ceras
Drogon-torch-serve
ceras | Drogon-torch-serve | |
---|---|---|
1 | 6 | |
113 | 26 | |
- | - | |
3.6 | 0.0 | |
about 1 year ago | over 1 year ago | |
C++ | C++ | |
- | MIT License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
ceras
-
Neural nets in production systems
Maybe you can take a look at my toy repo (https://github.com/fengwang/ceras) before getting started. This library heavily uses templated meta-programming techniques to build the computation graph at compile time, and is header only.
Drogon-torch-serve
-
Neural nets in production systems
https://github.com/SABER-labs/Drogon-torch-serve example for serving torchscript model using Drogon C++ server, can be modified to serve ONNX instead.
-
What projects are you currently working on?
Working on https://github.com/SABER-labs/Drogon-torch-serve, a libtorch + Drogon model server, currently 12x faster than fastapi + pytorch, learning c++20 co-routines, multi-threaded communication, batch inference etc while doing the project.
- Serve Torch models using Drogon for higher throughput / lower latency
- Show HN: Serve Torch Models Using Drogon
- GitHub - SABER-labs/Drogon-torch-serve: Serve torch models using Drogon
What are some alternatives?
tensorflow - An Open Source Machine Learning Framework for Everyone
LibtorchTutorials - This is a code repository for pytorch c++ (or libtorch) tutorial.
PaddlePaddle - PArallel Distributed Deep LEarning: Machine Learning Framework from Industrial Practice (『飞桨』核心框架,深度学习&机器学习高性能单机、分布式训练和跨平台部署)
AECforWebAssembly - A port of ArithmeticExpressionCompiler from x86 to WebAssembly, so that the programs written in the language can run in a browser. The compiler has been rewritten from JavaScript into C++.
frugally-deep - Header-only library for using Keras (TensorFlow) models in C++.
SegmentationCpp - A c++ trainable semantic segmentation library based on libtorch (pytorch c++). Backbone: VGG, ResNet, ResNext. Architecture: FPN, U-Net, PAN, LinkNet, PSPNet, DeepLab-V3, DeepLab-V3+ by now.
CNTK - Microsoft Cognitive Toolkit (CNTK), an open source deep-learning toolkit
torchRL - TorchRL is a C++ reinforcement library using PyTorch C++ backend LibTorch
geodesic_raytracing
RESTinio - Cross-platform, efficient, customizable, and robust asynchronous HTTP(S)/WebSocket server C++ library with the right balance between performance and ease of use
muzero-cpp - A C++ pytorch implementation of MuZero