stream 2.0-0 (09/01/22)

New Features

We now support %>% and defining pipelines with the new class DST_Runner.
New base class DSF for data stream filters with implementations for
DSF_Downsample, DSF_Convolve, DSF_Func, DSF_dplyr
New DST_Multi to run multiple tasks on a stream.
New DSD_NULL
New DSOutlier: DSOutlier_DBSTREAM, DSOutlier_DStream
DSD_ReadCSV gained parameter col.names
Added DSD_ReadStream as an alias for DSD_ReadCSV.
Added DST_WriteStream to write streams using update().
Added DSD_Mixture to combine streams.
update() now returns information like cluster assignments as a data.frame.
DSC implementations now have a formula argument to decide what variables should be used for
clustering.
DSD_ReadDB has now close_stream().
DSD_Memory, DSD_ReadDB and DSD_ReadStream have now a parameter called outofpoints to handle
the situation that get_points requests more points than available.

stream now implements a standard predict function (get_assignment() is now deprecated).
class information and extra information are now stored as columns starting with '.'
instead of as attributes. See get_points().
DSD_ScaleStream is now DSF_Scale and DSD_ScaleStream is deprecated.
DSO is now called DSAggregate.
The NAMESPACE is now managed using roxygen.
evaluate is now a generic.
noise and outliers are now the same concept. DS_Gaussian can make sure that noise points are
separated from clusters.
evaluate() is now called evaluate_static() and evaluate_cluster is evaluate_stream(), both are now generics.
plot for DSC now automatically finds micro and macro-clusters to plot.