Release Notes Version 6 2: Difference between revisions
Jump to navigation
Jump to search
imported>Jeremy No edit summary |
imported>Jeremy |
||
| Line 28: | Line 28: | ||
===Documentation=== | ===Documentation=== | ||
* General content improvements | * General content improvements | ||
* New demos for: [[fasternnls]], [[wlsbaseline]], | * New demos for: [[fasternnls]], [[wlsbaseline]], confusionmatrix, and [[stepwise_regrcls]] | ||
===PLS_Toolbox-Specific Enhancements=== | ===PLS_Toolbox-Specific Enhancements=== | ||
Revision as of 13:13, 8 March 2011
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, seeSolo Product Page.
(back to Release Notes PLS Toolbox and Solo)
NEW FEATURES
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).
Documentation
- General content improvements
- New demos for: fasternnls, wlsbaseline, confusionmatrix, and stepwise_regrcls
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
NEW FUNCTIONS AND FILES
- 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