Release Notes Version 8 0: Difference between revisions

From Eigenvector Research Documentation Wiki
Jump to navigation Jump to search
imported>Scott
No edit summary
imported>Scott
No edit summary
Line 10: Line 10:
* [[mlsca|MLSCA]] - Multi-level simultaneous component analysis.
* [[mlsca|MLSCA]] - Multi-level simultaneous component analysis.
* History now captured in history field of [[Model_object| Model Object]].
* History now captured in history field of [[Model_object| Model Object]].
** Add .scoredistance and .esterror as virtual fields for models.
* [[ann|ANN]] now supports custom cross-validation.
* [[ann|ANN]] now supports custom cross-validation.
* [[plsda|PLSDA]] varcap plot now available.
* [[plsda|PLSDA]] varcap plot now available.
* Better handle full-rank PCA sub-models where Q residuals are zero in [[simca]].
* Better handle full-rank PCA sub-models where Q residuals are zero in [[simca]].
* Better handling of errors during variable matching (test to model).


===Plotting===
===Plotting===
Line 18: Line 20:
* Better handling of font sizes.
* Better handling of font sizes.
* Smarter plot style (e.g., scatter vs. bar) assumptions in "automatic" mode.
* Smarter plot style (e.g., scatter vs. bar) assumptions in "automatic" mode.
* Fix rearranging of controls issue with newer versions of Matlab.
* ********{REV 19448} add viewcompressgaps option?
* Connect Classes button has drop-down menu to display connection types.
* Add logic to NORMALIZE score distance to maximum calibration value.
* Respect include field when calculating scores limit
* Show knn score distance limits on plotscores plots.
* Better handling of crossval when using PLSDA.
** Convert plsda regression method input to be 'sim' (to speed it up).
** Recognize when user has passed single-column (either logical or class) and force it to be multi-column logical.


===Importers===
===Importers===
* [[omnicreadr]] - Reads OMNICix HDF5 image files.
* [[omnicreadr]] - Reads OMNICix HDF5 image files.
* Improved importer behavior with mixed length data using [[matchvars]] during import.
* Improved importer behavior with mixed length data using [[matchvars]] during import.
* Updated libraries for Horiba file formats including 64 bit.
* Updated libraries to use new labspec 6 components including 64 bit support.
* [[xlsreadr]] - Add support for joining sheets in rows using matchvars.
** Add support for using the same graphically-selected parsing options on ALL sheets.


===Preprocessing===
===Preprocessing===
Line 37: Line 50:
* [[multiblocktool|Multiblock Tool]] - Interface to view, manipulate, and join data.
* [[multiblocktool|Multiblock Tool]] - Interface to view, manipulate, and join data.
* [[modeloptimizergui|Model Optimizer]] - Better handling of numeric data in comparison table, additional statistics, and improved handling of include field.
* [[modeloptimizergui|Model Optimizer]] - Better handling of numeric data in comparison table, additional statistics, and improved handling of include field.
 
** Add support for automatically using user-selected model groupings in PLSDA and SVMDA within model optimizer.
 
* Better help integration with newer version of Matlab.
* [[modelselectorgui|Hierarchical Model Builder]] - Add vertical scrolling.
==New Command-line Features and Functions==
==New Command-line Features and Functions==


===Misc New Functions===
===Misc New Functions===


* [[eemoutlier]] - New file for automatically removing outliers in fluorescence PARAFAC models.
* [[mlsca]] Multi-level Simultaneous Component Analysis.
* [[mlsca]] Multi-level Simultaneous Component Analysis.
* [[multiblock]] Create or apply a multiblock model for joining data.
* [[multiblock]] Create or apply a multiblock model for joining data.
Line 50: Line 65:
* [[mdcheck]] - Allow use of KNN as data replacement method (replace missing data with data from sample(s) which are closest).
* [[mdcheck]] - Allow use of KNN as data replacement method (replace missing data with data from sample(s) which are closest).
* [[jmlimit]] - Better handle degenerate cases when multiple confidence levels are requested (return VECTOR of zeros instead of single zero).
* [[jmlimit]] - Better handle degenerate cases when multiple confidence levels are requested (return VECTOR of zeros instead of single zero).
* [[cov_cv]] - Changed from SVDS to SVD.
* [[histaxes]] - Fix for when NaN's are present in data.
* [[als]] - Sort components by variance captured (if no constraints otherwise defining order).
* [[confusionmatrix]] - Fix labeling of confusion matrix (P and F1 swap).
** Standardize terminology: TP = count, TPR = proportion (rate) for confusion matrix quantities, and labels shown. 
* [[comparemodels]] - Report the mean (class count weighted) of Classification Error, Precision, F1 Score for classification models.

Revision as of 17:23, 10 June 2015

Version 8.0 of PLS_Toolbox and Solo was released in June, 2015.

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 in Solo and PLS_Toolbox

Analysis and Models

  • MLSCA - Multi-level simultaneous component analysis.
  • History now captured in history field of Model Object.
    • Add .scoredistance and .esterror as virtual fields for models.
  • ANN now supports custom cross-validation.
  • PLSDA varcap plot now available.
  • Better handle full-rank PCA sub-models where Q residuals are zero in simca.
  • Better handling of errors during variable matching (test to model).

Plotting

  • Improve handling of zoom status in newer versions of Matlab.
  • Better handling of font sizes.
  • Smarter plot style (e.g., scatter vs. bar) assumptions in "automatic" mode.
  • Fix rearranging of controls issue with newer versions of Matlab.
  • ********{REV 19448} add viewcompressgaps option?
  • Connect Classes button has drop-down menu to display connection types.
  • Add logic to NORMALIZE score distance to maximum calibration value.
  • Respect include field when calculating scores limit
  • Show knn score distance limits on plotscores plots.
  • Better handling of crossval when using PLSDA.
    • Convert plsda regression method input to be 'sim' (to speed it up).
    • Recognize when user has passed single-column (either logical or class) and force it to be multi-column logical.

Importers

  • omnicreadr - Reads OMNICix HDF5 image files.
  • Improved importer behavior with mixed length data using matchvars during import.
  • Updated libraries to use new labspec 6 components including 64 bit support.
  • xlsreadr - Add support for joining sheets in rows using matchvars.
    • Add support for using the same graphically-selected parsing options on ALL sheets.

Preprocessing

  • glog Generalized Log Transform added to preprocessing options.
  • pqnorm Probabilistic Quotient Normalization added to preprocessing options.
  • ******{REV r19390} glsw Combine ELS/EMM and EPO options (use "inf" to select all PCs to get old ELS behavior)
  • Add support for handling missing data in both normaliz and mscorr (median only).

Distribution Fitting Toolbox

  • kurtosis: Added kurtosis statistic function.
  • skewness: Added skewness statistic function.

Other Interfaces

  • Multiblock Tool - Interface to view, manipulate, and join data.
  • Model Optimizer - Better handling of numeric data in comparison table, additional statistics, and improved handling of include field.
    • Add support for automatically using user-selected model groupings in PLSDA and SVMDA within model optimizer.
  • Better help integration with newer version of Matlab.
  • Hierarchical Model Builder - Add vertical scrolling.

New Command-line Features and Functions

Misc New Functions

  • eemoutlier - New file for automatically removing outliers in fluorescence PARAFAC models.
  • mlsca Multi-level Simultaneous Component Analysis.
  • multiblock Create or apply a multiblock model for joining data.

Command-Line Changes

  • matchvars - Add logic to speed up some cases of joins.
  • mdcheck - Allow use of KNN as data replacement method (replace missing data with data from sample(s) which are closest).
  • jmlimit - Better handle degenerate cases when multiple confidence levels are requested (return VECTOR of zeros instead of single zero).
  • cov_cv - Changed from SVDS to SVD.
  • histaxes - Fix for when NaN's are present in data.
  • als - Sort components by variance captured (if no constraints otherwise defining order).
  • confusionmatrix - Fix labeling of confusion matrix (P and F1 swap).
    • Standardize terminology: TP = count, TPR = proportion (rate) for confusion matrix quantities, and labels shown.
  • comparemodels - Report the mean (class count weighted) of Classification Error, Precision, F1 Score for classification models.