diffcode

diffcode

Diffcode toolbox

Purpose

This toolbox provides forward automatic differentiation. It is based on operators and functions overloading.

Given a Scilab code computing a variable y depending on a variable x and a direction dx it allow evaluation of y together with the directional derivative Grad(y)*dx.

The module supports common arithmetic operations, common elementary functions and several matrix functions, including matrix inversion.

Type :

help diffcode_overview

for a quick start.

Features

The following is a list of the current functions :

  • diffcode_der - Create a new code differentiation object.
  • diffcode_CDcost - Objective function for optim.
  • diffcode_hessian — Compute the Hessian of the function.
  • diffcode_jacobian — Compute the Jacobian of the function.

Dependencies

  • This module depends on the helptbx module.
  • This module depends on the assert module.
  • This module depends on the apifun module.

History

This toolbox was first released at:

http://www.scilab.org/contrib/index_contrib.php?page=displayContribution&fileID=186

in August 2002.

Michael Baudin updated the module for Scilab v5.

Authors

  • Copyright (C) 2011 - DIGITEO - Michael Baudin
  • Copyright (C) 2002 - INRIA - Xavier Jonsson
  • Copyright (C) 2002-2009 - INRIA - Serge Steer

Licence

This toolbox is released under the CeCILL_V2 licence :

http://www.cecill.info/licences/Licence_CeCILL_V2-en.txt

Powered by InDefero,
a Céondo Ltd initiative.