v0.0.16rc2

New features and improvements

• New accessor dtoolkit.accessor.dataframe.fillna_regression (#556, #567).
• New unique option for dtoolkit.accessor.dataframe.values_to_dict (#548).
• Speed up dtoolkit.util._exception.find_stack_level (#546).
• dtoolkit.accessor.dataframe.filter_in's how only works on condition DataFrame's columns (#545).
• dtoolkit.accessor.series.to_set speeds up especial to large data (#542, #543).
• dtoolkit.accessor.dataframe.drop_inf's inf option supports + and - (#539).
• New accessor dtoolkit.accessor.dataframe.boolean for DataFrame (#537, #538).
• New complement option for dtoolkit.accessor.dataframe.filter_in (#533).
• New Index method dtoolkit.accessor.index.to_set (#529).
• New method dtoolkit.accessor.dataframe.decompose for DataFrame (#488).

API changes

• Add deprecated warning for dtoolkit.transformer.pipeline (#558).
• Split dtoolkit.transformer scripts into sub-pakcages (#557).
• Drop inplace for dtoolkit.accessor.dataframe.drop_inf (#540).
• Drop generic package (#535).
• Drop inplace option of dtoolkit.accessor.dataframe.filter_in (#518, #531, #559).