June 26, 2013

MACROS 1.9.6 Released

DATADVANCE development team is pleased to announce MACROS 1.9.6, the next release in MACROS 1.9 series.

This release implements several significant improvements to the GTApprox Gaussian Processes (GP) technique, increasing the technique's robustness and model quality. As a result, the recent update of internal GTApprox algorithms:

  • resolved the problem of GP models degrading to constant in certain parameter space investigation tasks,
  • allowed to handle noisy sample data better, in particular avoiding automatic exact fit when it is not wanted, and
  • allowed user to control the trade-off between model accuracy and robustness in GP models.

Moreover, this update also made it possible to improve GTDF, preventing a degenerate behavior of Data Fusion models in the cases when the high- and low-fidelity training samples are located in different design space regions.

An important fix of multithreading is included to GTOpt, repairing a few unstable methods in multithreading implementation which could lead to crashes.

The release also includes some lesser updates in GTApprox and GTDR:

  • An improvement in GTApprox model export to C - exported models now contain two dedicated methods to calculate model values and accuracy estimation. Both methods also allow to calculate values and gradients separately.
  • A fix to the incomplete Tensor Approximation technique resolving an issue with incorrect processing of sample point weights, which sometimes lead to an undefined model builder behavior.
  • Ability to manually select the algorithm for the internal approximator used by GTDR in the sample-based Feature Extraction mode.

The details on these updates may be found in the release changelog.

Important:
Since this release, MACROS for Python requires NumPy version 1.6.1 or newer. The MACROS for Python Manual has been updated with the regard for the new requirement; follow the links from the release changelog to find the details.
The decision to include NumPy into system requirements is due to the fact that it is a widely used package in scientific computing, and usually it is already installed on target systems; MACROS for Python, in turn, will benefit from the capabilities provided by NumPy. This update does not break version compatibility: scripts made for previous MACROS for Python versions will work as is.

Please contact us to receive more information and MACROS updates!

LinkedIn
VK