Release Notes Version 8 9

From Eigenvector Research Documentation Wiki
Revision as of 12:42, 15 September 2020 by Scott (talk | contribs) (Other Changes)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

Changes and Bug Fixes in Version 8.9

Version 8.9 of PLS_Toolbox and Solo was released in September, 2020.

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

  • LREG-DA - Logistic Regression for classification.
  • SIMCA - Allow ability to cross-validate individual models in interface.
  • CROSS-VALIDATION - Allow a class set to be used for custom cross validation.
  • PLSDA
    • Allow creating of new class set based on modeling groups.
    • Enable Variable Selection methods VIP and sRatios.
  • ASCA - Uses the ASCA+ extension as introduced by Thiel, Feraud, and Govaerts (2017) to handle an un-balanced experimental design dataset.
  • PLOTGUI
    • Allow for creating a new class set via selection when viewing an existing class set.
    • Allow for override of default class coloring using 'classcoloruser' setting (set to "yes") then selecting context menu on marker.
  • PREPROCESS - Add Filtering and Despiking Settings GUI, which uses the windowfilter function.

Other Changes

File Comment
editds
  • Add menu items for flipping data.
  • Added Save Selected Indices and Load Selected Indices options under File menu.
analysis
  • Allow reset of included data (via xblock/yblock context menu).
anglemapper
  • Classification based on angle measures between signals.
xgb/xgbda
  • Add menu item to allow setting of include based on top N variables.
evrimodel
  • Add default naming to componentnames so components are unique by default.
visionairxmlreadr
  • Update java library.
doerunsheet
  • Increase font size.
anovadoe
  • Normalize sums of squares so that they the sum over all factors and error sum to 100% (as in Thiel et al (2017)).
selectvars
  • Extended to work when calibration Y-block is a multi-column class logical dataset or array.
pls
  • Fix PLS Y-Block Scores calculation for LVs 2 and higher, to normalize over included data only.
wtfa
  • Fix, do not take square root of rho.