SlidingWindowSplitter#

class SlidingWindowSplitter(fh: Union[int, list, numpy.ndarray, pandas.core.indexes.base.Index, sktime.forecasting.base._fh.ForecastingHorizon] = 1, window_length: Union[int, float, pandas._libs.tslibs.timedeltas.Timedelta, datetime.timedelta, numpy.timedelta64, pandas._libs.tslibs.offsets.DateOffset] = 10, step_length: Union[int, pandas._libs.tslibs.timedeltas.Timedelta, datetime.timedelta, numpy.timedelta64, pandas._libs.tslibs.offsets.DateOffset] = 1, initial_window: Optional[Union[int, float, pandas._libs.tslibs.timedeltas.Timedelta, datetime.timedelta, numpy.timedelta64, pandas._libs.tslibs.offsets.DateOffset]] = None, start_with_window: bool = True)[source]#

Sliding window splitter.

Split time series repeatedly into a fixed-length training and test set.

For example for window_length = 5, step_length = 1 and fh = 3 here is a representation of the folds:

|-----------------------|
| * * * * * x x x - - - |
| - * * * * * x x x - - |
| - - * * * * * x x x - |
| - - - * * * * * x x x |

* = training fold.

x = test fold.

Parameters

fhint, list or np.array

Forecasting horizon

window_lengthint or timedelta or pd.DateOffset

Window length

step_lengthint or timedelta or pd.DateOffset, optional (default=1)

Step length between windows

initial_windowint or timedelta or pd.DateOffset, optional (default=None)

Window length of first window

start_with_windowbool, optional (default=False)

If True, starts with full window.
If False, starts with empty window.

Methods

`get_cutoffs`([y])	Return the cutoff points in .iloc[] context.
`get_fh`()	Return the forecasting horizon.
`get_n_splits`([y])	Return the number of splits.
`split`(y)	Split y into training and test windows.

get_cutoffs(y: Optional[Union[pandas.core.series.Series, pandas.core.frame.DataFrame, numpy.ndarray, pandas.core.indexes.base.Index]] = None) → numpy.ndarray[source]#

Return the cutoff points in .iloc[] context.

Parameters

ypd.Series or pd.Index, optional (default=None): Time series to split

Returns

cutoffs1D np.ndarray of int: iloc location indices, in reference to y, of cutoff indices

get_fh() → sktime.forecasting.base._fh.ForecastingHorizon[source]#

Return the forecasting horizon.

Returns

fhForecastingHorizon: The forecasting horizon

get_n_splits(y: Optional[Union[pandas.core.series.Series, pandas.core.frame.DataFrame, numpy.ndarray, pandas.core.indexes.base.Index]] = None) → int[source]#

Return the number of splits.

Parameters

ypd.Series or pd.Index, optional (default=None): Time series to split

Returns

n_splitsint: The number of splits.

split(y: Union[pandas.core.series.Series, pandas.core.frame.DataFrame, numpy.ndarray, pandas.core.indexes.base.Index]) → Generator[Tuple[numpy.ndarray, numpy.ndarray], None, None][source]#

Split y into training and test windows.

Parameters

ypd.Series or pd.Index: Time series to split

Yields

trainnp.ndarray: Training window indices
testnp.ndarray: Test window indices