resolve merge conflict in _typos.toml

fsylmxx · fsylmxx · commit 3ca827a5badb · 2025-10-20T15:04:41.000+08:00
diff --git a/.github/CODEOWNERS b/.github/CODEOWNERS
@@ -1,2 +1,2 @@
 # Paddle API Docs
-docs/api/paddle @jzhang533 @sunzhongkai588 @mattheliu @Echo-Nie
+docs/api/paddle @jzhang533 @sunzhongkai588 @mattheliu @Echo-Nie @ooooo-create
diff --git a/.gitignore b/.gitignore
@@ -6,6 +6,7 @@
 # virtualenv
 venv/
 ENV/
+.venv/
 
 # Compiled Python files
 __pycache__/
diff --git a/_typos.toml b/_typos.toml
@@ -27,21 +27,13 @@ Archetecture = "Archetecture"
 Asynchoronous = "Asynchoronous"
 Attrbute = "Attrbute"
 Attribtue = "Attribtue"
-Bounary = "Bounary"
 Classfication = "Classfication"
 Comparision = "Comparision"
 Contructing = "Contructing"
 Creenshot = "Creenshot"
-DELCARE = "DELCARE"
-Dateset = "Dateset"
-Discription = "Discription"
-Distrbuted = "Distrbuted"
-Driect = "Driect"
 Embeddding = "Embeddding"
 Embeding = "Embeding"
 Engish = "Engish"
-Fasle = "Fasle"
-Flase = "Flase"
 Generater = "Generater"
 Gloabal = "Gloabal"
 Imporvement = "Imporvement"
@@ -76,20 +68,10 @@ Traning = "Traning"
 Transfomed = "Transfomed"
 Tthe = "Tthe"
 Ture = "Ture"
-Varialble = "Varialble"
-Varible = "Varible"
-Varient = "Varient"
 Wether = "Wether"
 accordding = "accordding"
 accoustic = "accoustic"
 accpetance = "accpetance"
-baisc = "baisc"
-basci = "basci"
-beacuse = "beacuse"
-bechmark = "bechmark"
-benckmark = "benckmark"
-boradcast = "boradcast"
-brodcast = "brodcast"
 caculate = "caculate"
 cantains = "cantains"
 choosen = "choosen"
@@ -253,12 +235,6 @@ transfered = "transfered"
 trasformed = "trasformed"
 treshold = "treshold"
 trian = "trian"
-varialbes = "varialbes"
-varibale = "varibale"
-varibales = "varibales"
-varience = "varience"
-varient = "varient"
-visting = "visting"
 warpped = "warpped"
 wether = "wether"
 wiht = "wiht"
diff --git a/docs/api/paddle/incubate/autograd/Overview_cn.rst b/docs/api/paddle/incubate/autograd/Overview_cn.rst
@@ -225,7 +225,7 @@ _________________________
 - 输入数据不支持可变形状写法，如[None, 1]、[-1, 1]。如果训练数据形状是变化的，一种可行 Workaround 方案是根据不同数据形状创建不同网络，即在组网阶段将形状固定，具体参考附 1 代码。
 - 我们尚未在 windows 平台进行完整验证和支持。
 - 目前只支持使用 default_main_program 和 default_startup_program。
-- boradcast 语意尚未完整支持。
+- broadcast 语意尚未完整支持。
 
 
 .. _autograd_design_details:
diff --git a/docs/api/paddle/static/IpuStrategy_cn.rst b/docs/api/paddle/static/IpuStrategy_cn.rst
@@ -46,7 +46,7 @@ set_pipelining_config(self, enable_pipelining, batches_per_step, enable_gradient
 
     - **enable_pipelining** (bool，可选)- 是否使能子图之间的数据流水线。仅支持当 enable_manual_shard=True 时，enable_pipelining 可以置为 True。默认值为 False，表示不使能该功能。
     - **batches_per_step** (int，可选)- 指定数据流水线每次运算多少个 batch 的数据。默认值为 1，表示不使能数据流水线功能。
-    - **enable_gradient_accumulation** (bool，可选)- 是否使能梯度累积，只用于训练模式。默认值为 Flase，表示不使能梯度累积功能。
+    - **enable_gradient_accumulation** (bool，可选)- 是否使能梯度累积，只用于训练模式。默认值为 False，表示不使能梯度累积功能。
     - **accumulation_factor** (int，可选)- 指定累积运算多少个 batch 更新一次权重。默认值为 1，表示不使能权重累积更新功能。
 
 **代码示例**
diff --git a/docs/design/concepts/tensor.md b/docs/design/concepts/tensor.md
@@ -161,7 +161,7 @@ Please reference the section of `Learn from Majel` for more details.
 
 `ArrayView` is an encapsulation of `Array`， which introduces extra iterator methods, such as `begin()` and `end()`. The `begin()` method returns an iterator pointing to the first element in the ArrayView. And the `end()` method returns an iterator pointing to the pass-the-end element in the ArrayView.
 
-`ArrayView` make the visting and manipulating an array more efficiently, flexibly and safely.
+`ArrayView` make the visiting and manipulating an array more efficiently, flexibly and safely.
 
 
 A global function `make_view` is provided to transform an array to corresponding arrayview.
diff --git a/docs/design/concepts/tensor_array.md b/docs/design/concepts/tensor_array.md
@@ -212,7 +212,7 @@ class TensorArray:
 ```
 
 ## DenseTensor-related Supports
-The `RecurrentGradientMachine` in Paddle serves as a flexible RNN layer; it takes varience-length sequences as input, and output sequences too.
+The `RecurrentGradientMachine` in Paddle serves as a flexible RNN layer; it takes variable-length sequences as input, and output sequences too.
 
 Since each step of RNN can only take a tensor-represented batch of data as input,
 some preprocess should be taken on the inputs such as sorting the sentences by their length in descending order and cut each word and pack to new batches.
@@ -244,10 +244,10 @@ def pack(level, indices_map):
     pass
 ```
 
-With these two methods, a varience-length sentence supported RNN can be implemented like
+With these two methods, a variable-length sentence supported RNN can be implemented like
 
 ```c++
-// input is the varient-length data
+// input is the variable-length data
 LodTensor sentence_input(xxx);
 TensorArray ta;
 Tensor indice_map;
@@ -268,4 +268,4 @@ for (int step = 0; step = ta.size(); step++) {
 DenseTensor rnn_output = ta.pack(ta, indice_map);
 ```
 the code above shows that by embedding the DenseTensor-related preprocess operations into `TensorArray`,
-the implementation of a RNN that supports varient-length sentences is far more concise than `RecurrentGradientMachine` because the latter mixes all the codes together, hard to read and extend.
+the implementation of a RNN that supports variable-length sentences is far more concise than `RecurrentGradientMachine` because the latter mixes all the codes together, hard to read and extend.
diff --git a/docs/design/dist_train/mpi_enabled_design.md b/docs/design/dist_train/mpi_enabled_design.md
@@ -8,7 +8,7 @@ When we do distribute multi GPU training, the communication overhead between ser
 
 We will use OpenMPI API to PaddlePaddle, which can bring two benefits to PaddlePaddle:
 1. Enable RDMA with PaddlePaddle, which bring high-performance low latency networks.
-2. Enable GPUDriect with PaddlePaddle, which bring the highest throughput and lowest latency GPU read and write.
+2. Enable GPUDirect with PaddlePaddle, which bring the highest throughput and lowest latency GPU read and write.
 
 # Change list
 * Compile args: Need add compile args to enable MPI support.
diff --git a/docs/design/dynamic_rnn/rnn_design_en.md b/docs/design/dynamic_rnn/rnn_design_en.md
@@ -1,4 +1,4 @@
-# Varient Length supported RNN Design
+# Variable Length supported RNN Design
 For the learning of variable length sequences, the existing mainstream frameworks such as tensorflow, pytorch, caffe2, mxnet and so on all use padding.
 
 Different-length sequences in a mini-batch will be padded with zeros and transformed to same length.
diff --git a/docs/design/modules/backward.md b/docs/design/modules/backward.md
@@ -61,7 +61,7 @@ def _append_backward_ops_(target,
         target_block(Block): the block which is going to hold new generated grad ops
         no_grad_dict(dict):
             key(int)  block index
-            val(set) a set of varibale names. These varibales have no gradient
+            val(set) a set of variable names. These variables have no gradient
         grad_to_var(dict)(output argument):
             key(str): grad variable name
             val(str): corresponding forward variable name
diff --git a/docs/design/modules/net_op_design.md b/docs/design/modules/net_op_design.md
@@ -90,7 +90,7 @@ class PlainNet : public Net {
   // Create a network describe by `def`.  NetDesc is the definition of a network.
   PlainNet(const NetDesc &def);
 
-  // Infer all the operators' input and output varialbes' shapes, will be called before every mini-batch
+  // Infer all the operators' input and output variables' shapes, will be called before every mini-batch
   training.
   virtual Error InferShape(Scope *scope) override;
 
diff --git a/docs/design/others/gan_api.md b/docs/design/others/gan_api.md
@@ -58,7 +58,7 @@ class DCGAN:
     self.D_b0 = pd.Variable(np.zeros(128)) # variable also support initialization using a  numpy data
     self.D_W1 = pd.Variable(shape=[784, 128], data=pd.gaussian_normal_randomizer())
     self.D_b1 = pd.Variable(np.zeros(128)) # variable also support initialization using a  numpy data
-    self.D_W2 = pd.Varialble(np.random.rand(128, 1))
+    self.D_W2 = pd.Variable(np.random.rand(128, 1))
     self.D_b2 = pd.Variable(np.zeros(128))
     self.theta_D = [self.D_W0, self.D_b0, self.D_W1, self.D_b1, self.D_W2, self.D_b2]
 
@@ -67,7 +67,7 @@ class DCGAN:
     self.G_b0 = pd.Variable(np.zeros(128)) # variable also support initialization using a  numpy data
     self.G_W1 = pd.Variable(shape=[784, 128], data=pd.gaussian_normal_randomizer())
     self.G_b1 = pd.Variable(np.zeros(128)) # variable also support initialization using a  numpy data
-    self.G_W2 = pd.Varialble(np.random.rand(128, 1))
+    self.G_W2 = pd.Variable(np.random.rand(128, 1))
     self.G_b2 = pd.Variable(np.zeros(128))
     self.theta_G = [self.G_W0, self.G_b0, self.G_W1, self.G_b1, self.G_W2, self.G_b2]
 ```
diff --git a/docs/design/others/graph_survey.md b/docs/design/others/graph_survey.md
@@ -28,7 +28,7 @@ def get_symbol(num_classes=10, **kwargs):
 
 
 
-Varible here is actually a Symbol. Every basic Symbol will correspond to one Node, and every Node has its own AnyAttr. There is a op field in AnyAttr class, when a Symbol represents Variable(often input data), the op field is null.
+Variable here is actually a Symbol. Every basic Symbol will correspond to one Node, and every Node has its own AnyAttr. There is a op field in AnyAttr class, when a Symbol represents Variable(often input data), the op field is null.
 
 Symbol contains a data member, std::vector<NodeEntry> outputs, and NodeEntry cantains a poniter to Node. We can follow the Node pointer to get all the Graph.
 
@@ -168,9 +168,9 @@ Expression pred = W * in;
 Expression loss = square(pred - label);
 ```
 
-The input data and parameter are also represented by Expression. Every basci Expression corresponds to a Node. And input data is also a Node.
+The input data and parameter are also represented by Expression. Every basic Expression corresponds to a Node. And input data is also a Node.
 
-Expression has a data member ComputationGraph, and ComputationGraph will be modified in users' configuring process. Expression can be a running target, beacuse Expression contains all dependency.
+Expression has a data member ComputationGraph, and ComputationGraph will be modified in users' configuring process. Expression can be a running target, because Expression contains all dependency.
 
 
 Here is a detailed example:
diff --git a/docs/design/phi/design_cn.md b/docs/design/phi/design_cn.md
@@ -253,7 +253,7 @@ enum class Backend : uint8_t {
   NUM_BACKENDS,
 
   /**
-   * [ Why we need ALL in baisc kernel key member? ]
+   * [ Why we need ALL in basic kernel key member? ]
    *
    * For Tensor, ALL represents an illegal Backend, but for Kernel, some
    * kernels may be device-independent by nature, such as reshape;
@@ -382,7 +382,7 @@ enum class DataType {
   BFLOAT16,
 
   NUM_DATA_TYPES,
-  // See Note [ Why we need ALL in baisc kernel key member? ]
+  // See Note [ Why we need ALL in basic kernel key member? ]
   ALL_DTYPE = UNDEFINED,
 };
 ```
@@ -1134,7 +1134,7 @@ class CompatMetaTensor : public phi::MetaTensor {
 通过前面介绍的 PT_INFER_META 宏归一化函数形式，然后将`PT_INFER_META(***InferMeta)`包装到一个 functor 中，functor 中先将 InferShapeContext 转换为 InferMetaContext，再调用相应 InferMeta 函数，通过一个宏统一管理代码
 
 ```
-##define DELCARE_INFER_SHAPE_FUNCTOR(op_type, functor_name, fn)      \
+##define DECLARE_INFER_SHAPE_FUNCTOR(op_type, functor_name, fn)      \
   struct functor_name : public paddle::framework::InferShapeBase {  \
     void operator()(                                                \
         paddle::framework::InferShapeContext* ctx) const override { \
@@ -1176,7 +1176,7 @@ class SignOp : public framework::OperatorWithKernel {
   using framework::OperatorWithKernel::OperatorWithKernel;
 };
 
-DELCARE_INFER_SHAPE_FUNCTOR(
+DECLARE_INFER_SHAPE_FUNCTOR(
     sign, SignInferShapeFunctor, PT_INFER_META(phi::UnchangedInferMetaNew));
 REGISTER_OPERATOR(sign, ops::SignOp, ops::SignOpMaker<float>,
                   ops::SignGradMaker<paddle::framework::OpDesc>,
diff --git a/docs/design/phi/design_en.md b/docs/design/phi/design_en.md
@@ -253,7 +253,7 @@ enum class Backend : uint8_t {
   NUM_BACKENDS,
 
   /**
-   * [ Why we need ALL in baisc kernel key member? ]
+   * [ Why we need ALL in basic kernel key member? ]
    *
    * For Tensor, ALL represents an illegal Backend, but for Kernel, some
    * kernels may be device-independent by nature, such as reshape;
@@ -382,7 +382,7 @@ enum class DataType {
   BFLOAT16,
 
   NUM_DATA_TYPES,
-  // See Note [ Why we need ALL in baisc kernel key member? ]
+  // See Note [ Why we need ALL in basic kernel key member? ]
   ALL_DTYPE = UNDEFINED,
 };
 ```
@@ -1134,7 +1134,7 @@ Then, in order to transplant the functional `InferMeta` back to the Op architect
 Normalize the function form through the `PT_INFER_META` macro, and then wrap `PT_INFER_META(***InferMeta)` into a functor. The functor first converts the `InferShapeContext` to `InferMetaContext`, then calls the corresponding `InferMeta` function, and manages the code uniformly through a macro.
 
 ```
-##define DELCARE_INFER_SHAPE_FUNCTOR(op_type, functor_name, fn)      \
+##define DECLARE_INFER_SHAPE_FUNCTOR(op_type, functor_name, fn)      \
   struct functor_name : public paddle::framework::InferShapeBase {  \
     void operator()(                                                \
         paddle::framework::InferShapeContext* ctx) const override { \
@@ -1176,7 +1176,7 @@ class SignOp : public framework::OperatorWithKernel {
   using framework::OperatorWithKernel::OperatorWithKernel;
 };
 
-DELCARE_INFER_SHAPE_FUNCTOR(
+DECLARE_INFER_SHAPE_FUNCTOR(
     sign, SignInferShapeFunctor, PT_INFER_META(phi::UnchangedInferMetaNew));
 REGISTER_OPERATOR(sign, ops::SignOp, ops::SignOpMaker<float>,
                   ops::SignGradMaker<paddle::framework::OpDesc>,
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
@@ -629,7 +629,7 @@
 | 最近邻回归 | nearest neighbor regression | |
 | 最近邻 | nearest neighbor | |
 | 验证集 | validation set | |
-| 基准 | bechmark | |
+| 基准 | benchmark | |
 | 基准 | baseline | |
 | 点估计 | point estimator | |
 | 估计量 | estimator | |
diff --git a/docs/dev_guides/op_optimization/kernel_primitive_api/io_api_cn.md b/docs/dev_guides/op_optimization/kernel_primitive_api/io_api_cn.md
@@ -79,7 +79,7 @@ __device__ void ReadDataBc(T* dst, const T* src,
 
 ### 函数说明
 
-将需要进行 brodcast 的 2D 数据按照 T 类型从全局内存 src 中读取到寄存器 dst 中，其中 src 为原始输入数据指针，根据 config 计算当前输出数据对应的输入数据坐标，并将坐标对应的数据读取到寄存器中。</br>
+将需要进行 broadcast 的 2D 数据按照 T 类型从全局内存 src 中读取到寄存器 dst 中，其中 src 为原始输入数据指针，根据 config 计算当前输出数据对应的输入数据坐标，并将坐标对应的数据读取到寄存器中。</br>
 数据处理过程如下：</br>
 ![ReadDataBc](./images/io_read_data_broadcast_stride.png)
 
@@ -116,7 +116,7 @@ __device__ void ReadDataBc(T* dst, const T* src,
 
 ### 函数说明
 
-将需要进行 brodcast 的 1D 数据按照 T 类型从全局内存 src 中读取到寄存器 dst 中，其中 src 为原始输入数据指针，根据 config 计算当前输出数据对应的输入数据坐标，并将坐标对应的数据读取到寄存器中。</br>
+将需要进行 broadcast 的 1D 数据按照 T 类型从全局内存 src 中读取到寄存器 dst 中，其中 src 为原始输入数据指针，根据 config 计算当前输出数据对应的输入数据坐标，并将坐标对应的数据读取到寄存器中。</br>
 数据处理过程如下：</br>
 ![ReadDataBc](./images/io_read_data_broadcast.png)
 
diff --git a/docs/dev_guides/op_optimization/kernel_primitive_api/io_api_en.md b/docs/dev_guides/op_optimization/kernel_primitive_api/io_api_en.md
@@ -77,7 +77,7 @@ __device__ void ReadDataBc(T* dst, const T* src,
 
 ### Description
 
-Read the 2D data that needs to be brodcast from the global memory src into the register dst according to the T type, where src is the original input data pointer, calculate the input data coordinates corresponding to the current output data according to config, and read the data corresponding to the coordinates to the register. </br>
+Read the 2D data that needs to be broadcast from the global memory src into the register dst according to the T type, where src is the original input data pointer, calculate the input data coordinates corresponding to the current output data according to config, and read the data corresponding to the coordinates to the register. </br>
 The data processing process is as follows:</br>
 ![ReadDataBc](./images/io_read_data_broadcast_stride.png)
 
@@ -114,7 +114,7 @@ __device__ void ReadDataBc(T* dst, const T* src,
 
 ### Description
 
-Read the 1D data that needs to be brodcast from the global memory src into the register dst according to the T type, where src is the original input data pointer, calculate the input data coordinates corresponding to the current output data according to config, and read the data corresponding to the coordinates to the register. </br>
+Read the 1D data that needs to be broadcast from the global memory src into the register dst according to the T type, where src is the original input data pointer, calculate the input data coordinates corresponding to the current output data according to config, and read the data corresponding to the coordinates to the register. </br>
 The data processing process is as follows:</br>
 ![ReadDataBc](./images/io_read_data_broadcast.png)
 
diff --git a/docs/dev_guides/op_optimization/kernel_primitive_api/reduce_example_cn.md b/docs/dev_guides/op_optimization/kernel_primitive_api/reduce_example_cn.md
@@ -84,7 +84,7 @@ __global__ void ReduceHigherDimKernel(const Tx* x, Ty* y, ReduceOp reducer,
 
   } else {
 
-    // The remaining data is smaller than blockdim.x, IsBounary must be true
+    // The remaining data is smaller than blockdim.x, IsBoundary must be true
     HigherDimImpl<Tx, Ty, MPType, AddFunctor<Tx, Ty>, IdentityFunctor<Tx, Ty>, true>(
         x, y, AddFunctor<Tx, Ty>(), IdentityFunctor<Tx, Ty>(), init, reduce_num, left_num, blocking_num);
 
diff --git a/docs/dev_guides/op_optimization/kernel_primitive_api/reduce_example_en.md b/docs/dev_guides/op_optimization/kernel_primitive_api/reduce_example_en.md
@@ -83,7 +83,7 @@ __global__ void ReduceHigherDimKernel(const Tx* x, Ty* y, ReduceOp reducer,
 
   } else {
 
-    // The remaining data is smaller than blockdim.x, IsBounary must be true
+    // The remaining data is smaller than blockdim.x, IsBoundary must be true
     HigherDimImpl<Tx, Ty, MPType, AddFunctor<Tx, Ty>, IdentityFunctor<Tx, Ty>, true>(
         x, y, AddFunctor<Tx, Ty>(), IdentityFunctor<Tx, Ty>(), init, reduce_num, left_num, blocking_num);
 
diff --git a/docs/eval/evaluation_of_docs_system.md b/docs/eval/evaluation_of_docs_system.md
diff --git a/docs/guides/06_distributed_training/group_sharded_parallel_cn.rst b/docs/guides/06_distributed_training/group_sharded_parallel_cn.rst
diff --git a/docs/guides/infer/paddleslim/paddle_slim_en.rst b/docs/guides/infer/paddleslim/paddle_slim_en.rst
diff --git a/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.utils.data.BatchSampler.md b/docs/guides/model_convert/convert_from_pytorch/api_difference/paddle_more_args/torch.utils.data.BatchSampler.md
diff --git a/docs/install/Tables_en.md b/docs/install/Tables_en.md
diff --git a/docs/practices/gan/Pix2Pix/Pix2Pix_with_Cityscapes.ipynb b/docs/practices/gan/Pix2Pix/Pix2Pix_with_Cityscapes.ipynb
diff --git a/docs/practices/gan/cyclegan/cyclegan.ipynb b/docs/practices/gan/cyclegan/cyclegan.ipynb

Original file line number	Diff line number	Diff line change
`@@ -1,2 +1,2 @@`
`1`	`1`	`# Paddle API Docs`
`2`		`-docs/api/paddle @jzhang533 @sunzhongkai588 @mattheliu @Echo-Nie`
	`2`	`+docs/api/paddle @jzhang533 @sunzhongkai588 @mattheliu @Echo-Nie @ooooo-create`
Original file line number	Diff line number	Diff line change
`@@ -1,4 +1,4 @@`
`1`		`-# Varient Length supported RNN Design`
	`1`	`+# Variable Length supported RNN Design`
`2`	`2`	`For the learning of variable length sequences, the existing mainstream frameworks such as tensorflow, pytorch, caffe2, mxnet and so on all use padding.`
`3`	`3`
`4`	`4`	`Different-length sequences in a mini-batch will be padded with zeros and transformed to same length.`