elphick.sklearn_viz.model_selection.learning_curve.LearningCurve

class elphick.sklearn_viz.model_selection.learning_curve.LearningCurve(estimator, x, y, train_sizes=array([0.1, 0.325, 0.55, 0.775, 1.]), cv=5, metrics=None, scorer=None, random_state=42, n_jobs=1)[source]
__init__(estimator, x, y, train_sizes=array([0.1, 0.325, 0.55, 0.775, 1.]), cv=5, metrics=None, scorer=None, random_state=42, n_jobs=1)[source]
Parameters:

  • estimator – The scikit-learn model or pipeline.

  • x (DataFrame) – X values provided to calculate the learning curve.

  • y (Union[DataFrame, Series]) – y values provided to calculate the learning curve.

  • train_sizes (Iterable) – list of training sample counts (or fractions if < 1)

  • cv (Union[int, Any]) – The number of cross validation folds or a cv callable.

  • metrics (Optional[dict[str, Callable]]) – Optional Dict of callable metrics to calculate post-fitting

  • scorer (Optional[Any]) – The scoring method. If None, ‘accuracy’ is used for classifiers and ‘r2’ for regressors.

  • random_state (int) – Optional random seed

  • n_jobs (int) – Number of parallel jobs to run. If -1, then the number of jobs is set to the number of CPU cores. Recommend setting to -2 for large jobs to retain a core for system interaction.

  • verbosity – Verbosity level. 0 = silent, 1 = overall (start/finish), 2 = each cross-validation.

Methods

__init__(estimator, x, y[, train_sizes, cv, ...])

type estimator:

calculate_grid_and_subplot_order(...)

calculate_learning_curve()

rtype:

LearningCurveResult

custom_learning_curve()

rtype:

LearningCurveResult

plot([title, metrics, col_wrap, plot_scorer])

Create the plot

Attributes

n_cores

results

plot(title=None, metrics=None, col_wrap=1, plot_scorer=True)[source]

Create the plot

Parameters:

  • title (Optional[str]) – title for the plot

  • metrics (Optional[list[str]]) – Optional list of metric keys to plot

  • col_wrap (int) – The number of columns to use for the facet grid if plotting metrics.

  • plot_scorer (bool) – If True, plot the scorer. Use False to plot only the metrics.

Return type:

Figure

Returns:

a plotly GraphObjects.Figure