tidyselect
is introduced in a few verbs (methods) which supportstart_with
,ends_with
andcontains
.summarise
now supports returning an iterable within a list as a output, no longer restricts the output to a scalar.
- pivot_wider gains a new argument
id_expand
(False by default)
- slice_max and slice_min are rewritten to avoid pandas apply and using pandas.groupbBy.Series.rank, thereby being fast for larger datasets.
- tidyframe drops row indexes when constructed using a pandas dataframe
- User is expected to run tidypandas.tidy_utils.simplify() beforehand to handle complicated pandas dataframe input
- bugfix: show proper error message when tidyframe cannot be created from pandas dataframe with some column starting with underscore
- bugfix: tidyframe.separate now handles NAs
- minor bug in _is_kwargable is fixed, making mutate better
- minor bug in filter with mask is fixed
- CI is enabled with github actions
- glimpse by default shows 100 columns
- add_row_number is more efficient
- show method added (print specified number of rows)
- glimpse method added (to see dataframe in horizontal form)
- add_row_number is fast in grouped case (use of .cumcount)
- All changes of github only release v0.1.5
- show method added (print specified number of rows)
- glimpse method added (to see dataframe in horizontal form)
- add_row_number is fast in grouped case (use of .cumcount)
- All changes of github only release v0.1.5
- methods 'expand' and 'complete' are implemented
- 'rename' method gains arguments 'predicate' and 'func'
- 'pivot_wider' method gains argument 'names_prefix'
- 'skim' method now provides a warning when skimpy is not installed
- utility 'expand_grid' is implemented
- init now does not coerce column classes
- repr set right for dataframes with pandas version <= 1.4.0
- tidyframe init now accepts inputs and passes them to pd.DataFrame
- Along with accepting a pandas dataframe
- minor bugfix where
skimpy
is now optional!
- Patch for First release of
tidypandas
!
- First release of
tidypandas
!