/PaddleSharp

.NET/C# binding for Baidu paddle inference library and PaddleOCR

Primary LanguageC#Apache License 2.0Apache-2.0

PaddleSharp ๐ŸŒŸ main QQ

English | ็ฎ€ไฝ“ไธญๆ–‡

๐Ÿ’— .NET Wrapper for PaddleInference C API, support Windows(x64) ๐Ÿ’ป, NVIDIA Cuda 10.2+ based GPU ๐ŸŽฎ and Linux(Ubuntu-22.04 x64) ๐Ÿง, currently contained following main components:

  • PaddleOCR ๐Ÿ“– support 14 OCR languages model download on-demand, allow rotated text angle detection, 180 degree text detection, also support table recognition ๐Ÿ“Š.
  • PaddleDetection ๐ŸŽฏ support PPYolo detection model and PicoDet model ๐Ÿน.
  • RotationDetection ๐Ÿ”„ use Baidu's official text_image_orientation_infer model to detect text picture's rotation angle(0, 90, 180, 270).
  • Paddle2Onnx ๐Ÿ”„ Allow user export ONNX model using C#.

NuGet Packages/Docker Images ๐Ÿ“ฆ

Release notes ๐Ÿ“

Please checkout this page ๐Ÿ“„.

Infrastructure packages ๐Ÿ—๏ธ

NuGet Package ๐Ÿ’ผ Version ๐Ÿ“Œ Description ๐Ÿ“š
Sdcb.PaddleInference NuGet Paddle Inference C API .NET binding โš™๏ธ

Native packages ๐Ÿ—๏ธ

Package Version ๐Ÿ“Œ Description
Sdcb.PaddleInference.runtime.win64.mkl NuGet win64+mkldnn
Sdcb.PaddleInference.runtime.win64.openblas NuGet win64+openblas
Sdcb.PaddleInference.runtime.win64.openblas-noavx NuGet win64+openblas(no AVX, for old CPUs)
Sdcb.PaddleInference.runtime.win64.cuda102_cudnn76_tr72_sm61_75 NuGet win64/CUDA 10.2/cuDNN 7.6/TensorRT 7.2/sm61+sm75
Sdcb.PaddleInference.runtime.win64.cuda118_cudnn86_tr85_sm86_89 NuGet win64/CUDA 11.8/cuDNN 8.6/TensorRT 8.5/sm86+sm89

Some of packages already deprecated(Version <= 2.5.0):

Package Version ๐Ÿ“Œ Description
Sdcb.PaddleInference.runtime.win64.cuda117_cudnn84_tr84_sm86 NuGet win64/CUDA 11.7/cuDNN 8.4/TensorRT 8.4/sm86
Sdcb.PaddleInference.runtime.win64.cuda102_cudnn76_sm61_75 NuGet win64/CUDA 10.2/cuDNN 7.6/sm61+sm75
Sdcb.PaddleInference.runtime.win64.cuda116_cudnn84_sm86_onnx NuGet win64/CUDA 11.6/cuDNN 8.4/sm86/onnx

Any other packages that starts with Sdcb.PaddleInference.runtime might deprecated.

Baidu packages were downloaded from here: https://www.paddlepaddle.org.cn/inference/master/guides/install/download_lib.html#windows

My Sdcb packages were self compiled.

Baidu official GPU packages are too large(>1.5GB) to publish to nuget.org, there is a limitation of 250MB when upload to Github, there is some related issues to this:

But You're good to build your own GPU nuget package using 01-build-native.linq ๐Ÿ› ๏ธ.

Note: Linux does not need a native binding NuGet package like windows(Sdcb.PaddleInference.runtime.win64.mkl), instead, you can/should based from a Dockerfile๐Ÿณ to development:

Docker Images ๐Ÿณ Version ๐Ÿ“Œ Description ๐Ÿ“š
sdflysha/dotnet6-paddle Docker PaddleInference 2.5.0, OpenCV 4.7.0, based on official Ubuntu 22.04 .NET 6 Runtime ๐ŸŒ
sdflysha/dotnet6sdk-paddle Docker PaddleInference 2.5.0, OpenCV 4.7.0, based on official Ubuntu 22.04 .NET 6 SDK ๐ŸŒ

You can also build your docker images using these dockerfiles.

Paddle Devices

  • Mkldnn - PaddleDevice.Mkldnn()

    Based on Mkldnn, generally fast

  • Openblas - PaddleDevice.Openblas()

    Based on openblas, slower, but dependencies file smaller and consume lesser memory

  • Onnx - PaddleDevice.Onnx()

    Based on onnxruntime, is also pretty fast and consume less memory

  • Gpu - PaddleDevice.Gpu()

    Much faster but relies on NVIDIA GPU and CUDA

    If you wants to use GPU, you should refer to FAQ How to enable GPU? section, CUDA/cuDNN/TensorRT need to be installed manually.

  • TensorRT - PaddleDevice.Gpu().And(PaddleDevice.TensorRt("shape-info.txt"))

    Even faster than raw Gpu but need install TensorRT environment.

    Please refer to tensorrt section for more details

FAQ โ“

Why my code runs good in my windows machine, but DllNotFoundException in other machine: ๐Ÿ’ป

  1. Please ensure the latest Visual C++ Redistributable was installed in Windows (typically it should automatically installed if you have Visual Studio installed) ๐Ÿ› ๏ธ Otherwise, it will fail with the following error (Windows only):

    DllNotFoundException: Unable to load DLL 'paddle_inference_c' or one of its dependencies (0x8007007E)
    

    If it's Unable to load DLL OpenCvSharpExtern.dll or one of its dependencies, then most likely the Media Foundation is not installed in the Windows Server 2012 R2 machine: image

  2. Many old CPUs do not support AVX instructions, please ensure your CPU supports AVX, or download the x64-noavx-openblas DLLs and disable Mkldnn: PaddleDevice.Openblas() ๐Ÿš€

  3. If you're using Win7-x64, and your CPU does support AVX2, then you might also need to extract the following 3 DLLs into C:\Windows\System32 folder to make it run: ๐Ÿ’พ

    • api-ms-win-core-libraryloader-l1-2-0.dll
    • api-ms-win-core-processtopology-obsolete-l1-1-0.dll
    • API-MS-Win-Eventing-Provider-L1-1-0.dll

    You can download these 3 DLLs here: win7-x64-onnxruntime-missing-dlls.zip โฌ‡๏ธ

How to enable GPU? ๐ŸŽฎ

Enable GPU support can significantly improve the throughput and lower the CPU usage. ๐Ÿš€

Steps to use GPU in Windows:

  1. (for Windows) Install the package: Sdcb.PaddleInference.runtime.win64.cuda* instead of Sdcb.PaddleInference.runtime.win64.mkl, do not install both. ๐Ÿ“ฆ
  2. Install CUDA from NVIDIA, and configure environment variables to PATH or LD_LIBRARY_PATH (Linux) ๐Ÿ”ง
  3. Install cuDNN from NVIDIA, and configure environment variables to PATH or LD_LIBRARY_PATH (Linux) ๐Ÿ› ๏ธ
  4. Install TensorRT from NVIDIA, and configure environment variables to PATH or LD_LIBRARY_PATH (Linux) โš™๏ธ

You can refer to this blog page for GPU in Windows: ๅ…ณไบŽPaddleSharp GPUไฝฟ็”จ ๅธธ่ง้—ฎ้ข˜่ฎฐๅฝ• ๐Ÿ“

If you're using Linux, you need to compile your own OpenCvSharp4 environment following the docker build scripts and the CUDA/cuDNN/TensorRT configuration tasks. ๐Ÿง

After these steps are completed, you can try specifying PaddleDevice.Gpu() in the paddle device configuration parameter, then enjoy the performance boost! ๐ŸŽ‰

TensorRT ๐Ÿš„

To use TensorRT, just specify PaddleDevice.Gpu().And(PaddleDevice.TensorRt("shape-info.txt")) instead of PaddleDevice.Gpu() to make it work. ๐Ÿ’ก

Please be aware, this shape info text file **.txt is bound to your model. Different models have different shape info, so if you're using a complex model like Sdcb.PaddleOCR, you should use different shapes for different models like this:

using PaddleOcrAll all = new(model,
   PaddleDevice.Gpu().And(PaddleDevice.TensorRt("det.txt")),
   PaddleDevice.Gpu().And(PaddleDevice.TensorRt("cls.txt")),
   PaddleDevice.Gpu().And(PaddleDevice.TensorRt("rec.txt")))
{
   Enable180Classification = true,
   AllowRotateDetection = true,
};

In this case:

  • DetectionModel will use det.txt ๐Ÿ”
  • 180DegreeClassificationModel will use cls.txt ๐Ÿ”ƒ
  • RecognitionModel will use rec.txt ๐Ÿ”ก

NOTE ๐Ÿ“:

The first round of TensorRT running will generate a shape info **.txt file in this folder: %AppData%\Sdcb.PaddleInference\TensorRtCache. It will take around 100 seconds to finish TensorRT cache generation. After that, it should be faster than the general GPU. ๐Ÿš€

In this case, if something strange happens (for example, you mistakenly create the same shape-info.txt file for different models), you can delete this folder to generate TensorRT cache again: %AppData%\Sdcb.PaddleInference\TensorRtCache. ๐Ÿ—‘๏ธ

Thanks & Sponsors ๐Ÿ™

Contact ๐Ÿ“ž

QQ group of C#/.NET computer vision technical communication (C#/.NET่ฎก็ฎ—ๆœบ่ง†่ง‰ๆŠ€ๆœฏไบคๆต็พค): 579060605