Documentation for torchpipe

Ensemble Pipeline Serving with Pytorch Frontend

torchpipe 是 介于底层加速库（如tensorrt，opencv，torchscript）以及 RPC（如thrift, gRPC）之间并与他们严格解耦的多实例流水线并行库；对外提供面向pytorch前端的线程安全函数接口，对内提供面向用户的细粒度后端扩展。

Torchpipe is a multi-instance pipeline parallel library that acts as a bridge between lower-level acceleration libraries (such as TensorRT, OpenCV, TorchScript) and RPC frameworks (like Thrift, gRPC), ensuring a strict decoupling from them. It offers a thread-safe function interface for the PyTorch frontend at a higher level, while empowering users with fine-grained backend extension capabilities at a lower level.

这里是其文档站点

Here is its documentation site.

Edit

• Chinese version located at：./i18n/zh/docusaurus-plugin-content-docs/current

Compile

See ./compile.md