diff --git a/atvc/README.md b/atvc/README.md
index 620e28c13366d2afc4f0cb10eedbcaa8d9ce3b78..854dd6e79268b6a29ee4b18cbb36937bd9dfa8b8 100644
--- a/atvc/README.md
+++ b/atvc/README.md
@@ -82,7 +82,7 @@ Accuracy verification passed.
| [ops_aclnn](./examples/aclnn) | 使用ATVC基于自定义工程算子的实现以及调用样例 | 自定义工程算子 |
| [ops_pytorch](./examples/pytorch) | 使用ATVC基于[pytorch](https://gitee.com/ascend/pytorch)算子的实现以及调用样例 | pytorch算子 |
-更多算子类型介绍和如何选取模板参见参阅[快速入门](./docs/1_quick_start.md),其中add、sinh_custom、add_with_scalar、reduce_sum、broadcast_to是ATVC的直调样例,ops_aclnn为基于ATVC对接aclnn工程的算子目录,ops_pytorch为基于ATVC对接pytorch工程的算子目录。
+更多算子类型介绍和如何选取模板参见参阅[快速入门](./docs/1_quick_start.md),其中add、sinh_custom、add_with_scalar、reduce_sum、broadcast_to是ATVC的直调样例,ops_aclnn为基于ATVC对接aclnn工程的算子目录,ops_pytorch为基于ATVC对接pytorch工程的算子目录。其中,ops_aclnn和ops_pytorch样例需要进入到example路径下按照README.md描述执行。
diff --git a/atvc/examples/aclnn/README.md b/atvc/examples/aclnn/README.md
index 35df784ff75fa94f0c7b5109539eac6065ffaae7..3bfe4acfd50a90daf8b3c0abc7aa46e018c7ccf6 100644
--- a/atvc/examples/aclnn/README.md
+++ b/atvc/examples/aclnn/README.md
@@ -33,7 +33,7 @@
$ cd CustomOp/build_out
$ ./custom_opp*.run
# 样例运行
- $ cd ./atvc/examples/aclnn/add/AclNNInvocationNaive
+ $ cd ../../AclNNInvocationNaive
$ bash run.sh
...
test pass
diff --git a/atvc/examples/aclnn/add/AclNNInvocationNaive/README.md b/atvc/examples/aclnn/add/AclNNInvocationNaive/README.md
index e826fd573f773eb04fa6ade96590eadc9c750501..65acdddea5ed3db4d375d13a1dc2c66d1bd3f856 100644
--- a/atvc/examples/aclnn/add/AclNNInvocationNaive/README.md
+++ b/atvc/examples/aclnn/add/AclNNInvocationNaive/README.md
@@ -33,12 +33,6 @@
将CMakeLists.txt文件内"/usr/local/Ascend/ascend-toolkit/latest"替换为CANN软件包安装后的实际路径。
eg:/home/HwHiAiUser/Ascend/ascend-toolkit/latest
- - 环境变量配置
-
- 需要设置NPU_HOST_LIB环境变量,以x86为例
- ```bash
- export NPU_HOST_LIB=/home/HwHiAiUser/Ascend/ascend-toolkit/latest/x86_64-linux/lib64
- ```
- 样例执行
用户参考run.sh脚本进行编译与运行。
diff --git a/atvc/examples/aclnn/add/AclNNInvocationNaive/run.sh b/atvc/examples/aclnn/add/AclNNInvocationNaive/run.sh
index 7abad744e9245564546a2b97d8b1bba68b659dd5..b085a0612091c9ca77a1253690b3201e21ba6d93 100644
--- a/atvc/examples/aclnn/add/AclNNInvocationNaive/run.sh
+++ b/atvc/examples/aclnn/add/AclNNInvocationNaive/run.sh
@@ -1,4 +1,13 @@
#!/bin/bash
+# Copyright (c) 2025 Huawei Technologies Co., Ltd.
+# This file is a part of the CANN Open Software.
+# Licensed under CANN Open Software License Agreement Version 1.0 (the "License").
+# Please refer to the License for details. You may not use this file except in compliance with the License.
+# THIS SOFTWARE IS PROVIDED ON AN "AS IS" BASIS, WITHOUT WARRANTIES OF ANY KIND, EITHER EXPRESS OR IMPLIED,
+# INCLUDING BUT NOT LIMITED TO NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE.
+# See LICENSE in the root of the software repository for the full text of the License.
+# ======================================================================================================================
+
if [ -n "$ASCEND_INSTALL_PATH" ]; then
_ASCEND_INSTALL_PATH=$ASCEND_INSTALL_PATH
elif [ -n "$ASCEND_HOME_PATH" ]; then
diff --git a/atvc/examples/aclnn/add/README.md b/atvc/examples/aclnn/add/README.md
index 5eb359d02f3ee5c9e50a6c5f3a582b3f880af058..358d7b113437f5663b7e67eae32568dec07b4ec4 100644
--- a/atvc/examples/aclnn/add/README.md
+++ b/atvc/examples/aclnn/add/README.md
@@ -20,11 +20,11 @@ z = x + y
算子类型(OpType) | Add |
算子输入 | name | shape | data type | format |
-x | 8 * 2048 | float,int | ND |
-y | 8 * 2048 | float,int | ND |
+x | 8 * 2048 | float | ND |
+y | 8 * 2048 | float | ND |
-算子输出 | z | 8 * 2048 | float,int | ND |
+算子输出 | z | 8 * 2048 | float | ND |
核函数名 | add_custom |
diff --git a/atvc/examples/aclnn/add/install.sh b/atvc/examples/aclnn/add/install.sh
index 5ca57e7ebfea0ed405ccffa131a6c42ea94a5fe2..2e41482d35f1f6f0ef575eee1c8d696a48e9677c 100644
--- a/atvc/examples/aclnn/add/install.sh
+++ b/atvc/examples/aclnn/add/install.sh
@@ -1,4 +1,13 @@
#!/bin/bash
+# Copyright (c) 2025 Huawei Technologies Co., Ltd.
+# This file is a part of the CANN Open Software.
+# Licensed under CANN Open Software License Agreement Version 1.0 (the "License").
+# Please refer to the License for details. You may not use this file except in compliance with the License.
+# THIS SOFTWARE IS PROVIDED ON AN "AS IS" BASIS, WITHOUT WARRANTIES OF ANY KIND, EITHER EXPRESS OR IMPLIED,
+# INCLUDING BUT NOT LIMITED TO NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE.
+# See LICENSE in the root of the software repository for the full text of the License.
+# ======================================================================================================================
+
SHORT=v:,i:,
LONG=soc-version:,install-path:,
OPTS=$(getopt -a --options $SHORT --longoptions $LONG -- "$@")
diff --git a/atvc/examples/aclnn/reduce_sum/AclNNInvocationNaive/README.md b/atvc/examples/aclnn/reduce_sum/AclNNInvocationNaive/README.md
index 4bc6770ff8e0658d91afb7106907e6c2633c8b31..3fd20f3cac4dce4d504240835e5352e7a7015c8f 100644
--- a/atvc/examples/aclnn/reduce_sum/AclNNInvocationNaive/README.md
+++ b/atvc/examples/aclnn/reduce_sum/AclNNInvocationNaive/README.md
@@ -33,12 +33,6 @@
将CMakeLists.txt文件内"/usr/local/Ascend/ascend-toolkit/latest"替换为CANN软件包安装后的实际路径。
eg:/home/HwHiAiUser/Ascend/ascend-toolkit/latest
- - 环境变量配置
-
- 需要设置NPU_HOST_LIB环境变量,以x86为例
- ```bash
- export NPU_HOST_LIB=/home/HwHiAiUser/Ascend/ascend-toolkit/latest/x86_64-linux/lib64
- ```
- 样例执行
用户参考run.sh脚本进行编译与运行。
diff --git a/atvc/examples/aclnn/reduce_sum/AclNNInvocationNaive/run.sh b/atvc/examples/aclnn/reduce_sum/AclNNInvocationNaive/run.sh
index 9bb88021ddeeb0aaca15af2eb5f463cbe5a3a8fe..a950a390ef9ea4534b68834b1add420a882989b7 100644
--- a/atvc/examples/aclnn/reduce_sum/AclNNInvocationNaive/run.sh
+++ b/atvc/examples/aclnn/reduce_sum/AclNNInvocationNaive/run.sh
@@ -1,4 +1,13 @@
#!/bin/bash
+# Copyright (c) 2025 Huawei Technologies Co., Ltd.
+# This file is a part of the CANN Open Software.
+# Licensed under CANN Open Software License Agreement Version 1.0 (the "License").
+# Please refer to the License for details. You may not use this file except in compliance with the License.
+# THIS SOFTWARE IS PROVIDED ON AN "AS IS" BASIS, WITHOUT WARRANTIES OF ANY KIND, EITHER EXPRESS OR IMPLIED,
+# INCLUDING BUT NOT LIMITED TO NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE.
+# See LICENSE in the root of the software repository for the full text of the License.
+# ======================================================================================================================
+
if [ -n "$ASCEND_INSTALL_PATH" ]; then
_ASCEND_INSTALL_PATH=$ASCEND_INSTALL_PATH
elif [ -n "$ASCEND_HOME_PATH" ]; then
diff --git a/atvc/examples/aclnn/reduce_sum/README.md b/atvc/examples/aclnn/reduce_sum/README.md
index 693753811240e87033d47a734414852b0bad8299..3503c46e6991fe907459745f11f8905230cf6ccc 100644
--- a/atvc/examples/aclnn/reduce_sum/README.md
+++ b/atvc/examples/aclnn/reduce_sum/README.md
@@ -18,10 +18,10 @@ ReduceSum是对输入tensor的指定轴进行规约累加的计算并输出结
算子类型(OpType) | ReduceSum |
算子输入 | name | shape | data type | format |
-x | 8 * 2048 | float,int | ND |
+x | 8 * 2048 | float | ND |
-算子输出 | y | 1 * 2048 | float,int | ND |
+算子输出 | y | 1 * 2048 | float | ND |
核函数名 | reduce_sum_custom |
diff --git a/atvc/examples/aclnn/reduce_sum/install.sh b/atvc/examples/aclnn/reduce_sum/install.sh
index afe455ade5b53bd16ac6b4206650232bf32cf8e2..a3a01bc871550958b1b1dc625542080476edcde1 100644
--- a/atvc/examples/aclnn/reduce_sum/install.sh
+++ b/atvc/examples/aclnn/reduce_sum/install.sh
@@ -1,4 +1,13 @@
#!/bin/bash
+# Copyright (c) 2025 Huawei Technologies Co., Ltd.
+# This file is a part of the CANN Open Software.
+# Licensed under CANN Open Software License Agreement Version 1.0 (the "License").
+# Please refer to the License for details. You may not use this file except in compliance with the License.
+# THIS SOFTWARE IS PROVIDED ON AN "AS IS" BASIS, WITHOUT WARRANTIES OF ANY KIND, EITHER EXPRESS OR IMPLIED,
+# INCLUDING BUT NOT LIMITED TO NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE.
+# See LICENSE in the root of the software repository for the full text of the License.
+# ======================================================================================================================
+
SHORT=v:,i:,
LONG=soc-version:,install-path:,
OPTS=$(getopt -a --options $SHORT --longoptions $LONG -- "$@")
diff --git a/atvc/examples/pytorch/add/README.md b/atvc/examples/pytorch/add/README.md
index 5df331b32080080cb36959de9ae11e4f1adf8b9d..bbd61f3088f05ca4526e7a0aafba85703575a8a2 100644
--- a/atvc/examples/pytorch/add/README.md
+++ b/atvc/examples/pytorch/add/README.md
@@ -33,18 +33,6 @@ z = x + y
本样例支持如下产品型号:
- Atlas A2训练系列产品
-## 算子工程介绍
-其中,算子工程目录AddCustom包含算子的实现文件,如下所示:
-```
-├── add
-│ ├── add_custom_impl.h // 通过pytroch调用的方式调用Add算子
-│ ├── pytorch_ascendc_extension.cpp // pytorch调用入口
-│ ├── run_op.py // pytorch的测试用例
-│ └── run.sh // 脚本,编译需要的二进制文件,并测试
-```
-
-run.sh脚本编译pytorch所需要的算子依赖,并将执行测试用例。
-
## 编译运行样例算子
针对自定义算子工程,编译运行包含如下步骤:
- 完成算子pytorch入口和impl文件的实现;
@@ -54,10 +42,10 @@ run.sh脚本编译pytorch所需要的算子依赖,并将执行测试用例。
详细操作如下所示。
### 1. 获取源码包及环境配置
编译运行此样例前,请参考[准备:获取样例代码](../README.md#codeready)获取源码包及环境变量的准备。
-### 1. 安装pytorch环境
+### 2. 安装pytorch环境
参考[torch的安装](https://gitee.com/ascend/pytorch)进行安装torch、torch_npu环境
-### 2. 基于ATVC编写pytorch算子的实现
+### 3. 基于ATVC编写pytorch算子的实现
- 编写kernel侧函数,参考[add_custom_impl.h](./add_custom_impl.h)
```cpp
// 引入头文件
@@ -138,7 +126,7 @@ run.sh脚本编译pytorch所需要的算子依赖,并将执行测试用例。
run_tests()
```
-### 3. 基于ATVC编写pytorch算子的调用验证
+### 4. 基于ATVC编写pytorch算子的调用验证
- 调用脚本,生成pytorch算子,并运行测试用例
```bash
$ cd ./atvc/examples/pytorch/add
diff --git a/atvc/examples/pytorch/reduce_sum/README.md b/atvc/examples/pytorch/reduce_sum/README.md
index e5115295f6cb02bfc8665dc82c9e0faa4172f104..b951656b9c174399dcae3e8825294cbadc64a5a1 100644
--- a/atvc/examples/pytorch/reduce_sum/README.md
+++ b/atvc/examples/pytorch/reduce_sum/README.md
@@ -32,18 +32,6 @@ z = x + y
本样例支持如下产品型号:
- Atlas A2训练系列产品
-## 算子工程介绍
-其中,算子工程目录ReduceSum包含算子的实现文件,如下所示:
-```
-├── reduce_sum
-│ ├── reduce_sum_impl.h // 通过pytroch调用的方式调用ReduceSum算子
-│ ├── pytorch_ascendc_extension.cpp // pytorch调用入口
-│ ├── run_op.py // pytorch的测试用例
-│ └── run.sh // 脚本,编译需要的二进制文件,并测试
-```
-
-run.sh脚本编译pytorch所需要的算子依赖,并将执行测试用例。
-
## 编译运行样例算子
针对自定义算子工程,编译运行包含如下步骤:
- 完成算子pytorch入口和impl文件的实现;
@@ -53,10 +41,10 @@ run.sh脚本编译pytorch所需要的算子依赖,并将执行测试用例。
详细操作如下所示。
### 1. 获取源码包及环境配置
编译运行此样例前,请参考[准备:获取样例代码](../README.md#codeready)获取源码包及环境变量的准备。
-### 1. 安装pytorch环境
+### 2. 安装pytorch环境
参考[torch的安装](https://gitee.com/ascend/pytorch)进行安装torch、torch_npu环境
-### 2. 基于ATVC编写pytorch算子的实现
+### 3. 基于ATVC编写pytorch算子的实现
- 编写kernel侧函数,参考[reduce_sum_impl.h](./reduce_sum_impl.h)
```cpp
// 引入头文件
@@ -162,7 +150,7 @@ run.sh脚本编译pytorch所需要的算子依赖,并将执行测试用例。
run_tests()
```
-### 3. 基于ATVC编写pytorch算子的调用验证
+### 4. 基于ATVC编写pytorch算子的调用验证
- 调用脚本,生成pytorch算子,并运行测试用例
```bash
$ cd ./atvc/examples/pytorch/reduce_sum