The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.
Documentation is refactored using Roxygen2 and considerably enhanced.
All camelCase function names now have their equivalence in
snake_case, e.g., mlRforest ->
ml_rforest(), or confusionImage() ->
confusion_image() in order to adapt to the coding
preferences of the user.
mlRpart() function implements
rpart::rpart() for using decision trees.The description is extended.
A {pkgdown} site is added.
mlKnn() is implemented for K-nearest
neighbors.
Several adjustments were required for compatibility with R 4.2.0 (it is not allowed any more to use vectors > 1 with || and &&).
predict() was applied to an mlearning object build
with full formula (not the short one var ~ .), if the
dependent variable was not in newdata =, an error message
was raised (although this variable is not necessary at this point). Bug
identified by Damien Dumont, and corrected.mlSvm.formula(), arguments scale=,
type=, kernel= and classwt= were
not correctly used. Corrected.mlLvq() providing size = or
prior = led to an lvq object not found
message. Corrected.Sometimes, data was not found (e.g., when called inside a {learnr} tutorial).
In mlearning(), data is forced with
as.data.frame() (tibbles are not supported
internally).
In the mlXXX() function, it was not possible to
indicate something like mlLda(data = iris, Species ~ .).
Solved by adding train = argument in
mlXXX().
In summary.confusion() produced an error if more
than one type = was provided.
These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.