cinnamon.plot.drift.plot_drift.plot_feature_drift

cinnamon.plot.drift.plot_drift.plot_feature_drift(drift_explainer: AbstractDriftExplainer, feature: Union[int, str], max_n_cat: int = 20, figsize: Tuple[int, int] = (7, 5), as_discrete: bool = False, bins=10, legend_labels: Tuple[str, str] = ('Dataset 1', 'Dataset 2'))

Plot distributions of a given feature in order to visualize a potential data drift of this feature.

Parameters

drift_explainer: AbstractDriftExplainer

A AbstractDriftExplainer object.

featureUnion[int, str]

Either the column index or the name of the feature.

max_n_catint (default=20)

Maximum number of classes to represent on the plot (used only for categorical feature (not supported currently) or if as_discrete == True

binsint or sequence of scalars or str, optional (default=10)

For regression only. ‘two_heads’ corresponds to a number of bins which is the minimum of of the optimal number of bins for dataset 1 and dataset 2 taken separately. Other value of “bins” parameter passed to matplotlib.pyplot.hist function are also accepted.

figsizeTuple[int, int] (default=(7, 5))

Graphic size passed to matplotlib

as_discrete: bool (default=False)

If a numerical feature is discrete (has few unique values), consider it discrete to make the plot.

legend_labelsTuple[str, str] (default=(‘Dataset 1’, ‘Dataset 2’))

Legend labels used for dataset 1 and dataset 2

Returns

None