apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-06 09:07:10 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[Tutorial Doc] add tutorial for how to use c and c sharp api (#1392)

Browse Source 
* add tutorialfor how to use c and c sharp api

* update english doc

* update doc

* update doc

* update doc

* update doc

* update doc
This commit is contained in:
chenjian
2023-02-28 21:29:44 +08:00
committed by GitHub
parent c8bcada1a2
commit 4ee783b1ea
12 changed files with 36 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
c_api/README.md
View File

@@ -1,4 +1,4 @@
# C API implementation # C API tutorial
This directory is the implementation of FastDeploy C SDK, which provides solutions for scenarios where users need C API. This directory is the implementation of FastDeploy C SDK, which provides solutions for scenarios where users need C API.

2
c_api/README_CN.md
View File

@@ -1,4 +1,4 @@
# C API实现 # C API指南
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。 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。

2
csharp/README.md
View File

@@ -1,4 +1,4 @@
# C# API implementation # C# API tutorial
This directory is the implementation of FastDeploy C# SDK, which provides solutions for scenarios where users need C# API. This directory is the implementation of FastDeploy C# SDK, which provides solutions for scenarios where users need C# API.

2
csharp/README_CN.md
View File

@@ -1,4 +1,4 @@
# C# API实现 # C# API指南
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。 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。

6
docs/cn/build_and_install/cpu.md
View File

@@ -12,8 +12,8 @@
| ENABLE_OPENVINO_BACKEND | Linux(x64)/Windows(x64)/Mac OSX(x86) | 默认OFF，是否编译集成OpenVINO后端 | | ENABLE_OPENVINO_BACKEND | Linux(x64)/Windows(x64)/Mac OSX(x86) | 默认OFF，是否编译集成OpenVINO后端 |
| ENABLE_VISION | Linux(x64)/Windows(x64)/Mac OSX(x86) | 默认OFF，是否编译集成视觉模型的部署模块 | | ENABLE_VISION | Linux(x64)/Windows(x64)/Mac OSX(x86) | 默认OFF，是否编译集成视觉模型的部署模块 |
| ENABLE_TEXT | Linux(x64)/Windows(x64)/Mac OSX(x86) | 默认OFF，是否编译集成文本NLP模型的部署模块 | | ENABLE_TEXT | Linux(x64)/Windows(x64)/Mac OSX(x86) | 默认OFF，是否编译集成文本NLP模型的部署模块 |
| ENABLE_CAPI | Linux(x64)/Windows(x64)/Mac OSX(x86) | 默认OFF，是否编译集成C API | | WITH_CAPI | Linux(x64)/Windows(x64)/Mac OSX(x86) | 默认OFF，是否编译集成C API |
| ENABLE_CSHARPAPI | Windows(x64) | 默认OFF，是否编译集成C# API | | WITH_CSHARPAPI | Windows(x64) | 默认OFF，是否编译集成C# API |
第三方库依赖指定（不设定如下参数，会自动下载预编译库） 第三方库依赖指定（不设定如下参数，会自动下载预编译库）
| 选项 | 说明 | | 选项 | 说明 |
@@ -70,7 +70,7 @@ cmake .. -G "Visual Studio 16 2019" -A x64 ^
-DENABLE_VISION=ON ^ -DENABLE_VISION=ON ^
-DENABLE_TEXT=ON ^ -DENABLE_TEXT=ON ^
-DCMAKE_INSTALL_PREFIX="D:\Paddle\compiled_fastdeploy" -DCMAKE_INSTALL_PREFIX="D:\Paddle\compiled_fastdeploy"
% nuget restore （please execute it when ENABLE_CSHARPAPI to prepare dependencies in C#) % 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 fastdeploy.sln /m /p:Configuration=Release /p:Platform=x64
msbuild INSTALL.vcxproj /m /p:Configuration=Release /p:Platform=x64 msbuild INSTALL.vcxproj /m /p:Configuration=Release /p:Platform=x64
``` ```

6
docs/cn/build_and_install/gpu.md
View File

@@ -16,8 +16,8 @@
| ENABLE_TEXT | Linux(x64)/Windows(x64) | 默认OFF，是否编译集成文本NLP模型的部署模块 | | ENABLE_TEXT | Linux(x64)/Windows(x64) | 默认OFF，是否编译集成文本NLP模型的部署模块 |
| CUDA_DIRECTORY | Linux(x64)/Windows(x64) | 默认/usr/local/cuda，要求CUDA>=11.2 | | CUDA_DIRECTORY | Linux(x64)/Windows(x64) | 默认/usr/local/cuda，要求CUDA>=11.2 |
| TRT_DIRECTORY | Linux(x64)/Windows(x64) | 默认为空，要求TensorRT>=8.4， 指定路径如/Download/TensorRT-8.5 | | TRT_DIRECTORY | Linux(x64)/Windows(x64) | 默认为空，要求TensorRT>=8.4， 指定路径如/Download/TensorRT-8.5 |
| ENABLE_CAPI | Linux(x64)/Windows(x64)/Mac OSX(x86) | 默认OFF，是否编译集成C API | | WITH_CAPI | Linux(x64)/Windows(x64)/Mac OSX(x86) | 默认OFF，是否编译集成C API |
| ENABLE_CSHARPAPI | Windows(x64) | 默认OFF，是否编译集成C# API | | WITH_CSHARPAPI | Windows(x64) | 默认OFF，是否编译集成C# API |
第三方库依赖指定（不设定如下参数，会自动下载预编译库） 第三方库依赖指定（不设定如下参数，会自动下载预编译库）
| 选项 | 说明 | | 选项 | 说明 |
@@ -88,7 +88,7 @@ cmake .. -G "Visual Studio 16 2019" -A x64 ^
-DTRT_DIRECTORY="D:\Paddle\TensorRT-8.4.1.5" ^ -DTRT_DIRECTORY="D:\Paddle\TensorRT-8.4.1.5" ^
-DCUDA_DIRECTORY="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.2" ^ -DCUDA_DIRECTORY="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.2" ^
-DCMAKE_INSTALL_PREFIX="D:\Paddle\compiled_fastdeploy" -DCMAKE_INSTALL_PREFIX="D:\Paddle\compiled_fastdeploy"
% nuget restore （please execute it when ENABLE_CSHARPAPI to prepare dependencies in C#) % 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 fastdeploy.sln /m /p:Configuration=Release /p:Platform=x64
msbuild INSTALL.vcxproj /m /p:Configuration=Release /p:Platform=x64 msbuild INSTALL.vcxproj /m /p:Configuration=Release /p:Platform=x64
``` ```

6
docs/en/build_and_install/cpu.md
View File

@@ -13,8 +13,8 @@ Please do not modify other cmake paramters exclude the following options.
| ENABLE_OPENVINO_BACKEND | Linux(x64)/Windows(x64)/Mac OSX(x86) | Default OFF, whether to intergrate OpenVINO backend | | ENABLE_OPENVINO_BACKEND | Linux(x64)/Windows(x64)/Mac OSX(x86) | Default OFF, whether to intergrate OpenVINO backend |
| ENABLE_VISION | Linux(x64/aarch64)/Windows(x64)/Mac OSX(arm64/x86) | Default OFF, whether to intergrate vision models | | ENABLE_VISION | Linux(x64/aarch64)/Windows(x64)/Mac OSX(arm64/x86) | Default OFF, whether to intergrate vision models |
| ENABLE_TEXT | Linux(x64/aarch64)/Windows(x64)/Mac OSX(arm64/x86) | Default OFF, whether to intergrate text models | | ENABLE_TEXT | Linux(x64/aarch64)/Windows(x64)/Mac OSX(arm64/x86) | Default OFF, whether to intergrate text models |
| ENABLE_CAPI | Linux(x64)/Windows(x64)/Mac OSX(x86) | Default OFF, whether to intergrate C API | | WITH_CAPI | Linux(x64)/Windows(x64)/Mac OSX(x86) | Default OFF, whether to intergrate C API |
| ENABLE_CSHARPAPI | Windows(x64) | Default OFF, whether to intergrate C# API | | WITH_CSHARPAPI | Windows(x64) | Default OFF, whether to intergrate C# API |
The configuration for third libraries(Optional, if the following option is not defined, the prebuilt third libraries will download automaticly while building FastDeploy). The configuration for third libraries(Optional, if the following option is not defined, the prebuilt third libraries will download automaticly while building FastDeploy).
| Option | Description | | Option | Description |
@@ -70,7 +70,7 @@ cmake .. -G "Visual Studio 16 2019" -A x64 \
-DENABLE_OPENVINO_BACKEND=ON \ -DENABLE_OPENVINO_BACKEND=ON \
-DENABLE_VISION=ON \ -DENABLE_VISION=ON \
-DCMAKE_INSTALL_PREFIX="D:\Paddle\compiled_fastdeploy" -DCMAKE_INSTALL_PREFIX="D:\Paddle\compiled_fastdeploy"
% nuget restore （please execute it when ENABLE_CSHARPAPI to prepare dependencies in C#) % 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 fastdeploy.sln /m /p:Configuration=Release /p:Platform=x64
msbuild INSTALL.vcxproj /m /p:Configuration=Release /p:Platform=x64 msbuild INSTALL.vcxproj /m /p:Configuration=Release /p:Platform=x64
``` ```

6
docs/en/build_and_install/gpu.md
View File

@@ -16,8 +16,8 @@ Please do not modify other cmake paramters exclude the following options.
| ENABLE_TEXT | Linux(x64/Windows(x64) | Default OFF, whether to intergrate text models | | ENABLE_TEXT | Linux(x64/Windows(x64) | Default OFF, whether to intergrate text models |
| CUDA_DIRECTORY | Linux(x64/Windows(x64) | Default /usr/local/cuda，require CUDA>=11.2 | | CUDA_DIRECTORY | Linux(x64/Windows(x64) | Default /usr/local/cuda，require CUDA>=11.2 |
| TRT_DIRECTORY | Linux(x64/Windows(x64) | Default empty，require TensorRT>=8.4， e.g. /Download/TensorRT-8.5 | | TRT_DIRECTORY | Linux(x64/Windows(x64) | Default empty，require TensorRT>=8.4， e.g. /Download/TensorRT-8.5 |
| ENABLE_CAPI | Linux(x64)/Windows(x64)/Mac OSX(x86) | Default OFF, whether to intergrate C API | | WITH_CAPI | Linux(x64)/Windows(x64)/Mac OSX(x86) | Default OFF, whether to intergrate C API |
| ENABLE_CSHARPAPI | Windows(x64) | Default OFF, whether to intergrate C# API | | WITH_CSHARPAPI | Windows(x64) | Default OFF, whether to intergrate C# API |
The configuration for third libraries(Optional, if the following option is not defined, the prebuilt third libraries will download automaticly while building FastDeploy). The configuration for third libraries(Optional, if the following option is not defined, the prebuilt third libraries will download automaticly while building FastDeploy).
| Option | Description | | Option | Description |
@@ -87,7 +87,7 @@ cmake .. -G "Visual Studio 16 2019" -A x64 \
-DTRT_DIRECTORY="D:\Paddle\TensorRT-8.4.1.5" \ -DTRT_DIRECTORY="D:\Paddle\TensorRT-8.4.1.5" \
-DCUDA_DIRECTORY="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.2" \ -DCUDA_DIRECTORY="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.2" \
-DCMAKE_INSTALL_PREFIX="D:\Paddle\compiled_fastdeploy" -DCMAKE_INSTALL_PREFIX="D:\Paddle\compiled_fastdeploy"
% nuget restore （please execute it when ENABLE_CSHARPAPI to prepare dependencies in C#) % 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 fastdeploy.sln /m /p:Configuration=Release /p:Platform=x64
msbuild INSTALL.vcxproj /m /p:Configuration=Release /p:Platform=x64 msbuild INSTALL.vcxproj /m /p:Configuration=Release /p:Platform=x64
``` ```

1
tutorials/README.md
View File

@@ -7,3 +7,4 @@ This directory provides some tutorials for FastDeploy. For other model deploymen
- Intel independent graphics card/integrated graphics card deployment [see intel_gpu](intel_gpu) - Intel independent graphics card/integrated graphics card deployment [see intel_gpu](intel_gpu)
- Model multithreaded call [see multi_thread](multi_thread) - Model multithreaded call [see multi_thread](multi_thread)
- Image decoding, including hardward decoding, e.g. nvJPEG [image_decoder](image_decoder) - Image decoding, including hardward decoding, e.g. nvJPEG [image_decoder](image_decoder)
- Deploy models with C or C# API [use_c_csharp_sdk](use_c_sharp_sdk)

1
tutorials/README_CN.md
View File

@@ -8,3 +8,4 @@
- Intel独立显卡/集成显卡部署 [见intel_gpu](intel_gpu) - Intel独立显卡/集成显卡部署 [见intel_gpu](intel_gpu)
- 模型多线程调用 [见multi_thread](multi_thread) - 模型多线程调用 [见multi_thread](multi_thread)
- 图片解码（含nvJPEG硬解码） [见image_decoder](image_decoder) - 图片解码（含nvJPEG硬解码） [见image_decoder](image_decoder)
- 使用C或C# API进行模型部署 [见use_c_csharp_sdk](use_c_sharp_sdk)

8
tutorials/use_c_csharp_sdk/README.md Normal file
View File

@@ -0,0 +1,8 @@
English | [中文](README_CN.md)
# Model Deployment with C or C# API
Fastdeploy provideds C and C# API to support deployment with multiple programming languages, and this feature is included since version 1.0.4. C API is compiled and included in fastdeploy dynamic library by default. If you want to compile fastdeploy with C and C# API manually, compile it with options -DWITH_CAPI=ON and -DWITH_CSHARPAPI=ON. For more information about how to use C and C# API to deploy models, please refer to
- [C API tutorial](../../c_api/README.md)
- [C# API tutorial](../../csharp/README.md)

10
tutorials/use_c_csharp_sdk/README_CN.md Normal file
View File

@@ -0,0 +1,10 @@
[English](README.md) | 中文
# 使用C或C# API进行模型部署
FastDeploy提供了C API以及C# API满足多语言部署的需求，这一特性从FastDeploy v1.0.4版本开始进行了支持。默认情况下，C API被编译进入了fastdeploy的动态库。
如果希望手动编译FastDeploy，并且集成C和C# API进去，只需要打开编译开关-DWITH_CAPI=ON, -DWITH_CSHARPAPI=ON。关于如何安装和使用C和C# API，可以参考下列文档：
- [C API指南](../../c_api/README_CN.md)
- [C# API指南](../../csharp/README_CN.md)