pyspark.pandas.window.Rolling.quantile#
- Rolling.quantile(quantile, accuracy=10000)[source]#
Calculate the rolling quantile of the values.
New in version 3.4.0.
- Parameters
- quantilefloat
Value between 0 and 1 providing the quantile to compute.
Deprecated since version 4.0.0: This will be renamed to ‘q’ in a future version.
- accuracyint, optional
Default accuracy of approximation. Larger value means better accuracy. The relative error can be deduced by 1.0 / accuracy. This is a panda-on-Spark specific parameter.
- Returns
- Series or DataFrame
Returned object type is determined by the caller of the rolling calculation.
See also
pyspark.pandas.Series.rollingCalling rolling with Series data.
pyspark.pandas.DataFrame.rollingCalling rolling with DataFrames.
pyspark.pandas.Series.quantileAggregating quantile for Series.
pyspark.pandas.DataFrame.quantileAggregating quantile for DataFrame.
Notes
quantile in pandas-on-Spark are using distributed percentile approximation algorithm unlike pandas, the result might be different with pandas, also interpolation parameter is not supported yet.
the current implementation of this API uses Spark’s Window without specifying partition specification. This leads to move all data into single partition in single machine and could cause serious performance degradation. Avoid this method against very large dataset.
Examples
>>> s = ps.Series([4, 3, 5, 2, 6]) >>> s 0 4 1 3 2 5 3 2 4 6 dtype: int64
>>> s.rolling(2).quantile(0.5) 0 NaN 1 3.0 2 3.0 3 2.0 4 2.0 dtype: float64
>>> s.rolling(3).quantile(0.5) 0 NaN 1 NaN 2 4.0 3 3.0 4 5.0 dtype: float64
For DataFrame, each rolling quantile is computed column-wise.
>>> df = ps.DataFrame({"A": s.to_numpy(), "B": s.to_numpy() ** 2}) >>> df A B 0 4 16 1 3 9 2 5 25 3 2 4 4 6 36
>>> df.rolling(2).quantile(0.5) A B 0 NaN NaN 1 3.0 9.0 2 3.0 9.0 3 2.0 4.0 4 2.0 4.0
>>> df.rolling(3).quantile(0.5) A B 0 NaN NaN 1 NaN NaN 2 4.0 16.0 3 3.0 9.0 4 5.0 25.0