!68971 [Document Consistency] modify format

Merge pull request !68971 from 俞涵/code_docs_mst
2024-04-30 08:23:39 +00:00 · 2024-04-30 08:23:39 +00:00 · ece62d03e3
parent 109ab098cf f9d8b0fc78
commit ece62d03e3
13 changed files with 31 additions and 32 deletions
--- a/docs/api/api_python/mindspore/Tensor/mindspore.Tensor.register_hook.rst
+++ b/docs/api/api_python/mindspore/Tensor/mindspore.Tensor.register_hook.rst
@ -16,5 +16,5 @@ mindspore.Tensor.register_hook
    返回：
        返回与该hook_fn函数对应的handle对象。可通过调用handle.remove()来删除添加的hook_fn函数。

-     异常：
-         - **TypeError** - 如果 `hook_fn` 不是Python函数。
+    异常：
+        - **TypeError** - 如果 `hook_fn` 不是Python函数。
--- a/docs/api/api_python/mindspore/mindspore.recompute.rst
+++ b/docs/api/api_python/mindspore/mindspore.recompute.rst
@ -3,8 +3,7 @@ mindspore.recompute

 .. py:function:: mindspore.recompute(block, *args, **kwargs)

-    该函数用于减少显存的使用，当运行选定的模块时，不再保存其中的前向计算的产生的激活值，我们将在反向传播时，
-    重新计算前向的激活值。
+    该函数用于减少显存的使用，当运行选定的模块时，不再保存其中的前向计算的产生的激活值，我们将在反向传播时，重新计算前向的激活值。

    .. note::
        - 重计算函数只支持继承自Cell对象的模块，
--- a/docs/api/api_python/nn/mindspore.nn.Cell.rst
+++ b/docs/api/api_python/nn/mindspore.nn.Cell.rst
@ -279,11 +279,6 @@

        `parameter_layout_dict` 表示一个参数的张量layout，这种张量layout是由分片策略和分布式算子信息推断出来的。

-    .. py:method:: pipeline_stage
-        :property:
-
-        `pipeline_stage` 表示当前Cell所在的stage。
-
    .. py:method:: parameters_and_names(name_prefix='', expand=True)

        返回Cell中parameter的迭代器。
@ -320,6 +315,11 @@
        返回：
            OrderedDict类型，返回参数字典。

+    .. py:method:: pipeline_stage
+        :property:
+
+        `pipeline_stage` 表示当前Cell所在的stage。
+
    .. py:method:: place(role, rank_id)

        为该Cell中所有算子设置标签。此标签告诉MindSpore编译器此Cell在哪个进程上启动。
--- a/docs/api/api_python/nn/mindspore.nn.optim_arg_dynamic_lr.rst
+++ b/docs/api/api_python/nn/mindspore.nn.optim_arg_dynamic_lr.rst
@ -1,5 +1,5 @@
- **float** - <EFBFBD>̶<EFBFBD><EFBFBD><EFBFBD>ѧϰ<EFBFBD>ʡ<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>ڵ<EFBFBD><EFBFBD><EFBFBD><EFBFBD>㡣
- **int** - <EFBFBD>̶<EFBFBD><EFBFBD><EFBFBD>ѧϰ<EFBFBD>ʡ<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>ڵ<EFBFBD><EFBFBD><EFBFBD><EFBFBD>㡣<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>ͻᱻת<EFBFBD><EFBFBD>Ϊ<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
- **Tensor** - <EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>Ǳ<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>һά<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>ǹ̶<EFBFBD><EFBFBD><EFBFBD>ѧϰ<EFBFBD>ʡ<EFBFBD>һά<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>Ƕ<EFBFBD>̬<EFBFBD><EFBFBD>ѧϰ<EFBFBD>ʣ<EFBFBD><EFBFBD><EFBFBD>i<EFBFBD><EFBFBD><EFBFBD><EFBFBD>ȡ<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>е<EFBFBD>i<EFBFBD><EFBFBD>ֵ<EFBFBD><EFBFBD>Ϊѧϰ<EFBFBD>ʡ<EFBFBD>
- **Iterable** - <EFBFBD><EFBFBD>̬<EFBFBD><EFBFBD>ѧϰ<EFBFBD>ʡ<EFBFBD><EFBFBD><EFBFBD>i<EFBFBD><EFBFBD><EFBFBD><EFBFBD>ȡ<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>i<EFBFBD><EFBFBD>ֵ<EFBFBD><EFBFBD>Ϊѧϰ<EFBFBD>ʡ<EFBFBD>
- **LearningRateSchedule** - <EFBFBD><EFBFBD>̬<EFBFBD><EFBFBD>ѧϰ<EFBFBD>ʡ<EFBFBD><EFBFBD><EFBFBD>ѵ<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>У<EFBFBD><EFBFBD>Ż<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>ʹ<EFBFBD>ò<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>mindspore.cn/docs/zh-CN/masterateSchedule <https://www.mindspore.cn/docs/zh-CN/r2.3/api_python/mindspore.nn.html#learningrateschedule%E7%B1%BB>`_ ʵ<><CAB5><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>㵱ǰѧϰ<D1A7>ʡ<EFBFBD>
+- **float** - 固定的学习率。必须大于等于零。
+- **int** - 固定的学习率。必须大于等于零。整数类型会被转换为浮点数。
+- **Tensor** - 可以是标量或一维向量。标量是固定的学习率。一维向量是动态的学习率，第i步将取向量中第i个值作为学习率。
+- **Iterable** - 动态的学习率。第i步将取迭代器第i个值作为学习率。
+- **LearningRateSchedule** - 动态的学习率。在训练过程中，优化器将使用步数（step）作为输入，调用 `LearningRateSchedule <https://www.mindspore.cn/docs/zh-CN/master/api_python/mindspore.nn.html#learningrateschedule%E7%B1%BB>`_ 实例来计算当前学习率。
--- a/docs/api/api_python/ops/mindspore.ops.func_layer_norm.rst
+++ b/docs/api/api_python/ops/mindspore.ops.func_layer_norm.rst
@ -22,7 +22,7 @@ mindspore.ops.layer_norm
        - **bias** (Tensor, 可选) - 可学习的偏移值，shape为 `normalized_shape` ，默认值: ``None`` 。为 ``None`` 时，初始化为 ``0`` 。
        - **eps** (float, 可选) - 添加到分母中的值（:math:`\epsilon`），以确保数值稳定。默认值： ``1e-5`` 。

-    输出：
+    返回：
        Tensor，归一化后的Tensor，shape和数据类型与 `input` 相同。

    异常：
--- a/mindspore/core/ops/ops_def/doc/batch_matmul_doc.yaml
+++ b/mindspore/core/ops/ops_def/doc/batch_matmul_doc.yaml
@ -16,10 +16,10 @@ batch_mat_mul:

        Inputs:
            - **x** (Tensor) - The first tensor to be multiplied. The shape of the tensor is :math:`(*B, N, C)`,
-                  where :math:`*B` represents the batch size which can be multidimensional, :math:`N` and :math:`C` are the
-                  size of the last two dimensions. If `transpose_a` is ``True`` , its shape must be :math:`(*B, C, N)`.
+              where :math:`*B` represents the batch size which can be multidimensional, :math:`N` and :math:`C` are the
+              size of the last two dimensions. If `transpose_a` is ``True`` , its shape must be :math:`(*B, C, N)`.
            - **y** (Tensor) - The second tensor to be multiplied. The shape of the tensor is :math:`(*B, C, M)`. If
-                  `transpose_b` is ``True`` , its shape must be :math:`(*B, M, C)`.
+              `transpose_b` is ``True`` , its shape must be :math:`(*B, M, C)`.

        Outputs:
            Tensor, the shape of the output tensor is :math:`(*B, N, M)`.
--- a/mindspore/core/ops/ops_def/doc/matmul_doc.yaml
+++ b/mindspore/core/ops/ops_def/doc/matmul_doc.yaml
@ -19,9 +19,9 @@ matmul:

    Inputs:
        - **a** (Tensor) - The first tensor to be multiplied. The shape of the tensor is :math:`(N, C)`. If
-              `transpose_a` is ``True`` , its shape must be :math:`(C, N)` after transpose.
+          `transpose_a` is ``True`` , its shape must be :math:`(C, N)` after transpose.
        - **b** (Tensor) - The second tensor to be multiplied. The shape of the tensor is :math:`(C, M)`. If
-              `transpose_b` is ``True`` , its shape must be :math:`(M, C)` after transpose.
+          `transpose_b` is ``True`` , its shape must be :math:`(M, C)` after transpose.

    Outputs:
        Tensor, the shape of the output tensor is :math:`(N, M)`.
--- a/mindspore/core/ops/ops_def/doc/ones_doc.yaml
+++ b/mindspore/core/ops/ops_def/doc/ones_doc.yaml
@ -9,7 +9,7 @@ ones:
            For argument `shape`, Tensor type input will be deprecated in the future version.

        Args:
-            shape (Union[tuple[int], List[int], int, Tensor]): The specified shape of output tensor. Only positive integer or
+            shape (Union[tuple[int], list[int], int, Tensor]): The specified shape of output tensor. Only positive integer or
                tuple or Tensor containing positive integers are allowed. If it is a Tensor,
                it must be a 0-D or 1-D Tensor with int32 or int64 dtypes.
            dtype (:class:`mindspore.dtype`): The specified type of output tensor. If `dtype` is ``None`` ,
--- a/mindspore/core/ops/ops_def/doc/zeros_doc.yaml
+++ b/mindspore/core/ops/ops_def/doc/zeros_doc.yaml
@ -6,7 +6,7 @@ zeros:
            For argument `size`, Tensor type input will be deprecated in the future version.

        Args:
-            size (Union[tuple[int], List[int], int, Tensor]): The specified shape of output tensor. Only positive integer or
+            size (Union[tuple[int], list[int], int, Tensor]): The specified shape of output tensor. Only positive integer or
                tuple or Tensor containing positive integers are allowed. If it is a Tensor,
                it must be a 0-D or 1-D Tensor with int32 or int64 dtypes.
            dtype (:class:`mindspore.dtype`, optional): The specified type of output tensor. If `dtype` is ``None`` ,
--- a/mindspore/python/mindspore/common/recompute.py
+++ b/mindspore/python/mindspore/common/recompute.py
@ -38,17 +38,18 @@ def recompute(block, *args, **kwargs):
    storing the intermediate activation computed in forward pass, we will recompute it in backward pass.

    Note:
-     - Recompute function only support block which inherited from Cell object.
-     - This function interface now only support pynative mode. you can use Cell.recompute interface
-       in graph mode.
-     - When use recompute function, block object should not decorated by @jit.
+        - Recompute function only support block which inherited from Cell object.
+        - This function interface now only support pynative mode. you can use Cell.recompute interface
+          in graph mode.
+        - When use recompute function, block object should not decorated by @jit.

    Args:
        block (Cell): Block to be recompute.
        args(tuple): Inputs for block object to run forward pass.
        kwargs(dict): Optional input for recompute function.

-    Returns: Same as return type of block.
+    Returns:
+        Same as return type of block.

    Raises:
        TypeError: If `block` is not Cell object.
--- a/mindspore/python/mindspore/common/tensor.py
+++ b/mindspore/python/mindspore/common/tensor.py
@ -4445,7 +4445,7 @@ class Tensor(Tensor_, metaclass=_TensorMeta):
            For argument `size`, Tensor type input will be deprecated in the future version.

        Args:
-            size (Union[int, tuple, list]): An int, list or tuple of integers defining the output shape.
+            size (Union[int, tuple, list, Tensor]): An int, list or tuple of integers defining the output shape.
            dtype (mindspore.dtype, optional): The desired dtype of the output tensor. If None, the returned tensor has
                thesame dtype as `self`. Default: ``None``.

@ -4478,7 +4478,7 @@ class Tensor(Tensor_, metaclass=_TensorMeta):
            For argument `size`, Tensor type input will be deprecated in the future version.

        Args:
-            size (Union[int, tuple, list]): An int, list or tuple of integers defining the output shape.
+            size (Union[int, tuple, list, Tensor]): An int, list or tuple of integers defining the output shape.
            dtype (mindspore.dtype, optional): The desired dtype of the output tensor. If None, the returned
                tensor has the same dtype as `self`. Default: ``None``.

--- a/mindspore/python/mindspore/nn/extend/embedding.py
+++ b/mindspore/python/mindspore/nn/extend/embedding.py
@ -49,7 +49,7 @@ class Embedding(Cell):
            of the index in `input`. Default ``False``.
        _weight (Tensor, optional): Used to initialize the weight of Embedding. If ``None``, the weight will be
            initialized from normal distribution :math:`{N}(\text{sigma=1.0}, \text{mean=0.0})`. Default ``None``.
-        dtype (mindspore.dtype) : Dtype of Parameters. It is meaningless when `_weight` is not None.
+        dtype (mindspore.dtype, optional) : Dtype of Parameters. It is meaningless when `_weight` is not None.
            Default: ``mindspore.float32``.

    Inputs:
@ -83,7 +83,6 @@ class Embedding(Cell):
          [ 0.00233847 -0.00596091  0.00536799]
          [-0.0024154  -0.01203444  0.00811537]
          [-0.0024154  -0.01203444  0.00811537]]
-
         [[ 0.00233847 -0.00596091  0.00536799]
          [ 0.00233847 -0.00596091  0.00536799]
          [-0.0024154  -0.01203444  0.00811537]
--- a/mindspore/python/mindspore/ops/function/math_func.py
+++ b/mindspore/python/mindspore/ops/function/math_func.py
@ -3646,7 +3646,7 @@ def nanmedian(input, axis=-1, keepdims=False):

    .. warning::
        `indices` does not necessarily contain the first occurrence of each median value found in the `input`,
-          unless it is unique.
+        unless it is unique.

    Args:
        input (Tensor): The input tensor to calculate the median and indices.