modify format1116

This commit is contained in:
huodagu 2022-11-16 16:59:59 +08:00
parent 7190fff7cb
commit 8035203d14
22 changed files with 47 additions and 36 deletions

View File

@ -14,4 +14,5 @@ mindspore.dataset.audio.BorderType
- **BorderType.SYMMETRIC** - 以各边的边界为轴进行对称填充,包括边界像素值。
例如,向 [1, 2, 3, 4] 的两边分别填充2个元素结果为 [2, 1, 1, 2, 3, 4, 4, 3]。
.. note:: 该类派生自 :class:`str` 以支持 JSON 可序列化。
.. note::
该类派生自 `str` 以支持 JSON 可序列化。

View File

@ -8,7 +8,7 @@ mindspore.dataset.transforms.Mask
参数:
- **operator** (:class:`mindspore.dataset.transforms.Relational`) - 关系操作符可以取值为Relational.EQ、Relational.NE、Relational.LT、Relational.GT、Relational.LE、Relational.GE。以Relational.EQ为例将找出Tensor中与 `constant` 相等的元素。
- **constant** (Union[str, int, float, bool]) - 与输入Tensor进行比较的基准值。
- **dtype** (:class:`mindspore.dtype`, 可选) - 生成的掩码Tensor的数据类型。默认值:class:`mindspore.dtype.bool_`
- **dtype** (:class:`mindspore.dtype`, 可选) - 生成的掩码Tensor的数据类型。默认值mstype.bool\_ 。
异常:
- **TypeError** - 参数 `operator` 类型不为 :class:`mindspore.dataset.transforms.Relational`

View File

@ -30,6 +30,6 @@ mindspore.dataset.vision.AutoAugment
异常:
- **TypeError** - 如果 `policy` 不是 :class:`mindspore.dataset.vision.AutoAugmentPolicy` 类型。
- **TypeError** - 如果 `interpolation` 不是 :class:`mindsore.dataset.vision.Inter` 类型。
- **TypeError** - 如果 `interpolation` 不是 :class:`mindspore.dataset.vision.Inter` 类型。
- **TypeError** - 如果 `fill_value` 不是整数或长度为3的元组。
- **RuntimeError** - 如果给定的张量形状不是<H, W, C>。
- **RuntimeError** - 如果给定的张量shape不是<H, W, C>。

View File

@ -14,4 +14,5 @@ mindspore.dataset.vision.Border
- **Border.SYMMETRIC** - 以各边的边界为轴进行对称填充,包括边界像素值。
例如,对 [1,2,3,4] 的两侧分别填充2个元素结果为 [2,1,1,2,3,4,4,3]。
.. note:: 该类派生自 :class:`str` 以支持 JSON 可序列化。
.. note::
该类派生自 `str` 以支持 JSON 可序列化。

View File

@ -1,6 +1,6 @@
mindspore.Tensor.index_add
==========================
.. py:method:: mindspore.Tensor.index_addindex_add(dim, index, source, *, alpha=1)
.. py:method:: mindspore.Tensor.index_add(dim, index, source, *, alpha=1)
详情请参考 :func:`mindspore.ops.index_add`

View File

@ -11,7 +11,7 @@ mindspore.DatasetHelper
DatasetHelper的迭代将提供一个epoch的数据。
参数:
- **dataset** (Dataset) - 训练数据集迭代器。数据集可以由数据集生成器API在 :class:`mindspore.dataset` 中生成,例如 :class:`mindspore.dataset.ImageFolderDataset`
- **dataset** (Dataset) - 训练数据集迭代器。数据集可以由数据集生成器API在 `mindspore.dataset` 中生成,例如 :class:`mindspore.dataset.ImageFolderDataset`
- **dataset_sink_mode** (bool) - 如果值为True使用 :class:`mindspore.ops.GetNext` 在设备Device上通过数据通道中获取数据否则在主机Host直接遍历数据集获取数据。默认值True。
- **sink_size** (int) - 控制每个下沉中的数据量。如果 `sink_size` 为-1则下沉每个epoch的完整数据集。如果 `sink_size` 大于0则下沉每个epoch的 `sink_size` 数据。默认值:-1。
- **epoch_num** (int) - 控制待发送的epoch数据量。默认值1。

View File

@ -7,7 +7,7 @@ mindspore.data_sink
参数:
- **fn** (Function) - 将与数据集一起运行的函数。
- **dataset** (Dataset) - 训练数据集迭代器。数据集可以由数据集生成器API在 :class:`mindspore.dataset` 中生成,例如 :class:`mindspore.dataset.ImageFolderDataset`
- **dataset** (Dataset) - 训练数据集迭代器。数据集可以由数据集生成器API在 `mindspore.dataset` 中生成,例如 :class:`mindspore.dataset.ImageFolderDataset`
- **sink_size** (int) - 控制每次下沉的数据执行次数。 `sink_size` 必须为正整数。默认值1。
- **jit_config** (JitConfig) - 编译时所使用的JitConfig配置项详细可参考 :class:`mindspore.JitConfig` 。默认值None表示以PyNative模式运行。
- **input_signature** (Union[Tensor, List or Tuple of Tensors]) - 用于表示输入参数的Tensor。Tensor的shape和dtype将作为函数的输入shape和dtype。默认值None。

View File

@ -19,7 +19,7 @@ mindspore.load_checkpoint
- **specify_prefix** (Union[str, list[str], tuple[str]]) - 以 `specify_prefix` 开头的参数将会被加载。默认值None。
返回:
字典key是参数名称value是Parameter类型。当使用 :func:`mindspore.save_checkpoint``append_dict` 参数和 :class:`CheckpointConfig` 的 `append_info` 参数保存
字典key是参数名称value是Parameter类型。当使用 :func:`mindspore.save_checkpoint``append_dict` 参数和 :class:`mindspore.train.CheckpointConfig` 的 `append_info` 参数保存
checkpoint `append_dict``append_info` 是dict类型且它们的值value是string时加载checkpoint得到的返回值是string类型其它情况返回值均是Parameter类型。
异常:

View File

@ -8,7 +8,7 @@ mindspore.ops.Conv2DTranspose
参数:
- **out_channel** (int) - 输出的通道数。
- **kernel_size** (Union[int, tuple[int]]) - 卷积核的大小。
- **pad_mode** (str) - 填充的模式。它可以是"valid"、"same"或"pad"。默认值:"valid"。请参考 :class:`mindspore.nn.Conv2DTranspose` 了解更多 `pad_mode` 的使用规则。
- **pad_mode** (str) - 填充的模式。它可以是"valid"、"same"或"pad"。默认值:"valid"。请参考 :class:`mindspore.nn.Conv2dTranspose` 了解更多 `pad_mode` 的使用规则。
- **pad** (Union[int, tuple[int]]) - 指定要填充的填充值。默认值0。如果 `pad` 是整数,则顶部、底部、左侧和右侧的填充都等于 `pad` 。如果 `pad` 是四个整数的tuple则顶部、底部、左侧和右侧的填充分别等于pad[0]、pad[1]、pad[2]和pad[3]。
- **pad_list** (Union[str, None]) - 卷积填充方式顶部、底部、左、右。默认值None表示不使用此参数。
- **mode** (int) - 指定不同的卷积模式。当前未使用该值。默认值1。

View File

@ -1,7 +1,7 @@
mindspore.ops.addr
==================
.. py:function:: mindspore.ops.addr(x, vec1, vec2, beta=1, alpha=1)
.. py:function:: mindspore.ops.addr(vec1, vec2, beta=1, alpha=1)
计算 `vec1``vec2` 的外积,并将其添加到 `x` 中。

View File

@ -3,8 +3,10 @@ mindspore.ops.multi_label_margin_loss
.. py:function:: mindspore.ops.multi_label_margin_loss(inputs, target, reduction='mean')
用于优化多标签分类问题的铰链损失。
创建一个标准,用于优化输入 :math:`x` 一个2D小批量Tensor
和输出 :math:`y` 一个目标类别索引的2DTensor之间的多类分类铰链损失基于边距的损失
和输出 :math:`y` 一个目标类别索引的2DTensor之间的多标签分类铰链损失(基于边距的损失):
对于每个小批量样本:
.. math::

View File

@ -3,6 +3,8 @@ mindspore.ops.multi_margin_loss
.. py:function:: mindspore.ops.multi_margin_loss(inputs, target, p=1, margin=1, weight=None, reduction='mean')
用于优化多类分类问题的铰链损失。
创建一个标准,用于优化输入 :math:`x` 一个2D小批量Tensor
和输出 :math:`y` 一个目标类索引的1DTensor :math:`0 \leq y \leq \text{x.size}(1)-1`
之间的多类分类铰链损失(基于边距的损失):

View File

@ -2750,7 +2750,7 @@ class Tensor(Tensor_):
Args:
v (Union[int, float, bool, list, tuple, Tensor]): Values to insert into the tensor.
side ('left', 'right', optional): If 'left', the index of the first suitable
side (str, optional): If 'left', the index of the first suitable
location found is given. If 'right', return the last such index. If there is
no suitable index, return either 0 or N (where N is the length of the tensor).
Default: 'left'.

View File

@ -286,7 +286,7 @@ class Mask(TensorOperation):
operator (Relational): relational operators, it can be any of [Relational.EQ, Relational.NE, Relational.LT,
Relational.GT, Relational.LE, Relational.GE], take Relational.EQ as example, EQ refers to equal.
constant (Union[str, int, float, bool]): Constant to be compared to.
dtype (mindspore.dtype, optional): Type of the generated mask. Default: mindspore.dtype.bool\_.
dtype (mindspore.dtype, optional): Type of the generated mask. Default: mstype.bool_.
Raises:
TypeError: `operator` is not of type Relational.

View File

@ -701,7 +701,7 @@ class ConvertColor(ImageTensorOperation):
- ConvertMode.COLOR_RGBA2GRAY, Convert RGBA image to GRAY image.
Raises:
TypeError: If `convert_mode` is not of type :class:`mindspore.dataset.vision.transforms.ConvertMode` .
TypeError: If `convert_mode` is not of type :class:`mindspore.dataset.vision.ConvertMode` .
RuntimeError: If given tensor shape is not <H, W> or <H, W, C>.
Supported Platforms:

View File

@ -180,7 +180,7 @@ class EmbeddingLookup(Cell):
target (str): Specifies the target where the op is executed. The value must in
['DEVICE', 'CPU']. Default: 'CPU'.
slice_mode (str): The slicing way in semi_auto_parallel/auto_parallel. The value must get through
:class:`mindspore.nn.EmbeddingLookup`. Default: :class:`mindspore.nn.EmbeddingLookup.BATCH_SLICE`.
:class:`mindspore.nn.EmbeddingLookup`. Default: 'nn.EmbeddingLookup.BATCH_SLICE'.
manual_shapes (tuple): The accompaniment array in field slice mode. Default: None.
max_norm (Union[float, None]): A maximum clipping value. The data type must be float16, float32
or None. Default: None
@ -496,7 +496,7 @@ class MultiFieldEmbeddingLookup(EmbeddingLookup):
target (str): Specifies the target where the op is executed. The value must in
['DEVICE', 'CPU']. Default: 'CPU'.
slice_mode (str): The slicing way in semi_auto_parallel/auto_parallel. The value must get through
:class:`mindspore.nn.EmbeddingLookup`. Default: :class:`mindspore.nn.EmbeddingLookup.BATCH_SLICE`.
:class:`mindspore.nn.EmbeddingLookup`. Default: 'nn.EmbeddingLookup.BATCH_SLICE'.
feature_num_list (tuple): The accompaniment array in field slice mode. This is unused currently. Default: None.
max_norm (Union[float, None]): A maximum clipping value. The data type must be float16, float32
or None. Default: None

View File

@ -3988,7 +3988,7 @@ def addr(x, vec1, vec2, beta=1, alpha=1):
alpha (scalar[int, float, bool], optional): Multiplier for `vec1` `vec2` (α). The `alpha` must
be int or float or bool, Default: 1.
Outputs:
Returns:
Tensor, the shape of the output tensor is :vec1:`(N, M)`, has the same dtype as `x`.
Raises:

View File

@ -4501,6 +4501,8 @@ def glu(x, axis=-1):
def multi_margin_loss(inputs, target, p=1, margin=1, weight=None, reduction='mean'):
r"""
Hinge loss for optimizing a multi-class classification.
Creates a criterion that optimizes a multi-class classification hinge
loss (margin-based loss) between input :math:`x` (a 2D mini-batch `Tensor`) and
output :math:`y` (which is a 1D tensor of target class indices,
@ -4569,7 +4571,9 @@ def multi_margin_loss(inputs, target, p=1, margin=1, weight=None, reduction='mea
def multi_label_margin_loss(inputs, target, reduction='mean'):
r"""
Creates a criterion that optimizes a multi-class multi-classification
Hinge loss for optimizing a multi-label classification.
Creates a criterion that optimizes a multi-label multi-classification
hinge loss (margin-based loss) between input :math:`x` (a 2D mini-batch `Tensor`)
and output :math:`y` (which is a 2D `Tensor` of target class indices).
For each sample in the mini-batch:

View File

@ -32,12 +32,12 @@ class AdjustSaturation(Primitive):
adds an offset to the saturation channel, converts back to RGB and then back to the original data type.
If several adjustments are chained it is advisable to minimize the number of redundant conversions.
inputs:
Inputs:
- **image** (Tensor): Images to adjust. Must be one of the following types: float16, float32.
At least 3-D.The last dimension is interpreted as channels, and must be three.
- **scale** (Tensor): A float scale to add to the saturation. A Tensor of type float32. Must be 0-D.
Output:
Outputs:
Adjusted image(s), same shape and dtype as `image`.
Raises:
@ -81,12 +81,12 @@ class AdjustContrastv2(Primitive):
The other dimensions only represent a collection of images, such as [batch, height, width, channels].
Contrast is adjusted independently for each channel of each image.
inputs:
Inputs:
-**images**(tensor): Images to adjust. Must be one of the following types: float16, float32.
At least 3-D.The last dimension is interpreted as channels, and must be three.
-**contrast_factor**(tensor): A float multiplier for adjusting contrast. A Tensor of type float32. Must be 0-D.
Output:
Outputs:
Adjusted image(s), same shape and dtype as `images`.
Raises:
@ -135,7 +135,7 @@ class AdjustHue(Primitive):
the dtype is float16 or float32. At least 3-D.
- **delta** (Tensor): How much to add to the hue channel, the dtype is float32. Must be 0-D.
Output:
Outputs:
Adjusted image(s), same shape and dtype as `image`.
Raises:

View File

@ -7530,7 +7530,7 @@ class Qr(Primitive):
class Cauchy(Primitive):
r"""
Create a tensor of shape `size` with random numbers drawn from Cauchy distribution
Create a tensor of shape `size` with random numbers drawn from Cauchy distribution.
.. math::
\f(x)= \frac{1}{\pi} \frac{\sigma}{(x-median)^2 +\sigma^2}

View File

@ -7985,7 +7985,7 @@ class CTCLossV2(Primitive):
Args:
blank (int, optional): The blank label. Default: 0.
reduction (string, optional): Apply specific reduction method to the output. Currently only support 'none',
reduction (str, optional): Apply specific reduction method to the output. Currently only support 'none',
not case sensitive. Default: "none".
zero_infinity (bool, optional): Whether to set infinite loss and correlation gradient to zero. Default: False.

View File

@ -816,15 +816,16 @@ class BlackmanWindow(Primitive):
Args:
periodic (bool): If True, returns a window to be used as periodic function.
If False, return a symmetric window. Default: True.
dtype (mindspore.dtype): the desired data type of returned tensor. Only float16, float32 and float64 is allowed.
Default: mindspore.float32.
dtype (mindspore.dtype): the desired data type of returned tensor.
Only float16, float32 and float64 is allowed. Default: mindspore.float32.
Inputs:
- **window_length** (Tensor) - the size of returned window, with data type int32, int64.
The input data should be an integer with a value of [0, 1000000].
Outputs:
A 1-D tensor of size "window_length" containing the window. Its datatype is set by the attr 'dtype'
A 1-D tensor of size "window_length" containing the window. Its datatype is set by the attr 'dtype'.
Raises:
TypeError: If "window_length" is not a Tensor.
TypeError: If "periodic" is not a bool.