Skip to content

Clad version 1.0
Compare
Choose a tag to compare
@vgvassilev vgvassilev released this 07 Oct 00:02
· 280 commits to master since this release
v1.0
ca8b09d

Introduction

This document contains the release notes for the automatic differentiation plugin for clang Clad, release 1.0. Clad is built on top of Clang and LLVM compiler infrastructure. Here we describe the status of Clad in some detail, including major improvements from the previous release and new feature work.

Note that if you are reading this file from a git checkout, this document applies to the next release, not the current one.

What's New in Clad 1.0?

Some of the major new features and improvements to Clad are listed here. Generic improvements to Clad as a whole or to its underlying infrastructure are described first.

External Dependencies

  • Clad now works with clang-5.0 to clang-14

Forward Mode & Reverse Mode

  • Add support for pushforward- and pullback-style functions which allow to accumulate the results from the AD when required to correctly compute derivatives when arguments are passed by reference or pointers

Forward Mode

  • Add support for member variables in functors
  • Add basic support for virtual functions
  • Add support for reference arguments
  • Add basic support for AD of class types wrt scalars
  • Add support for member functions, pointers, overloaded operators, pointer arithmetic, nullptr, sizeof and pseudo objects,

Reverse Mode

  • Add support for while and do-while statements
  • Add initial support for AD of user-defined types allowing to differentiate scalar types wrt user-defined types

CUDA

  • Add forward mode support for basic CUDA programs. More can be seen here

Error Estimation

  • Developed an error estimation framework to perform AD-based error estimation. The new facility is available via the clad::estimate_error interface. See more in this demo

Misc

  • Developed user documentation available at clad.readthedocs.io
  • Developed developers documentation available at doxygen
  • Implement a fallback to numerical differentiation if Clad cannot differentiate a given function. To disable this behavior, please compile your programs with the -DCLAD_NO_NUM_DIFF.
  • Add benchmarking infrastructure based on google benchmark
  • Add integration with Enzyme via clad::gradient<clad::opts::use_enzyme>(...)

Fixed Bugs

28 281 353 368 386 387 393 440

Special Kudos

This release wouldn't have happened without the efforts of our contributors, listed in the form of Firstname Lastname (#contributions):

FirstName LastName (#commits)

Parth Arora (65)
Vassil Vassilev (49)
Garima Singh (17)
Baidyanath Kundu (13)
Nirhar (12)
Ioana Ifrim (9)
Alexander Penev (4)
RohitRathore1 (1)
David (1)

Assets 2