pyspark.pandas.Series.sort_index¶
-
Series.sort_index(axis: Union[int, str] = 0, level: Union[int, List[int], None] = None, ascending: bool = True, inplace: bool = False, kind: str = None, na_position: str = 'last', ignore_index: bool = False) → Optional[pyspark.pandas.series.Series][source]¶ Sort object by labels (along an axis)
- Parameters
- axisindex, columns to direct sorting. Currently, only axis = 0 is supported.
- levelint or level name or list of ints or list of level names
if not None, sort on values in specified index level(s)
- ascendingboolean, default True
Sort ascending vs. descending
- inplacebool, default False
if True, perform operation in-place
- kindstr, default None
pandas-on-Spark does not allow specifying the sorting algorithm now, default None
- na_position{‘first’, ‘last’}, default ‘last’
first puts NaNs at the beginning, last puts NaNs at the end. Not implemented for MultiIndex.
- ignore_indexbool, default False
If True, the resulting axis will be labeled 0, 1, …, n - 1.
New in version 3.4.0.
- Returns
- sorted_objSeries
Examples
>>> s = ps.Series([2, 1, np.nan], index=['b', 'a', np.nan])
>>> s.sort_index() a 1.0 b 2.0 None NaN dtype: float64
>>> s.sort_index(ignore_index=True) 0 1.0 1 2.0 2 NaN dtype: float64
>>> s.sort_index(ascending=False) b 2.0 a 1.0 None NaN dtype: float64
>>> s.sort_index(na_position='first') None NaN a 1.0 b 2.0 dtype: float64
>>> s.sort_index(inplace=True) >>> s a 1.0 b 2.0 None NaN dtype: float64
Multi-index series.
>>> s = ps.Series(range(4), index=[['b', 'b', 'a', 'a'], [1, 0, 1, 0]], name='0')
>>> s.sort_index() a 0 3 1 2 b 0 1 1 0 Name: 0, dtype: int64
>>> s.sort_index(level=1) a 0 3 b 0 1 a 1 2 b 1 0 Name: 0, dtype: int64
>>> s.sort_index(level=[1, 0]) a 0 3 b 0 1 a 1 2 b 1 0 Name: 0, dtype: int64