Release Notes PLS Toolbox Version 5 2 2: Difference between revisions

From Eigenvector Research Documentation Wiki
Jump to navigation Jump to search
imported>Scott
(New page: ==Changes and Bug Fixes in Version 5.2.2== ===Bug Fixes===)
 
imported>Scott
No edit summary
 
(One intermediate revision by the same user not shown)
Line 3: Line 3:


===Bug Fixes===
===Bug Fixes===
{|
|analysis
| Fix plot for preprocessed validation data.
|---
|mdcheck
| Fixed bug which would give error when viewing missing data map with excluded data on DSO.
|---
|parafac
| Fix for analysis of 2-way data.
|---
|plotscores_pls
| Show leverage even when showing test data.
|---
|plotscoreslimits
| Fix for bug when y-block includes any column above column #99.<br>Fix for bug when applying an old (pre 5.0) model to new data and viewing a scores plot.
|---
|plsda
| Fix bug involving test-set predictions with grouped classes (didn't correctly group classes in test data, thus leading to incorrect prediction statistics for misclassification and sensitivity/specificity).
|---
|plsdaroc
| Fix for bad threshold curves (e.g. NaN because all predicted values were the same).
|---
|xclreadr
| Better handle errors thrown during multi-file read. <br>Fix for label bug when first and second files read do not contain the same number of rows (and we're concatenating rows).
|----
|}

Latest revision as of 09:27, 11 June 2009

Changes and Bug Fixes in Version 5.2.2

Bug Fixes

analysis Fix plot for preprocessed validation data.
mdcheck Fixed bug which would give error when viewing missing data map with excluded data on DSO.
parafac Fix for analysis of 2-way data.
plotscores_pls Show leverage even when showing test data.
plotscoreslimits Fix for bug when y-block includes any column above column #99.
Fix for bug when applying an old (pre 5.0) model to new data and viewing a scores plot.
plsda Fix bug involving test-set predictions with grouped classes (didn't correctly group classes in test data, thus leading to incorrect prediction statistics for misclassification and sensitivity/specificity).
plsdaroc Fix for bad threshold curves (e.g. NaN because all predicted values were the same).
xclreadr Better handle errors thrown during multi-file read.
Fix for label bug when first and second files read do not contain the same number of rows (and we're concatenating rows).