DACE- a matlab Kriging toolbox

本文主要是介绍DACE- a matlab Kriging toolbox，希望对大家解决编程问题提供一定的参考价值，需要的开发者们随着小编来一起学习吧！

以下为转帖

http://www2.imm.dtu.dk/projects/dace/

http://www.imm.dtu.dk/~hbni/dace/【已经失效】

对比的工具：https://cn.mathworks.com/matlabcentral/fileexchange/59960-krigingtoolbox

DACE

A Matlab Kriging Toolbox

Description

DACE, Design and Analysis of Computer Experiments, is aMatlab toolbox for working with kriging approximations to computer models.

Typical use of this software is to construct a kriging approximation modelbased on data from a computer experiment, and to use this approximation model asa surrogate for the computer model.

The software also addresses the design of experiment problem, that ischoosing the inputs at which to evaluate the computer model for constructing thekriging approximation.

11 design sites, Gauss correlation model

Presentations

Kriging and radial basis functions. SIAM OP05.Stockholm, May 2005.
Surrogate models.ERCIM Copenhagen meeting, April 2005.
Surrogate models. IMM, DTU, November 2004.Focus on Kriging and radial basis functions.
Surrogate modelling by Kriging. ISMP2003, Copenhagen, August 2003.

Availability

The software and documentation are available for free. Selling the software and/or the documentation for commercial advantage, in part(e.g., included in other software) or as a whole, is allowed only with a priorwritten agreement with the authors (see feedback section below for contactinformation).

Software

The toolbox is available as a zip-archive: dace.zip (1.5Mb). Unzip the file and add the toolboxdirectory to the Matlab path (refer to the Matlab manual for help on this).

Current version is 2.5 from September 4, 2002. See the changelog for revision history.

Requirements: The software is developed and tested with Matlabversion R12.1, though it probably works with other (recent) versions of Matlab.

Documentation

The manual (including reference manual and examples) is available as a pdf-file (1.5Mb) (the file is also included in the archive with the software).

Further documentation of the algorithms is given in the report Aspects of the Matlab Toolbox DACE, which is available as aps-file (1Mb).

People involved:The team behind the software and documentation are