December 25, 2012

MACROS 1.8 Released

DATADVANCE development team is pleased to announce the release of MACROS 1.8.

This release brings many new features, performance and stability improvements. A brief summary of new features is given below.


  • Added surrogate model export to plain old C, Octave and MEX: constructed surrogate model can now be used completely without MACROS.
  • Added new approximation technique - Response Surface Model.
  • Added pre- and postprocessing capabilities which assists user to configure the tool to get the most accurate model using given data and to improve constructed surrogate model.
  • Significantly enhanced quality of approximation of noisy functions with Gaussian Process technique.
  • Added an optional parameter to specify the variance of response values in the training sample.
  • Added response noise variance information to the internal validation output.
  • Added method to calculate accuracy evaluation gradients.
  • Fixed the strong dependency of internal validation results on the random seed.


  • Implemented new internal solver algorithm (Sequential Quadratic Constraints Quadratic Programming) which further reduces the number of objective functions and constraints evaluations.
  • Added Surrogate Based Optimization strategy.
  • Added new globalization strategy based on lightweight version of Surrogate Based Optimization.
  • Implemented batch mode which allows to evaluate objective functions and constraints in multiple design points simultaneously. 


  • Added Adaptive Design of Experiments technique based on Gaussian Processes.
  • Added Optimal Design technique.

Other changes

  • Improved the support for non-English system locales.
  • Added new usage examples.
  • Regression tests results and details included into documentation.

Please contact us for more information or to get new MACROS versions!


Contact Information

location_on  31100, Toulouse, Avenue du Général de Croutte 42

phone  +33 (0) 5 82-95-59-68


Contact us navigate_next Distributors navigate_next