore.dataset.vision.c_transforms.RandomResizeWithBBox
|
||||
mindspore.dataset.vision.c_transforms.RandomRotation
|
||||
mindspore.dataset.vision.c_transforms.RandomSelectSubpolicy
|
||||
mindspore.dataset.vision.c_transforms.RandomSharpness
|
||||
mindspore.dataset.vision.c_transforms.RandomSolarize
|
||||
mindspore.dataset.vision.c_transforms.RandomVerticalFlip
|
||||
mindspore.dataset.vision.c_transforms.RandomVerticalFlipWithBBox
|
||||
mindspore.dataset.vision.c_transforms.Rescale
|
||||
mindspore.dataset.vision.c_transforms.Resize
|
||||
mindspore.dataset.vision.c_transforms.ResizeWithBBox
|
||||
mindspore.dataset.vision.c_transforms.Rotate
|
||||
mindspore.dataset.vision.c_transforms.SlicePatches
|
||||
mindspore.dataset.vision.c_transforms.SoftDvppDecodeRandomCropResizeJpeg
|
||||
mindspore.dataset.vision.c_transforms.SoftDvppDecodeResizeJpeg
|
||||
mindspore.dataset.vision.c_transforms.UniformAugment
|
||||
mindspore.dataset.vision.c_transforms.VerticalFlip
|
||||
mindspore.dataset.vision.AdjustGamma
|
||||
mindspore.dataset.vision.AutoAugment
|
||||
mindspore.dataset.vision.AutoContrast
|
||||
mindspore.dataset.vision.BoundingBoxAugment
|
||||
mindspore.dataset.vision.CenterCrop
|
||||
mindspore.dataset.vision.ConvertColor
|
||||
mindspore.dataset.vision.Crop
|
||||
mindspore.dataset.vision.CutMixBatch
|
||||
mindspore.dataset.vision.CutOut
|
||||
mindspore.dataset.vision.Decode
|
||||
mindspore.dataset.vision.Equalize
|
||||
mindspore.dataset.vision.FiveCrop
|
||||
mindspore.dataset.vision.GaussianBlur
|
||||
mindspore.dataset.vision.GrayScale
|
||||
mindspore.dataset.vision.HorizontalFlip
|
||||
mindspore.dataset.vision.HsvToRgb
|
||||
mindspore.dataset.vision.HWC2CHW
|
||||
mindspore.dataset.vision.Invert
|
||||
mindspore.dataset.vision.LinearTransformation
|
||||
mindspore.dataset.vision.MixUp
|
||||
mindspore.dataset.vision.MixUpBatch
|
||||
mindspore.dataset.vision.Normalize
|
||||
mindspore.dataset.vision.NormalizePad
|
||||
mindspore.dataset.vision.Pad
|
||||
mindspore.dataset.vision.PadToSize
|
||||
mindspore.dataset.vision.RandomAdjustSharpness
|
||||
mindspore.dataset.vision.RandomAffine
|
||||
mindspore.dataset.vision.RandomAutoContrast
|
||||
mindspore.dataset.vision.RandomColor
|
||||
mindspore.dataset.vision.RandomColorAdjust
|
||||
mindspore.dataset.vision.RandomCrop
|
||||
mindspore.dataset.vision.RandomCropDecodeResize
|
||||
mindspore.dataset.vision.RandomCropWithBBox
|
||||
mindspore.dataset.vision.RandomEqualize
|
||||
mindspore.dataset.vision.RandomErasing
|
||||
mindspore.dataset.vision.RandomGrayscale
|
||||
mindspore.dataset.vision.RandomHorizontalFlip
|
||||
mindspore.dataset.vision.RandomHorizontalFlipWithBBox
|
||||
mindspore.dataset.vision.RandomInvert
|
||||
mindspore.dataset.vision.RandomLighting
|
||||
mindspore.dataset.vision.RandomPerspective
|
||||
mindspore.dataset.vision.RandomPosterize
|
||||
mindspore.dataset.vision.RandomResizedCrop
|
||||
mindspore.dataset.vision.RandomResizedCropWithBBox
|
||||
mindspore.dataset.vision.RandomResize
|
||||
mindspore.dataset.vision.RandomResizeWithBBox
|
||||
mindspore.dataset.vision.RandomRotation
|
||||
mindspore.dataset.vision.RandomSelectSubpolicy
|
||||
mindspore.dataset.vision.RandomSharpness
|
||||
mindspore.dataset.vision.RandomSolarize
|
||||
mindspore.dataset.vision.RandomVerticalFlip
|
||||
mindspore.dataset.vision.RandomVerticalFlipWithBBox
|
||||
mindspore.dataset.vision.Rescale
|
||||
mindspore.dataset.vision.Resize
|
||||
mindspore.dataset.vision.ResizeWithBBox
|
||||
mindspore.dataset.vision.RgbToHsv
|
||||
mindspore.dataset.vision.Rotate
|
||||
mindspore.dataset.vision.SlicePatches
|
||||
mindspore.dataset.vision.TenCrop
|
||||
mindspore.dataset.vision.ToNumpy
|
||||
mindspore.dataset.vision.ToPIL
|
||||
mindspore.dataset.vision.ToTensor
|
||||
mindspore.dataset.vision.ToType
|
||||
mindspore.dataset.vision.UniformAugment
|
||||
mindspore.dataset.vision.VerticalFlip
|
||||
|
||||
mindspore.dataset.vision.py_transforms
|
||||
-------------------------------------------------
|
||||
|
||||
.. autosummary::
|
||||
:toctree: dataset_vision
|
||||
:nosignatures:
|
||||
:template: classtemplate.rst
|
||||
|
||||
mindspore.dataset.vision.py_transforms.AutoContrast
|
||||
mindspore.dataset.vision.py_transforms.CenterCrop
|
||||
mindspore.dataset.vision.py_transforms.Cutout
|
||||
mindspore.dataset.vision.py_transforms.Decode
|
||||
mindspore.dataset.vision.py_transforms.Equalize
|
||||
mindspore.dataset.vision.py_transforms.FiveCrop
|
||||
mindspore.dataset.vision.py_transforms.Grayscale
|
||||
mindspore.dataset.vision.py_transforms.HsvToRgb
|
||||
mindspore.dataset.vision.py_transforms.HWC2CHW
|
||||
mindspore.dataset.vision.py_transforms.Invert
|
||||
mindspore.dataset.vision.py_transforms.LinearTransformation
|
||||
mindspore.dataset.vision.py_transforms.MixUp
|
||||
mindspore.dataset.vision.py_transforms.Normalize
|
||||
mindspore.dataset.vision.py_transforms.NormalizePad
|
||||
mindspore.dataset.vision.py_transforms.Pad
|
||||
mindspore.dataset.vision.py_transforms.RandomAffine
|
||||
mindspore.dataset.vision.py_transforms.RandomColor
|
||||
mindspore.dataset.vision.py_transforms.RandomColorAdjust
|
||||
mindspore.dataset.vision.py_transforms.RandomCrop
|
||||
mindspore.dataset.vision.py_transforms.RandomErasing
|
||||
mindspore.dataset.vision.py_transforms.RandomGrayscale
|
||||
mindspore.dataset.vision.py_transforms.RandomHorizontalFlip
|
||||
mindspore.dataset.vision.py_transforms.RandomPerspective
|
||||
mindspore.dataset.vision.py_transforms.RandomResizedCrop
|
||||
mindspore.dataset.vision.py_transforms.RandomRotation
|
||||
mindspore.dataset.vision.py_transforms.RandomSharpness
|
||||
mindspore.dataset.vision.py_transforms.RandomVerticalFlip
|
||||
mindspore.dataset.vision.py_transforms.Resize
|
||||
mindspore.dataset.vision.py_transforms.RgbToHsv
|
||||
mindspore.dataset.vision.py_transforms.TenCrop
|
||||
mindspore.dataset.vision.py_transforms.ToPIL
|
||||
mindspore.dataset.vision.py_transforms.ToTensor
|
||||
mindspore.dataset.vision.py_transforms.ToType
|
||||
mindspore.dataset.vision.py_transforms.UniformAugment
|
||||
|
||||
mindspore.dataset.vision.utils
|
||||
-------------------------------
|
||||
Others
|
||||
------
|
||||
|
||||
.. autosummary::
|
||||
:toctree: dataset_vision
|
||||
:nosignatures:
|
||||
:template: classtemplate.rst
|
||||
|
||||
mindspore.dataset.vision.AutoAugmentPolicy
|
||||
mindspore.dataset.vision.Border
|
||||
mindspore.dataset.vision.ConvertMode
|
||||
mindspore.dataset.vision.ImageBatchFormat
|
||||
|
|
|
|||
|
|
@ -21,7 +21,7 @@ Common imported modules in corresponding API examples are as follows:
|
|||
.. code-block::
|
||||
|
||||
import mindspore.dataset as ds
|
||||
import mindspore.dataset.transforms
|
||||
import mindspore.dataset.transforms as transforms
|
||||
|
||||
Alternative and equivalent imported transforms is as follows:
|
||||
|
||||
|
|
|
|||
|
|
@ -523,9 +523,11 @@ class CutOut(TensorOperation):
|
|||
|
||||
Raises:
|
||||
TypeError: If `length` is not of type integer.
|
||||
TypeError: If `is_hwc` is not of type bool.
|
||||
TypeError: If `num_patches` is not of type integer.
|
||||
ValueError: If `length` is less than or equal 0.
|
||||
ValueError: If `num_patches` is less than or equal 0.
|
||||
RuntimeError: If given tensor shape is not <H, W, C>.
|
||||
|
||||
Supported Platforms:
|
||||
``CPU``
|
||||
|
|
@ -2847,7 +2849,7 @@ class RandomVerticalFlip(TensorOperation, PyTensorOperation):
|
|||
Randomly flip the input image vertically with a given probability.
|
||||
|
||||
Args:
|
||||
prob (float, optional): Probability of the image being flipped (default=0.5).
|
||||
prob (float, optional): Probability of the image being flipped. Default=0.5.
|
||||
|
||||
Raises:
|
||||
TypeError: If `prob` is not of type float.
|
||||
|
|
@ -2920,6 +2922,9 @@ class Rescale(TensorOperation):
|
|||
Rescale the input image with the given rescale and shift. This operator will rescale the input image
|
||||
with: output = image * rescale + shift.
|
||||
|
||||
Note:
|
||||
This operation supports running on Ascend or GPU platforms by Offload.
|
||||
|
||||
Args:
|
||||
rescale (float): Rescale factor.
|
||||
shift (float): Shift factor.
|
||||
|
|
@ -2929,7 +2934,7 @@ class Rescale(TensorOperation):
|
|||
TypeError: If `shift` is not of type float.
|
||||
|
||||
Supported Platforms:
|
||||
``CPU``
|
||||
``CPU`` ``Ascend`` ``GPU``
|
||||
|
||||
Examples:
|
||||
>>> transforms_list = [vision.Decode(), vision.Rescale(1.0 / 255.0, -1.0)]
|
||||
|
|
@ -2961,20 +2966,14 @@ class Resize(TensorOperation, PyTensorOperation):
|
|||
It can be any of [Inter.BILINEAR, Inter.LINEAR, Inter.NEAREST, Inter.BICUBIC, Inter.AREA, Inter.PILCUBIC,
|
||||
Inter.ANTIALIAS].
|
||||
|
||||
- Inter.BILINEAR, means interpolation method is bilinear interpolation.
|
||||
|
||||
- Inter.LINEAR, means interpolation method is bilinear interpolation, here is the same as Inter.BILINEAR.
|
||||
|
||||
- Inter.NEAREST, means interpolation method is nearest-neighbor interpolation.
|
||||
|
||||
- Inter.BICUBIC, means interpolation method is bicubic interpolation.
|
||||
|
||||
- Inter.AREA, means interpolation method is pixel area interpolation.
|
||||
|
||||
- Inter.PILCUBIC, means interpolation method is bicubic interpolation like implemented in Pillow, input
|
||||
should be in 3 channels format.
|
||||
|
||||
- Inter.ANTIALIAS, means antialias interpolation.
|
||||
- Inter.BILINEAR, bilinear interpolation.
|
||||
- Inter.LINEAR, bilinear interpolation, here is the same as Inter.BILINEAR.
|
||||
- Inter.NEAREST, nearest-neighbor interpolation.
|
||||
- Inter.BICUBIC, bicubic interpolation.
|
||||
- Inter.AREA, pixel area interpolation.
|
||||
- Inter.PILCUBIC, bicubic interpolation like implemented in Pillow, only valid when the input is
|
||||
a 3-channel image in the numpy.ndarray format.
|
||||
- Inter.ANTIALIAS, antialias interpolation.
|
||||
|
||||
Raises:
|
||||
TypeError: If `size` is not of type int or Sequence[int].
|
||||
|
|
@ -3086,6 +3085,9 @@ class RgbToHsv(PyTensorOperation):
|
|||
is_hwc (bool): If True, means the input image is in shape of (H, W, C) or (N, H, W, C).
|
||||
Otherwise, it is in shape of (C, H, W) or (N, C, H, W). Default: False.
|
||||
|
||||
Raises:
|
||||
TypeError: If `is_hwc` is not of type bool.
|
||||
|
||||
Supported Platforms:
|
||||
``CPU``
|
||||
|
||||
|
|
@ -3303,7 +3305,7 @@ class TenCrop(PyTensorOperation):
|
|||
|
||||
class ToNumpy(PyTensorOperation):
|
||||
"""
|
||||
Convert the PIL input image to NumPy array.
|
||||
Convert the PIL input image to numpy.ndarray image.
|
||||
|
||||
Supported Platforms:
|
||||
``CPU``
|
||||
|
|
@ -3385,11 +3387,11 @@ class ToPIL(PyTensorOperation):
|
|||
|
||||
class ToTensor(TensorOperation, PyTensorOperation):
|
||||
"""
|
||||
Rescale of pixel value range from [0, 255] to [0.0, 1.0] and change the shape from (H, W, C) to (C, H, W).
|
||||
Also convert the input PIL Image or numpy.ndarray to numpy.ndarray of the desired dtype.
|
||||
Convert the input PIL Image or numpy.ndarray to numpy.ndarray of the desired dtype, rescale the pixel value
|
||||
range from [0, 255] to [0.0, 1.0] and change the shape from (H, W, C) to (C, H, W).
|
||||
|
||||
Args:
|
||||
output_type (mindspore.dtype or numpy.dtype, optional): The desired dtype of the output image.
|
||||
output_type (Union[mindspore.dtype, numpy.dtype], optional): The desired dtype of the output image.
|
||||
Default: :class:`numpy.float32`.
|
||||
|
||||
Raises:
|
||||
|
|
@ -3427,18 +3429,19 @@ class ToTensor(TensorOperation, PyTensorOperation):
|
|||
|
||||
class ToType(TypeCast):
|
||||
"""
|
||||
Tensor operation to cast to a given MindSpore data type or NumPy data type.
|
||||
Note: This operation is an alias for TypeCast operation.
|
||||
Cast the input to a given MindSpore data type or NumPy data type.
|
||||
|
||||
It is the same as that of :class:`mindspore.dataset.transforms.TypeCast`.
|
||||
|
||||
Note:
|
||||
This operation supports running on Ascend or GPU platforms by Offload.
|
||||
|
||||
Args:
|
||||
data_type (mindspore.dtype or numpy.dtype): mindspore.dtype or numpy.dtype (e.g. :class:`numpy.float32`)
|
||||
to be cast to.
|
||||
data_type (Union[mindspore.dtype, numpy.dtype]): The desired data type of the output image,
|
||||
such as :class:`numpy.float32`.
|
||||
|
||||
Raises:
|
||||
TypeError: If `data_type` is not of MindSpore data type bool, int, float, string or type :class:`numpy.ndarray`.
|
||||
TypeError: If `data_type` is not of type :class:`mindspore.dtype` or :class:`numpy.dtype`.
|
||||
|
||||
Supported Platforms:
|
||||
``CPU`` ``Ascend`` ``GPU``
|
||||
|
|
|
|||
Loading…
Reference in New Issue