|
|
 
 

SOLO provides the Graphical Interfaces for quickly managing and analyzing data, authoring and applying models and interpreting results. Data can be imported from a variety of different file types and quickly assembled into convenient DataSet objects. Once in a DataSet object the user can easily manage labels, axis scales, and classes and can include/exclude data from the analysis with a click. Modeling and analyzing results is just a matter of drag-and-drop of the data to our comprehensive Analysis GUI.

The PLS_Toolbox is a collection of advanced chemometric routines, i.e., tools designed specifically for modeling data from chemical systems. Originally named for the partial least squares (PLS) routines that have become a mainstay for calibration model development, the PLS_Toolbox now contains over 100 functions that can be used in a variety of technical areas. This includes linear and nonlinear modeling tools, along with functions that help the user apply them in the chemical laboratory or process environment. Tools included in the toolbox are point and click graphical user interfaces for principal components analysis and partial least squares regression and a suite of tools for the analysis of multi-way data.
With MIA_Toolbox, multivariate images from remote sensing to microscopy can be easily analyzed using the same PLS_Toolbox tools you are already familiar with. MIA_Toolbox allows you to load, manipulate, and analyze multivariate images in the Analysis Graphical interface as well as using many of the higher-level command-line functions. Principal components analysis, multivariate curve resolution, SIMCA and PLSDA classification, K-Means clustering, and even regression can all be performed on images with this extension pack.
Extended Multiplicative Scatter Correction (EMSC) is a powerful preprocessing technique used to isolate and remove complicated multiplicative and additive effects, such as those caused by light scattering in reflectance spectroscopy. It expands on the popular Multiplicative Scatter/Signal Correction (MSC) technique by offering much improved flexibility in identifying known interferences to remove as well as scaling targets and known analyte spectra. Eigenvector Research is offering this powerful patented technique as an add-on for use with PLS_Toolbox as EMSC_Toolbox.
Model_Exporter converts models created within the PLS_Toolbox or Solo chemometrics modeling environments into an interpretable format for use outside of these products.

 


Export, Integrate, Deploy ... Easy


See what makes NeuroSolutions the premier neural network development environment. Take a few minutes to browse through the key benefits of the software, or simply visit the topics that interest you.
Model_Exporter
provides a unique and powerful means to add fast and flexible model prediction capability to any third-party software package without the need for outside libraries or toolboxes.

Model_Exporter, an add-on product for PLS_Toolbox and Solo, exports models into a stand-alone "predictor" script which includes a simple-to-implement formula to perform a model prediction.

Add the power of multivariate modeling to your proprietary application or simply apply models in Mathworks' MATLAB™, Octave, Tcl, or Symbion Systems' Symbion™ without the need for costly 3rd party toolboxes or libraries*!

See what makes NeuroSolutions the premier neural network development environment. Take a few minutes to browse through the key benefits of the software, or simply visit the topics that interest you.

Let Eigenvector do the math!

Instead of having to reinvent the mathematical steps and translate complicated proprietary file formats, you can use Model_Exporter to provide all the model parameters and numerical calculations you need to make a prediction for a PLS, PCR, or PCA model, including all data preprocessing!

Generic XML format: Predictor scripts in the generic XML format can be easily parsed without the need for external libraries or other additional software (* some standard free libraries may be required for implementing a parser in some languages). All you need to provide is the ability to parse XML, manage matrices, and do several simple (standard!) mathematical operations. Model_Exporter handles everything else.

LabView, MATLAB and Symbion Applications: In addition to the generic XML format, Model_Exporter also creates predictors for native use in LabView, MATLAB, and Symbion. These predictors can be run without the need for any additional toolboxes or purchased libraries. The MATLAB m-file script will even run in the free Octave package and the Symbion scripts will run in free Tcl parsers. These m-files run in any LabView package with MathScript textual math included (Full package or better.)

Plug-in modular design: By using a modular, self-describing script format, users can easily update models without the need to recompile your application; Simply make a new XML, m-file or tcl-file available to your program and the new model will automatically be used. You need know nothing about the new model -- it handles all the numerical variations.

Example applications:

  • Handheld or similar devices using embedded software (common languages for parsers include C++, C, C#, Java, Basic, Pearl, Tcl, -- essentially any language which allows mathematical calculations!
  • High-throughput applications requiring high-speed data manipulation and calculations.
  • Wide distributed applications in MathWorks’ MATLABTM or free Octave software.
  • On-line predictions using Symbion Systems’ SymbionTM software.

Model_Exporter allows you to be self-sufficient and run all the calculations in your own environment.


System Requirements

Model_Exporter can be executed from either the MATLAB computational environment (Mathworks, Inc., Natick, MA), or Solo (Eigenvector Research, Inc., Wenatchee, WA). Model_Exporter converts models created by PLS_Toolbox 3.5 or higher or Solo 4.0 or higher.

Matlab-Based Exporter Requirements
For execution of Model_Exporter within the MATLAB environment, the following is required:
Matlab 6.5 or higher
256 MB RAM (recommended – less may be required)

Solo-Based Exporter Requirements
For execution of the Model_Exporter, the following is recommended
Solo 4.1 or higher
Windows 2000, XP, 2003 server, Vista, or MAC OS X (intel)
200 MB Disk Space (for installation; some models may require additional space)
256 MB RAM (recommended – less may be required)

Requirements for Using Exported Models
The requirements to execute an exported model vary depending on the interpreter used, the number of variables in the modeled data, and the complexity of the model (i.e. the number of factors/components included in the model and the types of preprocessing used).

Memory requirements depend on the precision required for the application, the number of variables in the data and the total number of factors in the model. For example, a model working on 10,000 variables and 5 factors would require around 1MB for doubleprecision calculations and 500KB single-precision calculations.

The software which executes the specific file formats may have additional requirements. See the file format description sections later in this manual for where to locate model execution details.

copyright © 2008
SOLUTIONS4U MY Co REG. No: 706527A
SOLUTIONS4U SG Co REG. No: 200713296Z
ABOUT US sitemap contact us privacy policy terms & conditions