October 13, 2015
MACROS 6.2 Release
DATADVANCE development team is pleased to announce the release of MACROS 6.2, a new stable version of the MACROS library and featured plug-ins.
This release adds a new sample generation method to GTDoE that allows to update an existing LHS or OLHS sample, keeping its properties. The new method is a part of the sample-based adaptive DoE technique. It adds points to an initial sample and, if the latter is an (optimized) Latin hypercube, ensures that the new sample is also an (optimized) Latin hypercube. In particular, this method can be used to generate LHS and OLHS incrementally. This update also fixes a bug in GTDoE which rarely lead to incorrect OLHS generation.
Other notable updates in MACROS 6.2 include:
- Improvements and bugfixes in the surrogate based optimization method in GTOpt.
- Updated Gradient Boosting Regression Trees (GBRT) and Tensor Approximation techniques in GTApprox.
- Reduced size of GBRT and componentwise GP (Gaussian processes) approximation models.
- New capability to control the complexity of tensor approximation models.
- Updates to correlation analysis and sensitivity analysis methods in GTSDA.
MACROS 6.2 also fixes several minor bugs in GTApprox and GTSDA. For full details on recent updates and bugfixes you can refer to the latest changelog, or contact us to receive more information and MACROS updates!