zhujingxuan
5 years ago
4 changed files with 476 additions and 178 deletions
Split View
Diff Options
-
+174 -175mindspore/lite/micro/example/mnist/README.md
-
+125 -0mindspore/lite/micro/example/mnist/mnist.sh
-
+174 -0mindspore/lite/micro/example/mobilenetv2/README.md
-
+3 -3mindspore/lite/micro/example/mobilenetv2/mobilenetv2.sh
mindspore/lite/micro/example/lenet/README.md → mindspore/lite/micro/example/mnist/README.md
View File
| @@ -1,175 +1,174 @@ | |||
| # Arm Cortex-M编译部署 | |||
| `Linux` `Cortex-M` `IOT` `C/C++` `全流程` `模型编译` `模型代码生成` `模型部署` `推理应用` `初级` `中级` `高级` | |||
| <!-- TOC --> | |||
| - Arm Cortex-M编译部署 | |||
| - [STM32F746编译依赖](#STM32F746编译依赖) | |||
| - [STM32F746构建](#STM32F746构建) | |||
| - [STM32F746工程部署](#STM32F746工程部署) | |||
| - [更多详情](#更多详情) | |||
| - [Linux x86_64平台编译部署](#Linux x86_64平台编译部署) | |||
| - [Android平台编译部署](#STM32746平台编译部署) | |||
| <!-- /TOC --> | |||
| <a href="https://gitee.com/mindspore/docs/blob/master/tutorials/lite/source_zh_cn/quick_start/quick_start.md" target="_blank"><img src="../_static/logo_source.png"></a> | |||
| ## Arm Cortex-M编译部署 | |||
| 本教程以在STM32F746单板上编译部署生成模型代码为例，演示了codegen编译模型在Cortex-M平台的使用。更多关于Arm Cortex-M的详情可参见其[官网](https://developer.arm.com/ip-products/processors/cortex-m)。 | |||
| ### STM32F746编译依赖 | |||
| 模型推理代码的编译部署需要在windows上安装[Jlink]((https://www.segger.com/))、[STM32CubeMX](https://www.st.com/content/st_com/en.html)、[gcc-arm-none-ebai](https://developer.arm.com/tools-and-software/open-source-software/developer-tools/gnu-toolchain/gnu-rm)等工具来进行交叉编译。 | |||
| - [STM32CubeMX-Win](https://www.st.com/content/ccc/resource/technical/software/sw_development_suite/group0/0b/05/f0/25/c7/2b/42/9d/stm32cubemx_v6-1-1/files/stm32cubemx_v6-1-1.zip/jcr:content/translations/en.stm32cubemx_v6-1-1.zip) >= 6.0.1 | |||
| - [gcc-arm-none-eabi](https://developer.arm.com/tools-and-software/open-source-software/developer-tools/gnu-toolchain/gnu-rm/downloads) >= 9-2019-q4-major-win32 | |||
| - [JLink-windows](https://www.segger.com/downloads/jlink/) >= 6.56 | |||
| - [GCC](https://gcc.gnu.org/releases.html) >= 7.3.0 | |||
| - [CMake](https://cmake.org/download/) >= 3.18.3 | |||
| ### STM32F746构建与运行 | |||
| 首先使用codegen编译LeNet模型，生成对应的STM32F46推理代码。具体命令如下： | |||
| ```bash | |||
| ./codegen --codeMode=Inference --codePath=. --modelPath=LeNet.ms --moduleName=LeNet --target=ARM32M | |||
| ``` | |||
| #### 代码工程说明 | |||
| ```bash | |||
| ├── LeNet | |||
| └── operator_library | |||
| ``` | |||
| ##### 算子静态库目录说明 | |||
| 在编译此工程之前需要预先获取Cortex-M 平台对应的[算子库]()。 | |||
| 预置算子静态库的目录如下: | |||
| ```bash | |||
| ├── operator_library # 对应平台算子库目录 | |||
| ├── include # 对应平台算子库头文件目录 | |||
| └── lib # 对应平台算子库静态库目录 | |||
| ``` | |||
| 生成代码工程目录如下： | |||
| ```bash | |||
| ├── LeNet # 生成代码的根目录 | |||
| ├── benchmark # 生成代码的benchmark目录 | |||
| ├── include # 模型推理代码对外暴露头文件目录 | |||
| └── src # 模型推理代码目录 | |||
| ``` | |||
| #### 代码工程编译 | |||
| ##### 环境测试 | |||
| 安装好交叉编译所需环境后，需要在windows环境中依次将其加入到环境变量中 | |||
| ```bash | |||
| gcc -v # 查看GCC版本 | |||
| arm-none-eabi-gdb -v # 查看交叉编译环境 | |||
| jlink -v # 查看jlink版本 | |||
| make -v # 查看make版本 | |||
| ``` | |||
| 以上的命令均有成功返回值时，表明环境准备ok，可以继续进入下一步，否则先安装上述环境！！！ | |||
| ##### 生成STM32F746单板初始化代码([详情示例代码]()) | |||
| 1. 启动 STM32CubeMX，新建project，选择单板STM32F746IG | |||
| 2. 成功以后，选择`Makefile` ，`generator code` | |||
| 3. 在生成的工程目录下打开`cmd`，执行`make`，测试初始代码是否成功编译。 | |||
| ```bash | |||
| # make成功结果 | |||
| arm-none-eabi-size build/test_stm32f746.elf | |||
| text data bss dec hex filename | |||
| 3660 20 1572 5252 1484 build/test_stm32f746.elf | |||
| arm-none-eabi-objcopy -O ihex build/test_stm32f746.elf build/test_stm32f746.hex | |||
| arm-none-eabi-objcopy -O binary -S build/test_stm32f746.elf build/test_stm32f746.bin | |||
| ``` | |||
| ##### 编译生成模型静态库 | |||
| 1. 拷贝mindspore团队提供的cortex-m7的算子静态库以及对应头文件到STM32CubeMX生成的工程目录中。 | |||
| 2. 拷贝codegen生成模型推理代码到 STM32CubeMX生成的代码工程目录中 | |||
| ```bash | |||
| ├── .mxproject | |||
| └── build # 工程编译目录最终的elf文件存在于此 | |||
| └── Core | |||
| └── Drivers | |||
| └── LeNet # codegen生成的cortex-m7 模型推理代码 | |||
| └── Makefile # 组织工程makefile文件需要用户自己修改组织lenet && operator_library到工程目录中 | |||
| └── operator_library # mindspore团队提供的对应平台算子库 | |||
| └── startup_stm32f746xx.s | |||
| └── STM32F746IGKx_FLASH.ld | |||
| └── test_stm32f746.ioc | |||
| ``` | |||
| 3. 修改makefile文件，组织算子静态库以及模型推理代码 | |||
| ```bash | |||
| # C includes | |||
| C_INCLUDES = \ | |||
| -ICore/Inc \ | |||
| -IDrivers/STM32F7xx_HAL_Driver/Inc \ | |||
| -IDrivers/STM32F7xx_HAL_Driver/Inc/Legacy \ | |||
| -IDrivers/CMSIS/Device/ST/STM32F7xx/Include \ | |||
| -Ioperator_library/include \ # 新增，指定算子库头文件目录 | |||
| -ILeNet/include \ # 新增，指定模型推理代码头文件 | |||
| -ILeNet/src # 新增，指定模型推理代码头文件 | |||
| # libraries | |||
| LIBS = -lc -lm -lnosys -lops # 修改，导入mindspore团队提供算子库 | |||
| LIBDIR = -Ioperator_library/lib/arm32m # 新增，指定算子库所在路径 | |||
| ``` | |||
| 4. 在工程目录的Core/Src的main.c编写模型调用代码，具体代码新增如下： | |||
| ```cpp | |||
| ``` | |||
| 5. 在工程跟目中目录使用管理员权限打开`cmd` 执行 `make`进行编译 | |||
| ```bash | |||
| make | |||
| ``` | |||
| ### STM32F746工程部署 | |||
| 使用jlink 将可执行文件拷贝到单板上并做推理 | |||
| ```bash | |||
| jlinkgdbserver # 启动jlinkgdbserver 选定target device为STM32F746IG | |||
| jlinkRTTViewer # 启动jlinkRTTViewer 选定target devices为STM32F746IG | |||
| arm-none-eabi-gdb # 启动arm-gcc gdb服务 | |||
| file build/target.elf # 打开调测文件 | |||
| target remote 127.0.0.1 # 连接jlink服务器 | |||
| monitor reset # 重置单板 | |||
| monitor halt # 挂起单板 | |||
| load # 加载可执行文件到单板 | |||
| c # 执行模型推理 | |||
| ``` | |||
| #### 执行结果 | |||
| ```bash | |||
| ``` | |||
| ## 更多详情 | |||
| ### [Linux x86_64平台编译部署]() | |||
| ### [Android平台编译部署]() | |||
| # Arm Cortex-M编译部署 | |||
| `Linux` `Cortex-M` `IOT` `C/C++` `全流程` `模型编译` `模型代码生成` `模型部署` `推理应用` `初级` `中级` `高级` | |||
| <!-- TOC --> | |||
| - Arm Cortex-M编译部署 | |||
| - [STM32F746编译依赖](#STM32F746编译依赖) | |||
| - [STM32F746构建](#STM32F746构建) | |||
| - [STM32F746工程部署](#STM32F746工程部署) | |||
| - [更多详情](#更多详情) | |||
| - [Linux x86_64平台编译部署](#Linux x86_64平台编译部署) | |||
| - [Android平台编译部署](#STM32746平台编译部署) | |||
| <!-- /TOC --> | |||
| ## Arm Cortex-M编译部署 | |||
| 本教程以在STM32F746单板上编译部署生成模型代码为例，演示了codegen编译模型在Cortex-M平台的使用。更多关于Arm Cortex-M的详情可参见其[官网](https://developer.arm.com/ip-products/processors/cortex-m)。 | |||
| ### STM32F746编译依赖 | |||
| 模型推理代码的编译部署需要在windows上安装[Jlink]((https://www.segger.com/))、[STM32CubeMX](https://www.st.com/content/st_com/en.html)、[gcc-arm-none-ebai](https://developer.arm.com/tools-and-software/open-source-software/developer-tools/gnu-toolchain/gnu-rm)等工具来进行交叉编译。 | |||
| - [STM32CubeMX-Win](https://www.st.com/content/ccc/resource/technical/software/sw_development_suite/group0/0b/05/f0/25/c7/2b/42/9d/stm32cubemx_v6-1-1/files/stm32cubemx_v6-1-1.zip/jcr:content/translations/en.stm32cubemx_v6-1-1.zip) >= 6.0.1 | |||
| - [gcc-arm-none-eabi](https://developer.arm.com/tools-and-software/open-source-software/developer-tools/gnu-toolchain/gnu-rm/downloads) >= 9-2019-q4-major-win32 | |||
| - [JLink-windows](https://www.segger.com/downloads/jlink/) >= 6.56 | |||
| - [GCC](https://gcc.gnu.org/releases.html) >= 7.3.0 | |||
| - [CMake](https://cmake.org/download/) >= 3.18.3 | |||
| ### STM32F746构建与运行 | |||
| 首先使用codegen编译LeNet模型，生成对应的STM32F46推理代码。具体命令如下： | |||
| ```bash | |||
| ./codegen --codePath=. --modelPath=LeNet.ms --moduleName=LeNet --target=ARM32M | |||
| ``` | |||
| #### 代码工程说明 | |||
| ```bash | |||
| ├── LeNet | |||
| └── operator_library | |||
| ``` | |||
| ##### 算子静态库目录说明 | |||
| 在编译此工程之前需要预先获取Cortex-M 平台对应的[算子库]()。 | |||
| 预置算子静态库的目录如下: | |||
| ```bash | |||
| ├── operator_library # 对应平台算子库目录 | |||
| ├── include # 对应平台算子库头文件目录 | |||
| └── lib # 对应平台算子库静态库目录 | |||
| ``` | |||
| 生成代码工程目录如下： | |||
| ```bash | |||
| ├── LeNet # 生成代码的根目录 | |||
| ├── benchmark # 生成代码的benchmark目录 | |||
| ├── include # 模型推理代码对外暴露头文件目录 | |||
| └── src # 模型推理代码目录 | |||
| ``` | |||
| #### 代码工程编译 | |||
| ##### 环境测试 | |||
| 安装好交叉编译所需环境后，需要在windows环境中依次将其加入到环境变量中 | |||
| ```bash | |||
| gcc -v # 查看GCC版本 | |||
| arm-none-eabi-gdb -v # 查看交叉编译环境 | |||
| jlink -v # 查看jlink版本 | |||
| make -v # 查看make版本 | |||
| ``` | |||
| 以上的命令均有成功返回值时，表明环境准备ok，可以继续进入下一步，否则先安装上述环境！！！ | |||
| ##### 生成STM32F746单板初始化代码([详情示例代码]()) | |||
| 1. 启动 STM32CubeMX，新建project，选择单板STM32F746IG | |||
| 2. 成功以后，选择`Makefile` ，`generator code` | |||
| 3. 在生成的工程目录下打开`cmd`，执行`make`，测试初始代码是否成功编译。 | |||
| ```bash | |||
| # make成功结果 | |||
| arm-none-eabi-size build/test_stm32f746.elf | |||
| text data bss dec hex filename | |||
| 3660 20 1572 5252 1484 build/test_stm32f746.elf | |||
| arm-none-eabi-objcopy -O ihex build/test_stm32f746.elf build/test_stm32f746.hex | |||
| arm-none-eabi-objcopy -O binary -S build/test_stm32f746.elf build/test_stm32f746.bin | |||
| ``` | |||
| ##### 编译生成模型静态库 | |||
| 1. 拷贝mindspore团队提供的cortex-m7的算子静态库以及对应头文件到STM32CubeMX生成的工程目录中。 | |||
| 2. 拷贝codegen生成模型推理代码到 STM32CubeMX生成的代码工程目录中 | |||
| ```bash | |||
| ├── .mxproject | |||
| └── build # 工程编译目录最终的elf文件存在于此 | |||
| └── Core | |||
| └── Drivers | |||
| └── LeNet # codegen生成的cortex-m7 模型推理代码 | |||
| └── Makefile # 组织工程makefile文件需要用户自己修改组织lenet && operator_library到工程目录中 | |||
| └── operator_library # mindspore团队提供的对应平台算子库 | |||
| └── startup_stm32f746xx.s | |||
| └── STM32F746IGKx_FLASH.ld | |||
| └── test_stm32f746.ioc | |||
| ``` | |||
| 3. 修改makefile文件，组织算子静态库以及模型推理代码 | |||
| ```bash | |||
| # C includes | |||
| C_INCLUDES = \ | |||
| -ICore/Inc \ | |||
| -IDrivers/STM32F7xx_HAL_Driver/Inc \ | |||
| -IDrivers/STM32F7xx_HAL_Driver/Inc/Legacy \ | |||
| -IDrivers/CMSIS/Device/ST/STM32F7xx/Include \ | |||
| -Ioperator_library/include \ # 新增，指定算子库头文件目录 | |||
| -ILeNet/include \ # 新增，指定模型推理代码头文件 | |||
| -ILeNet/src # 新增，指定模型推理代码头文件 | |||
| # libraries | |||
| LIBS = -lc -lm -lnosys -lops # 修改，导入mindspore团队提供算子库 | |||
| LIBDIR = -Ioperator_library/lib/arm32m # 新增，指定算子库所在路径 | |||
| ``` | |||
| 4. 在工程目录的Core/Src的main.c编写模型调用代码，具体代码新增如下： | |||
| ```cpp | |||
| ``` | |||
| 5. 在工程跟目中目录使用管理员权限打开`cmd` 执行 `make`进行编译 | |||
| ```bash | |||
| make | |||
| ``` | |||
| ### STM32F746工程部署 | |||
| 使用jlink 将可执行文件拷贝到单板上并做推理 | |||
| ```bash | |||
| jlinkgdbserver # 启动jlinkgdbserver 选定target device为STM32F746IG | |||
| jlinkRTTViewer # 启动jlinkRTTViewer 选定target devices为STM32F746IG | |||
| arm-none-eabi-gdb # 启动arm-gcc gdb服务 | |||
| file build/target.elf # 打开调测文件 | |||
| target remote 127.0.0.1 # 连接jlink服务器 | |||
| monitor reset # 重置单板 | |||
| monitor halt # 挂起单板 | |||
| load # 加载可执行文件到单板 | |||
| c # 执行模型推理 | |||
| ``` | |||
| #### 执行结果 | |||
| ```bash | |||
| ``` | |||
| ## 更多详情 | |||
| ### [Linux x86_64平台编译部署]() | |||
| ### [Android平台编译部署]() | |||
+ 125
- 0
mindspore/lite/micro/example/mnist/mnist.sh
View File
| @@ -0,0 +1,125 @@ | |||
| #!/bin/bash | |||
| # Copyright 2021 Huawei Technologies Co., Ltd | |||
| # | |||
| # Licensed under the Apache License, Version 2.0 (the "License"); | |||
| # you may not use this file except in compliance with the License. | |||
| # You may obtain a copy of the License at | |||
| # | |||
| # http://www.apache.org/licenses/LICENSE-2.0 | |||
| # | |||
| # Unless required by applicable law or agreed to in writing, software | |||
| # distributed under the License is distributed on an "AS IS" BASIS, | |||
| # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. | |||
| # See the License for the specific language governing permissions and | |||
| # limitations under the License. | |||
| # ============================================================================ | |||
| set -e | |||
| CURRENT_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" &> /dev/null && pwd )" | |||
| MINDSPORE_ROOT_DIR=${${CURRENT_DIR}%%/mindspore/lite/micro/example/mnist} | |||
| OUTPUT_DIR=${1:-${MINDSPORE_ROOT_DIR}/output} | |||
| THREAD_NUM=${2:-32} | |||
| MODULE_NAME=mnist | |||
| OUTPUT_IR=Reshape-64.ir | |||
| CALIB_OUT=${CURRENT_DIR}/Reshape-64.out | |||
| echo "current dir is: ${CURRENT_DIR}" | |||
| echo "packed output dir is :${OUTPUT_DIR}" | |||
| if [ ! -d "${OUTPUT_DIR}" ]; then | |||
| echo "folder ${OUTPUT_DIR} does not exist" | |||
| return 1 | |||
| fi | |||
| # rm if already exist | |||
| WORKSPACE=${CURRENT_DIR}/build | |||
| rm -rf ${WORKSPACE} | |||
| mkdir ${WORKSPACE} || exit 1 | |||
| PROJECT_DIR=${WORKSPACE}/${MODULE_NAME} | |||
| compare_output() { | |||
| local OUTPUT_FILE=$1 | |||
| local CALIB_FILE=$2 | |||
| if [[ ! -f "${OUTPUT_FILE}" || ! -f "${CALIB_FILE}" ]]; then | |||
| echo "file ${OUTPUT_FILE}, ${CALIB_FILE} does not exist, pwd $(pwd)" | |||
| exit 1 | |||
| fi | |||
| lines=$(cat ${CALIB_FILE} | wc -l) | |||
| for ((i = 1; i <= $lines; i++)); do | |||
| line1=$(awk 'NR=="'${i}'"{print $0}' ${CALIB_FILE}) | |||
| line2=$(awk 'NR=="'${i}'"{print $0}' ${OUTPUT_FILE}) | |||
| if [[ "${line1}" != "${line2}" ]]; then | |||
| echo -e "file ${OUTPUT_FILE}, ${CALIB_FILE}, compare failed! line: ${i}" | |||
| exit 1 | |||
| fi | |||
| done | |||
| echo -e "compare success, ${OUTPUT_FILE}, ${CALIB_FILE}" | |||
| } | |||
| # cp oplib and codegen | |||
| cp ${OUTPUT_DIR}/mindspore-lite-*-codegen-linux-x64.tar.gz ${WORKSPACE}/ || exit 1 | |||
| cd ${WORKSPACE} || exit 1 | |||
| tar -zxf mindspore-lite-*-codegen-linux-x64.tar.gz || exit 1 | |||
| cd mindspore-lite-*-codegen-linux-x64 || exit 1 | |||
| mv operator_library/ ${WORKSPACE}/ || exit 1 | |||
| mv codegen ${WORKSPACE}/ || exit 1 | |||
| cd - | |||
| rm -r mindspore-lite-*-codegen-linux-x64 || exit 1 | |||
| rm mindspore-lite-*-codegen-linux-x64.tar.gz || exit 1 | |||
| # convert model | |||
| cp ${OUTPUT_DIR}/mindspore-lite-*-converter-linux-x64.tar.gz ${WORKSPACE}/ || exit 1 | |||
| cd ${WORKSPACE} || exit 1 | |||
| tar -zxf mindspore-lite-*-converter-linux-x64.tar.gz || exit 1 | |||
| rm mindspore-lite-*-converter-linux-x64.tar.gz || exit 1 | |||
| cd mindspore-lite-*-converter-linux-x64 || exit 1 | |||
| export LD_LIBRARY_PATH=./lib/:./third_party/protobuf/lib:./third_party/flatbuffers/lib:./third_party/glog/lib | |||
| converter/converter_lite --fmk=TFLITE \ | |||
| --modelFile=${CURRENT_DIR}/mnist.tflite \ | |||
| --outputFile=${WORKSPACE}/mnist | |||
| cd - | |||
| rm -rf mindspore-lite-*-converter-linux-x64 || exit 1 | |||
| # generate code | |||
| ${WORKSPACE}/codegen --modelPath=${WORKSPACE}/mnist.ms \ | |||
| --moduleName=${MODULE_NAME} \ | |||
| --isWeightFile=true \ | |||
| --debugMode=true | |||
| rm codegen | |||
| if [ ! -d "${PROJECT_DIR}" ]; then | |||
| echo "folder ${PROJECT_DIR} does not exist" | |||
| return 1 | |||
| fi | |||
| cd ${PROJECT_DIR} || exit 1 | |||
| # 1. build static lib.a | |||
| echo -e "building static library" | |||
| mkdir -p src/build && cd src/build || exit 1 | |||
| OP_HEADER_PATH=${WORKSPACE}/operator_library/include | |||
| OP_LIB=${WORKSPACE}/operator_library/lib/x86/libops.a | |||
| echo "Head Path: ${OP_HEADER_PATH}" | |||
| echo "Lib Path: ${OP_LIB}" | |||
| cmake -DCMAKE_BUILD_TYPE=Debug \ | |||
| -DOP_LIB=${OP_LIB} \ | |||
| -DOP_HEADER_PATH=${OP_HEADER_PATH} .. | |||
| make -j${THREAD_NUM} | |||
| # 2. build benchmark | |||
| cd ${PROJECT_DIR}/benchmark && mkdir -p build && cd build || exit 1 | |||
| cmake -DMODEL_LIB="${PROJECT_DIR}/src/build/libnet.a" .. | |||
| make -j${THREAD_NUM} | |||
| echo "net file: ${PROJECT_DIR}/src/${MODULE_NAME}.net" | |||
| # 3. run benchmark | |||
| ./benchmark ${CURRENT_DIR}/input_1_224_224_3_uint8.bin ${PROJECT_DIR}/src/${MODULE_NAME}.net | |||
| compare_output ${OUTPUT_IR} ${CALIB_OUT} | |||
| RET=$? | |||
| if [[ "${RET}" -eq 0 ]]; then | |||
| echo -e "run benchmark success: ${MODULE_NAME}" | |||
| else | |||
| echo -e "run benchmark failed: ${MODULE_NAME}" | |||
| exit 1 | |||
| fi | |||
+ 174
- 0
mindspore/lite/micro/example/mobilenetv2/README.md
View File
| @@ -0,0 +1,174 @@ | |||
| # Android编译部署 | |||
| `Linux` `Cortex-M` `IOT` `C/C++` `全流程` `模型编译` `模型代码生成` `模型部署` `推理应用` `初级` `中级` `高级` | |||
| <!-- TOC --> | |||
| - Arm Cortex-M编译部署 | |||
| - [STM32F746编译依赖](#STM32F746编译依赖) | |||
| - [STM32F746构建](#STM32F746构建) | |||
| - [STM32F746工程部署](#STM32F746工程部署) | |||
| - [更多详情](#更多详情) | |||
| - [Linux x86_64平台编译部署](#Linux x86_64平台编译部署) | |||
| - [Android平台编译部署](#STM32746平台编译部署) | |||
| <!-- /TOC --> | |||
| ## Arm Cortex-M编译部署 | |||
| 本教程以在STM32F746单板上编译部署生成模型代码为例，演示了codegen编译模型在Cortex-M平台的使用。更多关于Arm Cortex-M的详情可参见其[官网](https://developer.arm.com/ip-products/processors/cortex-m)。 | |||
| ### STM32F746编译依赖 | |||
| 模型推理代码的编译部署需要在windows上安装[Jlink]((https://www.segger.com/))、[STM32CubeMX](https://www.st.com/content/st_com/en.html)、[gcc-arm-none-ebai](https://developer.arm.com/tools-and-software/open-source-software/developer-tools/gnu-toolchain/gnu-rm)等工具来进行交叉编译。 | |||
| - [STM32CubeMX-Win](https://www.st.com/content/ccc/resource/technical/software/sw_development_suite/group0/0b/05/f0/25/c7/2b/42/9d/stm32cubemx_v6-1-1/files/stm32cubemx_v6-1-1.zip/jcr:content/translations/en.stm32cubemx_v6-1-1.zip) >= 6.0.1 | |||
| - [gcc-arm-none-eabi](https://developer.arm.com/tools-and-software/open-source-software/developer-tools/gnu-toolchain/gnu-rm/downloads) >= 9-2019-q4-major-win32 | |||
| - [JLink-windows](https://www.segger.com/downloads/jlink/) >= 6.56 | |||
| - [GCC](https://gcc.gnu.org/releases.html) >= 7.3.0 | |||
| - [CMake](https://cmake.org/download/) >= 3.18.3 | |||
| ### STM32F746构建与运行 | |||
| 首先使用codegen编译LeNet模型，生成对应的STM32F46推理代码。具体命令如下： | |||
| ```bash | |||
| ./codegen --codePath=. --modelPath=LeNet.ms --moduleName=LeNet --target=ARM32M | |||
| ``` | |||
| #### 代码工程说明 | |||
| ```bash | |||
| ├── LeNet | |||
| └── operator_library | |||
| ``` | |||
| ##### 算子静态库目录说明 | |||
| 在编译此工程之前需要预先获取Cortex-M 平台对应的[算子库]()。 | |||
| 预置算子静态库的目录如下: | |||
| ```bash | |||
| ├── operator_library # 对应平台算子库目录 | |||
| ├── include # 对应平台算子库头文件目录 | |||
| └── lib # 对应平台算子库静态库目录 | |||
| ``` | |||
| 生成代码工程目录如下： | |||
| ```bash | |||
| ├── LeNet # 生成代码的根目录 | |||
| ├── benchmark # 生成代码的benchmark目录 | |||
| ├── include # 模型推理代码对外暴露头文件目录 | |||
| └── src # 模型推理代码目录 | |||
| ``` | |||
| #### 代码工程编译 | |||
| ##### 环境测试 | |||
| 安装好交叉编译所需环境后，需要在windows环境中依次将其加入到环境变量中 | |||
| ```bash | |||
| gcc -v # 查看GCC版本 | |||
| arm-none-eabi-gdb -v # 查看交叉编译环境 | |||
| jlink -v # 查看jlink版本 | |||
| make -v # 查看make版本 | |||
| ``` | |||
| 以上的命令均有成功返回值时，表明环境准备ok，可以继续进入下一步，否则先安装上述环境！！！ | |||
| ##### 生成STM32F746单板初始化代码([详情示例代码]()) | |||
| 1. 启动 STM32CubeMX，新建project，选择单板STM32F746IG | |||
| 2. 成功以后，选择`Makefile` ，`generator code` | |||
| 3. 在生成的工程目录下打开`cmd`，执行`make`，测试初始代码是否成功编译。 | |||
| ```bash | |||
| # make成功结果 | |||
| arm-none-eabi-size build/test_stm32f746.elf | |||
| text data bss dec hex filename | |||
| 3660 20 1572 5252 1484 build/test_stm32f746.elf | |||
| arm-none-eabi-objcopy -O ihex build/test_stm32f746.elf build/test_stm32f746.hex | |||
| arm-none-eabi-objcopy -O binary -S build/test_stm32f746.elf build/test_stm32f746.bin | |||
| ``` | |||
| ##### 编译生成模型静态库 | |||
| 1. 拷贝mindspore团队提供的cortex-m7的算子静态库以及对应头文件到STM32CubeMX生成的工程目录中。 | |||
| 2. 拷贝codegen生成模型推理代码到 STM32CubeMX生成的代码工程目录中 | |||
| ```bash | |||
| ├── .mxproject | |||
| └── build # 工程编译目录最终的elf文件存在于此 | |||
| └── Core | |||
| └── Drivers | |||
| └── LeNet # codegen生成的cortex-m7 模型推理代码 | |||
| └── Makefile # 组织工程makefile文件需要用户自己修改组织lenet && operator_library到工程目录中 | |||
| └── operator_library # mindspore团队提供的对应平台算子库 | |||
| └── startup_stm32f746xx.s | |||
| └── STM32F746IGKx_FLASH.ld | |||
| └── test_stm32f746.ioc | |||
| ``` | |||
| 3. 修改makefile文件，组织算子静态库以及模型推理代码 | |||
| ```bash | |||
| # C includes | |||
| C_INCLUDES = \ | |||
| -ICore/Inc \ | |||
| -IDrivers/STM32F7xx_HAL_Driver/Inc \ | |||
| -IDrivers/STM32F7xx_HAL_Driver/Inc/Legacy \ | |||
| -IDrivers/CMSIS/Device/ST/STM32F7xx/Include \ | |||
| -Ioperator_library/include \ # 新增，指定算子库头文件目录 | |||
| -ILeNet/include \ # 新增，指定模型推理代码头文件 | |||
| -ILeNet/src # 新增，指定模型推理代码头文件 | |||
| # libraries | |||
| LIBS = -lc -lm -lnosys -lops # 修改，导入mindspore团队提供算子库 | |||
| LIBDIR = -Ioperator_library/lib/arm32m # 新增，指定算子库所在路径 | |||
| ``` | |||
| 4. 在工程目录的Core/Src的main.c编写模型调用代码，具体代码新增如下： | |||
| ```cpp | |||
| ``` | |||
| 5. 在工程跟目中目录使用管理员权限打开`cmd` 执行 `make`进行编译 | |||
| ```bash | |||
| make | |||
| ``` | |||
| ### STM32F746工程部署 | |||
| 使用jlink 将可执行文件拷贝到单板上并做推理 | |||
| ```bash | |||
| jlinkgdbserver # 启动jlinkgdbserver 选定target device为STM32F746IG | |||
| jlinkRTTViewer # 启动jlinkRTTViewer 选定target devices为STM32F746IG | |||
| arm-none-eabi-gdb # 启动arm-gcc gdb服务 | |||
| file build/target.elf # 打开调测文件 | |||
| target remote 127.0.0.1 # 连接jlink服务器 | |||
| monitor reset # 重置单板 | |||
| monitor halt # 挂起单板 | |||
| load # 加载可执行文件到单板 | |||
| c # 执行模型推理 | |||
| ``` | |||
| #### 执行结果 | |||
| ```bash | |||
| ``` | |||
| ## 更多详情 | |||
| ### [Linux x86_64平台编译部署]() | |||
| ### [Android平台编译部署]() | |||
mindspore/lite/micro/example/mobilenetv2_quant/run.sh → mindspore/lite/micro/example/mobilenetv2/mobilenetv2.sh
View File
| @@ -1,5 +1,5 @@ | |||
| #!/bin/bash | |||
| # Copyright 2019 Huawei Technologies Co., Ltd | |||
| # Copyright 2021 Huawei Technologies Co., Ltd | |||
| # | |||
| # Licensed under the Apache License, Version 2.0 (the "License"); | |||
| # you may not use this file except in compliance with the License. | |||
| @@ -16,11 +16,11 @@ | |||
| set -e | |||
| CURRENT_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" &> /dev/null && pwd )" | |||
| MINDSPORE_ROOT_DIR=${${CURRENT_DIR}%%/mindspore/lite/micro/example/mobilenetv2_quant} | |||
| MINDSPORE_ROOT_DIR=${${CURRENT_DIR}%%/mindspore/lite/micro/example/mobilenetv2} | |||
| OUTPUT_DIR=${1:-${MINDSPORE_ROOT_DIR}/output} | |||
| THREAD_NUM=${2:-32} | |||
| MODULE_NAME=mobilenetv2_quant | |||
| MODULE_NAME=mobilenetv2 | |||
| OUTPUT_IR=Reshape-64.ir | |||
| CALIB_OUT=${CURRENT_DIR}/Reshape-64.out | |||