diff --git a/c_api/README.md b/c_api/README.md new file mode 100644 index 000000000..58dd70658 --- /dev/null +++ b/c_api/README.md @@ -0,0 +1,39 @@ +# C API implementation + +This directory is the implementation of FastDeploy C SDK, which provides solutions for scenarios where users need C API. + +## How to compile and install + +When compiling the fastdeploy library, turn on the compile option --WITH_CAPI=ON to start compiling for this directory. The compiled C library and fastdeploy library are in the same dynamic library file, such as libfastdeploy.a under Linux. For example, to compile the CPU version of fastdeploy and compile the C API into the library, you can use the following command + +```bash +git clone https://github.com/PaddlePaddle/FastDeploy.git +cd FastDeploy +mkdir build && cd build +cmake .. -DENABLE_ORT_BACKEND=ON \ + -DENABLE_PADDLE_BACKEND=ON \ + -DENABLE_OPENVINO_BACKEND=ON \ + -DCMAKE_INSTALL_PREFIX=${PWD}/compiled_fastdeploy_sdk \ + -DENABLE_VISION=ON \ + -DWITH_CAPI=ON \ + -DOPENCV_DIRECTORY=/usr/lib/x86_64-linux-gnu/cmake/opencv4 \ + -DENABLE_TEXT=ON +make -j12 +make install +``` +For more details about compiling the fastdeploy library, please refer to the documentation +- [FastDeploy installation](../docs/en/build_and_install/README.md) + +## How to use + + The header files of the provided C API can be used. If it is manually compiled and installed, such as the above command, after `make install`, the header files will be in the directory ${PWD}/compiled_fastdeploy_sdk/include/fastdeploy_capi/. If it is a downloaded precompiled library, after decompression, include/fastdeploy_capi/ is the header file of the C API. For usage examples, please refer to the use cases under examples + +- [paddleclas](../examples/vision/classification/paddleclas/c/README.md) +- [paddledetection](../examples/vision/detection/paddledetection/c/README.md) +- [pp-ocrv2](../examples/vision/ocr/PP-OCRv2/c/README.md) +- [paddleseg](../examples/vision/segmentation/paddleseg/cpu-gpu/c/README.md) + +## Other documents + +- [How to Develop C API for a New Model](../docs/en/faq/develop_c_api_for_a_new_model.md) +- [Vision Results](../docs/api/vision_results/README.md) diff --git a/c_api/README_CN.md b/c_api/README_CN.md new file mode 100644 index 000000000..21429d0d6 --- /dev/null +++ b/c_api/README_CN.md @@ -0,0 +1,39 @@ +# C API实现 + +该目录下为FastDeploy C SDK的接口实现，为用户需要C API的场景提供解决方案。 + +## 如何编译安装 + +在编译fastdeploy库的时候，打开编译选项 --WITH_CAPI=ON，即可开启对于该目录的编译。编译后的C库和fastdeploy库在同一个动态库文件中，如linux下为libfastdeploy.a。比如，编译CPU版本的fastdeploy并且将C API一同编入库可以使用如下命令 + +```bash +git clone https://github.com/PaddlePaddle/FastDeploy.git +cd FastDeploy +mkdir build && cd build +cmake .. -DENABLE_ORT_BACKEND=ON \ + -DENABLE_PADDLE_BACKEND=ON \ + -DENABLE_OPENVINO_BACKEND=ON \ + -DCMAKE_INSTALL_PREFIX=${PWD}/compiled_fastdeploy_sdk \ + -DENABLE_VISION=ON \ + -DWITH_CAPI=ON \ + -DOPENCV_DIRECTORY=/usr/lib/x86_64-linux-gnu/cmake/opencv4 \ + -DENABLE_TEXT=ON +make -j12 +make install +``` +关于编译fastdeploy库的更多详情信息可以参考文档 +- [FastDeploy安装](../docs/cn/build_and_install/README.md) + +## 如何使用 + +因为提供的C API的头文件即可进行使用，如果是手动编译安装，比如上述命令，`make install`之后，头文件将会在目录${PWD}/compiled_fastdeploy_sdk/include/fastdeploy_capi/下。如果是下载的预编译好的库，解压缩之后include/fastdeploy_capi/下即是C API的头文件。关于使用示例可以参考examples下的用例 + +- [paddleclas](../examples/vision/classification/paddleclas/c/README_CN.md) +- [paddledetection](../examples/vision/detection/paddledetection/c/README_CN.md) +- [pp-ocrv2](../examples/vision/ocr/PP-OCRv2/c/README_CN.md) +- [paddleseg](../examples/vision/segmentation/paddleseg/cpu-gpu/c/README_CN.md) + +## 其它文档 + +- [如何给新模型增加C API](../docs/cn/faq/develop_c_api_for_a_new_model.md) +- [Vision Results](../docs/api/vision_results/README_CN.md) diff --git a/csharp/README.md b/csharp/README.md new file mode 100644 index 000000000..053f3df45 --- /dev/null +++ b/csharp/README.md @@ -0,0 +1,41 @@ +# C# API implementation + +This directory is the implementation of FastDeploy C# SDK, which provides solutions for scenarios where users need C# API. + +## How to compile and install + +When compiling the fastdeploy library, enable the compilation option --WITH_CSHARPAPI=ON to enable compilation for this directory. The compiled C# library is in the csharp_lib directory after installation. For example, to compile the CPU version of the fastdeploy library under Windows and add the C# API library, you can use the following command + +```shell +git clone https://github.com/PaddlePaddle/FastDeploy.git +cd FastDeploy +mkdir build && cd build +cmake .. -G "Visual Studio 16 2019" -A x64 ^ + -DENABLE_ORT_BACKEND=ON ^ + -DENABLE_PADDLE_BACKEND=ON ^ + -DENABLE_OPENVINO_BACKEND=ON ^ + -DENABLE_VISION=ON ^ + -DENABLE_TEXT=ON ^ + -DWITH_CAPI=ON ^ + -DWITH_CSHARPAPI=ON ^ + -DCMAKE_INSTALL_PREFIX="D:\Paddle\compiled_fastdeploy" +nuget restore #（please execute it when WITH_CSHARPAPI=ON to prepare dependencies in C#) +msbuild fastdeploy.sln /m /p:Configuration=Release /p:Platform=x64 +msbuild INSTALL.vcxproj /m /p:Configuration=Release /p:Platform=x64 +``` +For more details about compiling the fastdeploy library, please refer to the documentation +- [FastDeploy installation](../docs/cn/build_and_install/README.md) + +## how to use + +Use the compiled C# dynamic library file fastdeploy_csharp.dll as a reference to use the corresponding API. For usage examples, please refer to the use cases under examples + +- [paddleclas](../examples/vision/classification/paddleclas/csharp/README.md) +- [paddledetection](../examples/vision/detection/paddledetection/csharp/README.md) +- [pp-ocrv2](../examples/vision/ocr/PP-OCRv2/csharp/README.md) +- [paddleseg](../examples/vision/segmentation/paddleseg/cpu-gpu/csharp/README.md) + +## Other documents + +- [How to add C# API to a new model](../docs/en/faq/develop_c_sharp_api_for_a_new_model.md) +- [Vision Results](../docs/api/vision_results/README.md) diff --git a/csharp/README_CN.md b/csharp/README_CN.md new file mode 100644 index 000000000..3dc3ec553 --- /dev/null +++ b/csharp/README_CN.md @@ -0,0 +1,41 @@ +# C# API实现 + +该目录下为FastDeploy C# SDK的接口实现，为用户需要C# API的场景提供解决方案。 + +## 如何编译安装 + +在编译fastdeploy库的时候，打开编译选项 --WITH_CSHARPAPI=ON，即可开启对于该目录的编译。编译后的C#库安装后在csharp_lib目录下。例如在Windows下编译CPU版本的fastdeploy库并且加上C# API库可以使用如下命令 + +```shell +git clone https://github.com/PaddlePaddle/FastDeploy.git +cd FastDeploy +mkdir build && cd build +cmake .. -G "Visual Studio 16 2019" -A x64 ^ + -DENABLE_ORT_BACKEND=ON ^ + -DENABLE_PADDLE_BACKEND=ON ^ + -DENABLE_OPENVINO_BACKEND=ON ^ + -DENABLE_VISION=ON ^ + -DENABLE_TEXT=ON ^ + -DWITH_CAPI=ON ^ + -DWITH_CSHARPAPI=ON ^ + -DCMAKE_INSTALL_PREFIX="D:\Paddle\compiled_fastdeploy" +nuget restore #（please execute it when WITH_CSHARPAPI=ON to prepare dependencies in C#) +msbuild fastdeploy.sln /m /p:Configuration=Release /p:Platform=x64 +msbuild INSTALL.vcxproj /m /p:Configuration=Release /p:Platform=x64 +``` +关于编译fastdeploy库的更多详情信息可以参考文档 +- [FastDeploy安装](../docs/cn/build_and_install/README.md) + +## 如何使用 + +将编译出来的C#动态库文件fastdeploy_csharp.dll作为引用即可以使用相应的API。关于使用示例可以参考examples下的用例 + +- [paddleclas](../examples/vision/classification/paddleclas/csharp/README_CN.md) +- [paddledetection](../examples/vision/detection/paddledetection/csharp/README_CN.md) +- [pp-ocrv2](../examples/vision/ocr/PP-OCRv2/csharp/README_CN.md) +- [paddleseg](../examples/vision/segmentation/paddleseg/cpu-gpu/csharp/README_CN.md) + +## 其它文档 + +- [如何给新模型增加C# API](../docs/cn/faq/develop_c_sharp_api_for_a_new_model.md) +- [Vision Results](../docs/api/vision_results/README_CN.md)