Release Notes Version 5 8

From Eigenvector Research Documentation Wiki
Jump to navigation Jump to search
The printable version is no longer supported and may have rendering errors. Please update your browser bookmarks and please use the default browser print function instead.

Version 5.8 of PLS_Toolbox and Solo was released in February, 2010.

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)

NEW FEATURES

Compatibility

  • Updated for Matlab 2010a compatibility.

Model Building

  • Support Vector Machine Regression and Classification non-linear modeling methods available in Analysis GUI (Solo and PLS_Toolbox) and command-line (PLS_Toolbox).
  • Locally Weighted Regression (LWR) non-linear modeling method available in Analysis GUI and significantly revised at the command-line to add new features, preprocessing support, and compatibility with Solo_Predictor 2.1
  • Permutation test added to Analysis GUI to allow easy testing for over-fit regression models (randomized y-block).
  • TrendTool outputs model structures which can be used to make predictions on new data and are compatible with Solo_Predictor 2.1. Interface is also now accessible through EVRIGUI_Objects
  • PLS-DA prior probability settings now available in Analysis GUI.

Scores Plots

  • Prediction intervals (Estimated prediction error) accessible in Scores plots

Preprocessing

  • Add "Transmission to Absorbance" preprocessing method (-log10(T)).

Importers/Exporters

  • Analytical Spectral Devices (ASD) Indico (Versions 6 and 7) data file reader added.
  • Horiba JY file importer fixes to data type handling and converts all data to double precision when importing.
  • Fix Excel reader problem which makes some XLS files unreadable on 64-bit systems.

Other Enhancements

  • Demo datasets now easily available in Model Cache.
  • Clearer deliniation of method categories and access to help in Analysis GUI.
  • Improved appearance and Copy/Paste functionality for MAC and Linux platforms.
  • NND Score Distance calculation now uses Mahalanobis distance
  • Improved Copy/Paste functionality from DataSet Editor data table (labels included and formatted for Excel compatibility)
  • Improved "single-window" feature - have all PLS_Toolbox or Solo windows "docked" in a single frame.

NEW FUNCTIONS AND FILES

asdreadr - Imports data from Analytical Spectral Devices (ASD) Indico (Versions 6 and 7) data files.
clipboard_image - Copy and paste images to/from the system clipboard.
ismodel - Returns boolean TRUE if input object is a standard model structure.
lwr - Locally Weighted Regression (LWR).
svm - Support Vector Machine (LIBSVM) for regression or classification.
svmda - Support Vector Machine (LIBSVM) for classification.