Skip to content

forward and reverse mode automatic differentiation primitives for Julia Base + StdLibs

License

Notifications You must be signed in to change notification settings

mhauru/ChainRules.jl

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ChainRules

CI Travis Coveralls PkgEval Code Style: Blue ColPrac: Contributor's Guide on Collaborative Practices for Community Packages DOI

Docs:

⚠️ The ChainRules project is currently seeking a new primary maintainer ⚠️
See this Discourse thread. We currently have many secondary maintainers and contributors with commit rights, but noone who has committed to leading the efforts. Until this is resolved PRs and issues may not recieve timely response.

The ChainRules package provides a variety of common utilities that can be used by downstream automatic differentiation (AD) tools to define and execute forward-, reverse-, and mixed-mode primitives.

The core logic of ChainRules is implemented in ChainRulesCore.jl. To add ChainRules support to your package, by defining new rrules or frules, you only need to depend on the very light-weight package ChainRulesCore.jl. This repository contains ChainRules.jl, which is what people actually use directly. ChainRules reexports all the ChainRulesCore functionality, and has all the rules for the Julia standard library.

Here are some of the core features of the package:

  • Mixed-mode composability without being coupled to a specific AD implementation.
  • Extensible rules: package authors can add rules (and thus AD support) to the functions in their packages, without needing to make a PR to ChainRules.jl .
  • Control-inverted design: rule authors can fully specify derivatives in a concise manner that supports computational efficiencies, so we will only compute as much as the user requests.
  • Propagation semantics built-in, with default implementations that allow rule authors to easily opt-in to common optimizations (fusion, increment elision, memoization, etc.).

About

forward and reverse mode automatic differentiation primitives for Julia Base + StdLibs

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Julia 100.0%