plot_windows#

plot_windows(cv, y, title='', ax=None)[source]#

Plot training and test windows.

Plots the training and test windows for each split of a time series, subject to an sktime time series splitter.

x-axis: time, ranging from start to end of y

y-axis: window number, starting at 0

plot elements: training split (orange) and test split (blue)

dots indicate index in the training or test split will be plotted on top of each other if train/test split is not disjoint

Parameters:

ypd.Series: Time series to split
cvsktime splitter object, descendant of BaseSplitter: Time series splitter, e.g., temporal cross-validation iterator
titlestr: Plot title
axmatplotlib.axes.Axes, optional (default=None): Axes on which to plot. If None, axes will be created and returned.

Returns:

figmatplotlib.figure.Figure, returned only if ax is None: matplotlib figure object
axmatplotlib.axes.Axes: matplotlib axes object with the figure

Examples

>>> from sktime.split import ExpandingWindowSplitter
>>> from sktime.utils.plotting import plot_windows
>>> from sktime.datasets import load_airline
>>> import numpy as np

>>> fh = np.arange(1, 13)
>>> cv = ExpandingWindowSplitter(step_length=1, fh=fh, initial_window=24)
>>> y = load_airline()
>>> plot_windows(cv, y.iloc[:50])  