v0.2.5

New Features

• feat: add forecast bias metric @jmoralez (#122)
• feat: support dask and spark dataframes in evaluate @jmoralez (#121)

v0.2.4

New Features

• feat: add future_exog_to_historic function @jmoralez (#119)

v0.2.3

Bug Fixes

• support single ax object in plot_series @jmoralez (#116)
• generate ranges with time_unit for polars in fill_gaps @jmoralez (#112)

Enhancement

• use TypeVar to get correct output type @jmoralez (#117)

v0.2.2

Enhancement

• support returning multiple time features from function @jmoralez (#111)

v0.2.1

New Features

• add time_features to feature engineering @jmoralez (#108)

Enhancement

• accept plot object in plot_series @jmoralez (#109)
• use utc to handle timezones in fill_gaps @jmoralez (#107)

v0.2.0

New Features

• add agg_fn to evaluate @jmoralez (#99)

Breaking Change

• remove target transforms @jmoralez (#90)
• use single baseline in rmae @jmoralez (#105)
• use current stylesheet's color palette by default @elephaint (#92)

Bug Fixes

• drop na for pandas in counts_by_id @jmoralez (#106)

Enhancement

• make df optional in plot_series @jmoralez (#98)

v0.1.12

Enhancement

• support lists in assign_columns @jmoralez (#94)
• add ensure_sorted and ProcessedDF named tuple @jmoralez (#93)
• ignore zeros on mape calculation @jmoralez (#89)

v0.1.11

Bug Fixes

• support timezones for pandas in fill_gaps @jmoralez (#85)

v0.1.10

Bug Fixes

• fix timestamp check in time_ranges for timestamps with timezone @jmoralez (#84)

v0.1.9

Enhancement

• faster counts_by_id for pandas @jmoralez (#83)