- Added
drop = FALSEparameter where needed insidesparsify()to force inputted data.table with 1 row to be returned as a sparse matrix - Fixed bug where
sparsify(dt, sparsifyNAs = FALSE)would error if dt had a factor column of all NAs (sometimes) - Added parameters
sparsifyColsandmemEfficienttosparsify().sparsifyColslet's the user choose a subset of the inputted table's columns to be sparsified andmemEfficient = TRUEwill run a new, memory efficient version of sparsify(). - Added parameter
alphatormsle()andmsle()functions to specify a custom offset instead of the deault (1). - Added parameter
roundbinstobin_data()to allow bin values to be rounded to a specified decimal place - Fixed bug in
empirical_cdf()which occurred when a boundary exactly matched multiple target records
- Added and organize evaluation metrics, mean square error: mse(), mean square logarithmic error: msle(), root mean square error: rmse(), and root mean square logarithmic error: rmsle(). Each metric now includes an option to pass weights and ignore NA values.
- Added unit tests
- Support for multiclass mcc() and user may now provide a confusion matrix as input
- Avoid calling utils::modifyList() on POSIXlt objects
- Added and organize evaluation metrics, mean square error: mse(), mean square logarithmic error: msle(), root mean square error: rmse(), and root mean square logarithmic error: rmsle(). Each metric now includes an option to pass weights and ignore NA values.
- Added unit tests
- In
date_factor(dateVec, ...)fixed bug in "character string is not in a standard unambiguous format" produced by some date values
- In
folds(x, ...)xcan now be a positive integer specifying the number of fold IDs to generate - In
date_factor(dateVec, ...)the argumentfullyearshas been dropped and replaced by the more flexible pair of argumentsminDateandmaxDatefor determining resulting vector levels. Additionally a bug regardingtype=yearquartershas been fixed. - In
bin_data()added some input validation
- Fixed bug in
bin_data()occurring whenxis integer type andbinsincludesInfor-Inf
- Fixed bug in
date_factor()occuring when type = "yearquarter" and fullyears = FALSE - Fixed bugs involving unsorted or duplicate ubounds in
empirical_cdf()
- Added
date_factor()for converting dates to a grouped ordered factor (e.g. months, yearmonths, yearquarters, etc.) - Implemented stratification for numeric
xinfolds(x, stratified=TRUE, ...)
Initial Release