SerialParam() (fix by
ococrook, see #142).mclust::mclustBICmvtnorm::dmvnorm <2019-06-21 Fri>mrkHClust now uses mrkConsProfiles and returns the hclust
object - see issue #130 for details and background <2018-12-19 Wed>fcol argument to plotDist to plot and colour all profiles
<2018-08-09 Thu>plotEllipse function to visualise and assess TAGM models
<2018-05-18 Fri>dplyr::left_join without attaching dplyr to avoid collision
between Biobase::exprs and dplyr::exprs <2018-04-04 Wed>.nipals method in dimensionality reduction for plot2D (closes
issue #103) <2018-01-16 Tue>Added keepNA argument to goTermToId so that if a GO term becomes obsolete and you cannot replace it with the ID name, you have the option to either replace it with a NA (previous and current default option) or now with keepNA = FALSE the term name will be replaced with the id name <2016-05-25 Wed>
Bump version of all packages that use knitr for vignettes. This is because of an issue (now fixed) in knitr which failed to create purl()'ed R files from vignette sources and include them in the package. This version bump will cause these packages to propagate with those R files included. d.tenenbaum
svmOpt sigma defaults changed from 10^(-2:3) to 10^(-3:2) [2014-08-15 Fri]
in xxxOptimisation, the best parameter(s) for the validation classification runs are now chosen at random instead of using the first best param (see change in pRoloc:::getBestParam that got a sample argument defaulted to TRUE) [2014-08-15 Fri]
When calculating macroF1 scores (xval and validation), NAs are set to 0 (via MLInterfaces:::.macroF1(..., naAs0 = TRUE)). The macro F1 will not be NA (when mean of F1s is calculated) but lowered. This avoids having an NA macro F1 when 1 (or more) classe(s) end(s) up with NA (also set to 0) precision(s) or recall(s) [2014-08-15 Fri]