Release Notes Version 6 2

From Eigenvector Research Documentation Wiki
Jump to navigation Jump to search

Version 6.2 of PLS_Toolbox and Solo was released in March, 2011.

For general product information, see PLS_Toolbox Product Page. For information on Solo, see Solo Product Page.

(back to Release Notes PLS Toolbox and Solo)


Model Building

  • Stepwise CLS added as algorithm to CLS.
  • Orthgonalized PLSDA (O-PLSDA) models now fully supported.
  • Full cross-validation support of weighted regression.

Ease-of-Use Features

  • Magnify Tool - magnifying glass for any image or data.
  • New "Export Figure" option in FigBrowser menu to allow easy exporting of any figure.
  • Easy export of figures to HTML with several file types.

Functionality Improvements

  • Importer for StellarNet ABS files.
  • Improved Model Cache performance.
  • New and configurable statistics on regression plots.
  • Improved Support Vector Machine handling of unusual cases.
  • Improved performance of multi-axes scores plots.
  • Shortcut keys and menu to automatically select all members of the next/previous class.
  • Color-by and class symbol improvements (full support on standard plot types).


PLS_Toolbox-Specific Enhancements

  • Full Matlab 2011a support.
  • Confusion Tables and Matrices for SVMDA and PLSDA models (confusionmatrix, confusiontable, and getmisclassifieds)
  • Improved EVRIScript object access.
  • Inverse calculation of confidence limits from T^2 and Q values (tsqlim and chilimit).
  • Added Skewed Gaussian as peak option in FitPeaks.
  • Improved boxplot support.

DataSet Object Changes

  • No longer mistakenly allows NaNs in class assignments.
  • Add sparse, issparse, and full methods.


stepwise_regrcls -Stepwise CLS engine.
snabsreadr -StellarNet ABS file importer.
peakgaussianskew -Engine to allow use of skewed Gaussian peaks in peak fitting.
confusionmatrix -Engine for calculating confusion matrices.
confusiontable -Engine for calculating confusion tables.
getmisclassifieds -Engine for determining which samples were misclassified.