Whats New in Version 5 0

Version 5.0 of PLS_Toolbox and Solo was released in October, 2008.

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

NEW FEATURES

• Analysis GUI has undergone a major renovation allowing for much more flexible data management, plot and data "linking" (connections between data and model results like selections and exclusions in scores and loadings), and data manipulation.
• Analysis GUI now allows simultaneous loading of Calibration and Validation data and has a revised interface to make loading and working with data and models easier and more intuitive
• Improved interactivity for multi-way data in Analysis GUI
• Several preprocessing methods and options have been added including EPO, GLSW, windowed normalization, and baselining options.
• Additional importing methods including a greatly improved automatic text importer and support for several Hamilton Sundstrand file formats
• Improved editing, sorting, and visualization tools for DataSets
• Model robustness tests in Analysis GUI and at the command line
• Export model to "model building function" which can be used to reproduce a given model
• New clustering method "dbscan"
• Added "Color By" function for colorized plots and improved 3D visualization to Plot Controls.
• Sorting of data by rows, columns, or any label field.
• Simple "Open in Analysis GUI" option on any data to allow for simple multi-block analyses (e.g. combination of Scores from multiple models)
• Improved missing data support (automatic exclusion of un-replaceable data)
• Simplified window-docking options allow for "docking" all windows in a single Figures window.
• Improved automatic factor selection
• IPLSDA (Interval PLSDA variable selection) support
• Batch and Wafer digestion support for "Batch" DataSet objects (supports different length batches)
• New DataSet Object features (DSO version 5.1)

NEW FUNCTIONS AND FILES

asfreadr - Reads a Hamilton Sundstrand data file.

dbscan - Density-based automatic sample clustering.

encodemodelbuilder - Create MATLAB m-code which will regenerate a given model.

ils_esterror - Estimation error for ILS models.

matchrows - Matches up rows from two DataSet objects using labels or sizes.

testrobustness - Test regression model for robustness to various effects.

tsqqmtx - Calculates matrix for T^2+Q contributions for PCA and MPCA.

@dataset\sortby.m

@dataset\sortcols.m