About PLS Toolbox and Solo
About PLS_Toolbox and Solo
PLS_Toolbox is a collection of essential and advanced chemometric routines built within the MATLAB® computational environment. It contains the tools required by engineers and scientists to explore their data and build predictive models. PLS_Toolbox gets its name from the Partial Least Squares (PLS) regression method, which has become the standard calibration method in many chemical applications. Although PLS_Toolbox has grown far beyond just PLS, the name remains.
Solo is PLS_Toolbox without MATLAB. We use the MATLAB® Compiler™ to compile all of the PLS_Toolbox GUI interfaces into a standalone application. The only difference between the two products is access to the MATLAB command line.
Users of earlier versions have provided much of the impetus for inclusion of additional routines and improvements to existing routines. We encourage you to make suggestions for routines you would find valuable for inclusion in future versions, and appreciate your comments on how we can improve PLS_Toolbox. We are especially indebted to those of you who have shared your routines and improvements to our routines with us.
PLS_Toolbox is frequently updated with new routines as research in the field of chemometrics continues. It is the intention Eigenvector Research Inc. to keep PLS_Toolbox as close to the state-of-the-art as practical. As our own research in chemometrics continues we will add routines that we find useful. Every attempt will be made to turn significant research results, whether our own or those of others, into working methods in the shortest time possible. New routines will initially be distributed to our users through our web site at www.eigenvector.com, and then incorporated into future versions of the toolbox. It is our goal to make PLS_Toolbox a repository of current state of the art techniques, and include these techniques long before stand-alone packages appear on the market.
MATLAB, an acronym for MATrix LABoratory, is a product of The MathWorks, Inc., of Natick, MA. It is a computational environment especially suited to problems in matrix algebra, allowing the user to do matrix computations using a command language that looks very much like standard linear algebra notation. MATLAB, which is required to run the toolbox routines, is not an inexpensive package. However, because of its large collection of built-in functions, easy to learn language and flexible environment, it is a great value. Another positive aspect of MATLAB is that it is available for many platforms, including Windows, Unix, Linux and Mac OS X. Thus, PLS_Toolbox can be run on any of these platforms, as can any other user defined functions.
The inexperienced MATLAB user will probably want to become familiar with the MATLAB package before attempting to use PLS_Toolbox. We have provided a chapter in this manual that introduces the basic features of MATLAB required to run PLS_Toolbox. We also suggest reading the Getting Started with MATLAB guide that comes with MATLAB. While much of this guide covers material beyond what a PLS_Toolbox user requires, being knowledgeable about MATLAB will help improve the overall usefulness of PLS_Toolbox.
The user can obtain help on any of the functions in PLS_Toolbox simply by typing help functionname at the MATLAB prompt. A brief message will be printed to the screen that contains information about the function and gives its I/O format. Short help, which includes only the I/O format, can be obtained by typing just functionname at the prompt for functions that do not have a GUI interface. The MATLAB ‘lookfor’ command can be used to find routines in PLS_Toolbox that perform specific functions. In addition, the ‘helppls’ function, (executed by typing helppls at the command line), opens the MATLAB Help Browser with a list of PLS_Toolbox help topics.
We have written all the functions in PLS_Toolbox so that they use only the functions in the “kernel” of MATLAB versions 6.5 and above. None of the functions in this package require any functions from other MATLAB Toolboxes, (though in some instances they will call functions from other toolboxes if they are found). There are several of these Toolboxes that the user may find helpful, however. We have used the System Identification, Signal Processing, Optimization, and Control Toolboxes extensively and have found them to be very useful. These toolboxes are available through The MathWorks. They provide many tools that are complementary to those in PLS_Toolbox.