Merge branch 'develop' into ApiEnhance30

Manfredss · web-flow · commit ad606f035e35 · 2025-11-24T18:29:42.000-08:00
diff --git a/_typos.toml b/_typos.toml
@@ -36,16 +36,7 @@ setted = "setted"
 shoule = "shoule"
 similary = "similary"
 simplier = "simplier"
-skiped = "skiped"
 softwares = "softwares"
-specail = "specail"
 sperated = "sperated"
-splited = "splited"
 splitted = "splitted"
-stirng = "stirng"
-stradegy = "stradegy"
-subract = "subract"
-sucess = "sucess"
-sucessor = "sucessor"
-sucessors = "sucessors"
 szie = "szie"
diff --git a/ci_scripts/check_api_cn.sh b/ci_scripts/check_api_cn.sh
@@ -15,7 +15,7 @@ function filter_cn_api_files() {
     local __resultvar=$2
     local need_check_files=""
     for file in `echo $git_files`;do
-        grep 'code-block:: python' ../docs/$file > /dev/null
+        grep -E 'code-block:: (python|pycon)' ../docs/$file > /dev/null
         if [ $? -eq 0 ] ;then
             api_file=`echo $file | sed 's#api/##g'`
             grep -w "${api_file}" ${DIR_PATH}/api_white_list.txt > /dev/null
diff --git a/ci_scripts/check_api_parameters.py b/ci_scripts/check_api_parameters.py
@@ -184,7 +184,7 @@ def check_api_parameters(rstfiles, apiinfo):
     """check function's parameters same as its origin definition.
 
     TODO:
-    1. All the documents of classes are skiped now. As
+    1. All the documents of classes are skipped now. As
         (1) there ars many class methods in documents, may break the scripts.
         (2) parameters of Class should be checked with its `__init__` method.
     2. Some COMPLICATED annotations may break the scripts.
diff --git a/docs/api/copy_codes_from_en_doc.py b/docs/api/copy_codes_from_en_doc.py
@@ -105,7 +105,7 @@ def extract_code_blocks_from_docstr(docstr, google_style=True):
 
     lastlineindex = len(docstr_list) - 1
 
-    cb_start_pat = re.compile(r"code-block::\s*python")
+    cb_start_pat = re.compile(r"code-block::\s*(python|python-console|pycon)")
     cb_param_pat = re.compile(r"^\s*:(\w+):\s*(\S*)\s*$")
 
     cb_info = {}
@@ -261,7 +261,7 @@ def insert_codes_into_cn_rst_if_need(cnrstfilename):
         cb_new = []
         indent = cf_info["indent"]
         cb_new.append("")  # insert a empty line in the frontend
-        cb_new.append(" " * indent + ".. code-block:: python")
+        cb_new.append(" " * indent + ".. code-block:: pycon")
         if cf_info["cb_name"]:
             cb_new.append(" " * (indent + 3) + ":name: " + cf_info["cb_name"])
         cb_new.append("")
diff --git a/docs/api/extract_api_from_docs.py b/docs/api/extract_api_from_docs.py
@@ -55,7 +55,9 @@ def extract_code_blocks_from_rst(docstr):
     ds_list = docstr.expandtabs(tabsize=4).split("\n")
     lastlineindex = len(ds_list) - 1
     cb_started = False
-    cb_start_pat = re.compile(r"((code)|(code-block))::\s*i?python[23]?")
+    cb_start_pat = re.compile(
+        r"((code)|(code-block))::\s*i?(python|python-console|pycon)"
+    )
     cb_cur = []
     cb_cur_indent = -1
     for lineno, linecont in enumerate(ds_list):
diff --git a/docs/api/gen_doc.py b/docs/api/gen_doc.py
@@ -262,8 +262,8 @@ def parse_module_file(mod):
             return
         logger.debug("parsing %s:%s", mod_name, src_file)
         if len(mod_name) >= 6 and mod_name[:6] == "paddle":
-            fn_splited = os.path.splitext(src_file)
-            if len(fn_splited) > 1 and fn_splited[1].lower() == ".py":
+            fn_split = os.path.splitext(src_file)
+            if len(fn_split) > 1 and fn_split[1].lower() == ".py":
                 mod_ast = ast.parse(open(src_file, "r").read())
                 for node in mod_ast.body:
                     short_names = []
diff --git a/docs/api/paddle/cosh_cn.rst b/docs/api/paddle/cosh_cn.rst
@@ -3,7 +3,7 @@
 cosh
 -------------------------------
 
-.. py:function:: paddle.cosh(x, name=None)
+.. py:function:: paddle.cosh(x, name=None, *, out=None)
 
 
 
@@ -21,6 +21,7 @@ cosh
 
     - **x** (Tensor) - 输入的 Tensor，数据类型为：float32、float64 、float16、complex64、complex128。
     - **name** (str，可选) - 具体用法请参见 :ref:`api_guide_Name`，一般无需设置，默认值为 None。
+    - **out** (Tensor，可选)- 输出的结果。该参数为仅关键字参数，默认值为 None。
 
 返回
 ::::::::::::
diff --git a/docs/api/paddle/device/memory_allocated_cn.rst b/docs/api/paddle/device/memory_allocated_cn.rst
@@ -22,7 +22,7 @@ memory_allocated
 
 代码示例
 ::::::::::::
-.. code-block:: python
+.. code-block:: pycon
 
     >>> import paddle
     >>> paddle.device.memory_allocated('npu:0')
diff --git a/docs/api/paddle/device/memory_reserved_cn.rst b/docs/api/paddle/device/memory_reserved_cn.rst
@@ -19,7 +19,7 @@ memory_reserved
 
 代码示例
 ::::::::::::
-.. code-block:: python
+.. code-block:: pycon
 
     >>> import paddle
     >>> paddle.device.memory_reserved('npu:0')
diff --git a/docs/api/paddle/device/reset_max_memory_allocated_cn.rst b/docs/api/paddle/device/reset_max_memory_allocated_cn.rst
@@ -19,7 +19,7 @@ reset_max_memory_allocated
 
 代码示例
 ::::::::::::
-.. code-block:: python
+.. code-block:: pycon
 
     >>> import paddle
     >>> paddle.device.reset_max_memory_allocated('npu:0')
diff --git a/docs/api/paddle/device/reset_max_memory_reserved_cn.rst b/docs/api/paddle/device/reset_max_memory_reserved_cn.rst
@@ -19,7 +19,7 @@ reset_max_memory_reserved
 
 代码示例
 ::::::::::::
-.. code-block:: python
+.. code-block:: pycon
 
     >>> import paddle
     >>> paddle.device.reset_max_memory_reserved('npu:0')
diff --git a/docs/api/paddle/nn/LayerNorm_cn.rst b/docs/api/paddle/nn/LayerNorm_cn.rst
@@ -3,7 +3,7 @@
 LayerNorm
 -------------------------------
 
-.. py:class:: paddle.nn.LayerNorm(normalized_shape, epsilon=1e-05, weight_attr=None, bias_attr=None, name=None)
+.. py:class:: paddle.nn.LayerNorm(normalized_shape, epsilon=1e-05, *, elementwise_affine=True, bias=True, device=None, dtype=None, weight_attr=None, bias_attr=None, name=None)
 
 构建 ``LayerNorm`` 类的一个可调用对象，具体用法参照 ``代码示例``。其中实现了层归一化层（Layer Normalization Layer）的功能，其可以应用于小批量输入数据。更多详情请参考：`Layer Normalization <https://arxiv.org/pdf/1607.06450v1.pdf>`_ 。
 
@@ -28,9 +28,21 @@ LayerNorm
 
     - **normalized_shape** (int|list|tuple) – 需规范化的 shape，期望的输入 shape 为 ``[*, normalized_shape[0], normalized_shape[1], ..., normalized_shape[-1]]`` 。如果是单个整数，则此模块将在最后一个维度上规范化（此时最后一维的维度需与该参数相同）。
     - **epsilon** (float，可选) - 指明在计算过程中是否添加较小的值到方差中以防止除零。默认值：1e-05。
+      ``别名: eps``
+    - **elementwise_affine** (bool，可选) - 该模块是否具有可学习的仿射参数（weight 和 bias）。如果设置为 False，将不会创建可学习的参数，无论 bias、weight_attr 和 bias_attr 如何设置。默认值：True。
+      ``注意： 此参数必须以关键字参数的形式传入``
+    - **bias** (bool，可选) - 该模块是否具有可学习的偏置（bias）。如果设置为 False，将不会创建可学习的偏置，无论 bias_attr 如何设置。默认值：True。
+      ``注意： 此参数必须以关键字参数的形式传入``
+    - **device** (PlaceLike，可选) - 计算发生的设备。默认值：None。
+      ``注意： 此参数必须以关键字参数的形式传入``
+    - **dtype** (DTypeLike，可选) - 权重和偏置的数据类型。默认值：None。
+      ``注意： 此参数必须以关键字参数的形式传入``
     - **weight_attr** (ParamAttr|bool|None, 可选) - 用于指定可训练的增益参数 :math:`g` 的属性。如果为 False，则不使用权重（即权重参数为 None）；如果为 None，则会使用一个默认的 ParamAttr 作为权重的属性设置，该属性将权重初始化为 1。默认值为 None，表示使用默认的权重属性。具体用法请参见 :ref:`cn_api_paddle_ParamAttr`。
+      ``注意： 此参数必须以关键字参数的形式传入``
     - **bias_attr** (ParamAttr|bool|None, 可选) - 用于指定可训练的偏置参数 :math:`b` 的属性。如果为 False，则不使用偏置（即偏置参数为 None）；如果为 None，则会使用一个默认的 ParamAttr 作为偏置的属性设置，该属性将偏置初始化为 0。默认值为 None，表示使用默认的偏置属性。具体用法请参见 :ref:`cn_api_paddle_ParamAttr`。
+      ``注意： 此参数必须以关键字参数的形式传入``
     - **name** (str，可选) - 具体用法请参见 :ref:`api_guide_Name`，一般无需设置，默认值为 None。
+      ``注意： 此参数必须以关键字参数的形式传入``
 
 形状
 ::::::::::::
diff --git a/docs/design/memory/memory_optimization.md b/docs/design/memory/memory_optimization.md
@@ -3,7 +3,7 @@
 
 ## Problem
 
-In a lecture from Andrew Ng, he attributes the recent sucess of AI due to a combination of these:
+In a lecture from Andrew Ng, he attributes the recent success of AI due to a combination of these:
 
 - Availability of Big Data
 - Supercomputing power to process this Big Data over very large neural networks
@@ -74,7 +74,7 @@ A simple way to perform data-flow analysis of programs is to set up dataflow equ
 
 - Flow Graph Terminology
 
-A flow graph node has out-edges that lead to sucessor nodes, and in-edges that come from predecessor nodes. The set *pred[n]* is all the predecessors of node n, and *succ[n]* is the set of sucessors.
+A flow graph node has out-edges that lead to successor nodes, and in-edges that come from predecessor nodes. The set *pred[n]* is all the predecessors of node n, and *succ[n]* is the set of successors.
 In former control flow graph, the out-edges of node 5 are 5 --> 6 and 5 --> 2, and *succ[5]* = {2, 6}. The in-edges of 2 are 5 --> 2 and 1 --> 2, and *pred[2]* = {1, 5}.
 
 - Uses and Defs
@@ -147,7 +147,7 @@ We can transfer all the operators and variables in ProgramDesc to build a contro
 ```python
 class ControlFlowGraph:
     def __init__(self, Program):
-        self._sucessors = defaultdict(set)
+        self._successors = defaultdict(set)
         self._presucessors = defaultdict(set)
         self._uses = defaultdict(set)
         self._defs = defaultdict(set)
diff --git a/docs/design/modules/python_api.md b/docs/design/modules/python_api.md
@@ -116,7 +116,7 @@ class Operator:
                  block,  # Block
                  type,   # string
                  inputs, # dict<string, Variable>
-                 outputs,# dict<stirng, Variable>
+                 outputs,# dict<string, Variable>
                  attrs   # dict<string, Any>
                  ):
         self.desc = core.NewOpDesc(block.desc, type, inputs, outputs, attrs)
diff --git a/docs/dev_guides/api_contributing_guides/api_design_guidelines_standard_cn.md b/docs/dev_guides/api_contributing_guides/api_design_guidelines_standard_cn.md
@@ -242,7 +242,7 @@
    |  paddle2.0 之前  | pytorch |  numpy   | tensorflow  |   paddle2.0 之后   |
    | :------------- | :----- | :------ | :--------- | :--------------- |
    | elementwise_add |   add   |   add    |     add     |        add        |
-   | elementwise_sub |   sub   | subtract |  subtract   |      subract      |
+   | elementwise_sub |   sub   | subtract |  subtract   |      subtract      |
    | elementwise_mul |   mul   | multiply |  multiply   |     multiply      |
    | elementwise_div |   div   |  divide  |   divide    | divide |
    | elementwise_min |   min   | minimum  |   minimum   |      minimum      |
diff --git a/docs/dev_guides/api_contributing_guides/api_docs_guidelines_cn.md b/docs/dev_guides/api_contributing_guides/api_docs_guidelines_cn.md
@@ -52,7 +52,7 @@
             N-D Tensor. A location into which the result is stored. It’s dimension equals with :attr:`x`.
 
         Examples:
-            .. code-block:: python
+            .. code-block:: pycon
 
                 >>> import paddle
 
@@ -305,7 +305,7 @@ API 抛出异常部分，由于历史原因写在文档中，建议在源码的
 def api():
     """
     Examples:
-        .. code-block:: python
+        .. code-block:: pycon
 
             示例代码位置
 
@@ -318,12 +318,12 @@ def api():
 def api():
     """
     Examples:
-        .. code-block:: python
+        .. code-block:: pycon
             :name: example-1
 
             示例代码位置
 
-        .. code-block:: python
+        .. code-block:: pycon
             :name: example-2
 
             示例代码位置 (存在多个示例代码)
@@ -336,7 +336,7 @@ def api():
 def multiply(x, y, axis=-1, name=None):
     """
     Examples:
-        .. code-block:: python
+        .. code-block:: pycon
 
             >>> import paddle
 
diff --git a/docs/dev_guides/api_contributing_guides/new_cpp_op_cn.md b/docs/dev_guides/api_contributing_guides/new_cpp_op_cn.md
@@ -818,7 +818,7 @@ def trace(
         Tensor: the output data type is the same as input data type.
 
     Examples:
-        .. code-block:: python
+        .. code-block:: pycon
 
             >>> import paddle
 
diff --git a/docs/eval/【Hackathon No.113】 PR.md b/docs/eval/【Hackathon No.113】 PR.md
@@ -105,13 +105,13 @@ else:
 ```python
 if self._nranks > 1:
     dist.init_parallel_env()
-    stradegy = fluid.dygraph.parallel.ParallelStrategy()
-    stradegy.nranks = ParallelEnv().nranks
-    stradegy.local_rank = ParallelEnv().local_rank
-    stradegy.trainer_endpoints = ParallelEnv().trainer_endpoints
-    stradegy.current_endpoint = ParallelEnv().current_endpoint
+    strategy = fluid.dygraph.parallel.ParallelStrategy()
+    strategy.nranks = ParallelEnv().nranks
+    strategy.local_rank = ParallelEnv().local_rank
+    strategy.trainer_endpoints = ParallelEnv().trainer_endpoints
+    strategy.current_endpoint = ParallelEnv().current_endpoint
     self.ddp_model = fluid.dygraph.parallel.DataParallel(
-        self.model.network, stradegy)
+        self.model.network, strategy)
 
 ```
 这样的话，分布式的环境与分布式的模型就已经在 paddle.Model 内的内部构建完成了。
diff --git a/docs/guides/advanced/auto_type_promotion_en.md b/docs/guides/advanced/auto_type_promotion_en.md
@@ -230,7 +230,7 @@ Number | API | Tensor-to-Tensor | Tensor-to-Scalar |
 31 | huber_loss | Common Rule | - |
 32 | mse_loss | Common Rule | - |
 
-There are two specail rules in this table above:
+There are two special rules in this table above:
 
 -  Divide Rule: For divide API, it will not return dtype smaller than float. Such as int32 / Scalar returns float32.
 
diff --git a/docs/guides/custom_op/new_cpp_op_cn.md b/docs/guides/custom_op/new_cpp_op_cn.md
diff --git a/docs/hardware_support/hardware_info_cn.md b/docs/hardware_support/hardware_info_cn.md
