Umbrella for improvements to RFormula and to how R handles feature and label processing in general
getModelFeatures of ml.api.r.SparkRWrapper cannot (always) reveal the original column names
RFormula output wrong features when formula w/o intercept
SparkR spark.naiveBayes throws error when label is numeric type
Expose ColumnPruner as feature transformer
SparkR formula syntax to turn strings/factors into numerics
OneHotEncoder support drop first category alphabetically in the encoded vector
R MLlib algorithms should support input columns "features" and "label"
Should ML Models contains single models or Pipelines?