PredictionErrorDisplay

class skore.PredictionErrorDisplay(*, y_true, y_pred, residuals, range_y_true, range_y_pred, range_residuals, estimator_names, data_source, ml_task, report_type)

Visualization of the prediction error of a regression model.

This tool can display “residuals vs predicted” or “actual vs predicted” using scatter plots to qualitatively assess the behavior of a regressor, preferably on held-out data points.

An instance of this class is should created by EstimatorReport.metrics.prediction_error(). You should not create an instance of this class directly.

Parameters:

y_truelist of ndarray of shape (n_samples,)

True values.

y_predlist of ndarray of shape (n_samples,)

Prediction values.

residualslist of ndarray of shape (n_samples,)

Residuals. Equal to y_true - y_pred.

range_y_trueRangeData

Global range of the true values.

range_y_predRangeData

Global range of the predicted values.

range_residualsRangeData

Global range of the residuals.

estimator_nameslist of str

Name of the estimators.

data_source{“train”, “test”, “X_y”}

The data source used to display the prediction error.

ml_task{“regression”, “multioutput-regression”}

The machine learning task.

report_type{“cross-validation”, “estimator”, “comparison-estimator”}

The type of report.

Attributes:

line_matplotlib Artist

Optimal line representing y_true == y_pred. Therefore, it is a diagonal line for kind="predictions" and a horizontal line for kind="residuals".

errors_lines_matplotlib Artist or None

Residual lines. If with_errors=False, then it is set to None.

scatter_list of matplotlib Artist

Scatter data points.

ax_matplotlib Axes

Axes with the different matplotlib axis.

figure_matplotlib Figure

Figure containing the scatter and lines.

Examples

>>> from sklearn.datasets import load_diabetes
>>> from sklearn.linear_model import Ridge
>>> from sklearn.model_selection import train_test_split
>>> from skore import EstimatorReport
>>> X_train, X_test, y_train, y_test = train_test_split(
...     *load_diabetes(return_X_y=True), random_state=0
... )
>>> classifier = Ridge()
>>> report = EstimatorReport(
...     classifier,
...     X_train=X_train,
...     y_train=y_train,
...     X_test=X_test,
...     y_test=y_test,
... )
>>> display = report.metrics.prediction_error()
>>> display.plot(kind="actual_vs_predicted")

help()

Display available attributes and methods using rich.

plot(ax=None, *, estimator_name=None, kind='residual_vs_predicted', data_points_kwargs=None, perfect_model_kwargs=None, despine=True)

Plot visualization.

Extra keyword arguments will be passed to matplotlib’s plot.

Parameters:

axmatplotlib axes, default=None

Axes object to plot on. If None, a new figure and axes is created.

estimator_namestr

Name of the estimator used to plot the prediction error. If None, we used the inferred name from the estimator.

kind{“actual_vs_predicted”, “residual_vs_predicted”}, default=”residual_vs_predicted”

The type of plot to draw:

• “actual_vs_predicted” draws the observed values (y-axis) vs. the predicted values (x-axis).

• “residual_vs_predicted” draws the residuals, i.e. difference between observed and predicted values, (y-axis) vs. the predicted values (x-axis).

data_points_kwargsdict, default=None

Dictionary with keywords passed to the matplotlib.pyplot.scatter call.

perfect_model_kwargsdict, default=None

Dictionary with keyword passed to the matplotlib.pyplot.plot call to draw the optimal line.

despinebool, default=True

Whether to remove the top and right spines from the plot.

Examples

>>> from sklearn.datasets import load_diabetes
>>> from sklearn.linear_model import Ridge
>>> from sklearn.model_selection import train_test_split
>>> from skore import EstimatorReport
>>> X_train, X_test, y_train, y_test = train_test_split(
...     *load_diabetes(return_X_y=True), random_state=0
... )
>>> classifier = Ridge()
>>> report = EstimatorReport(
...     classifier,
...     X_train=X_train,
...     y_train=y_train,
...     X_test=X_test,
...     y_test=y_test,
... )
>>> display = report.metrics.prediction_error()
>>> display.plot(kind="actual_vs_predicted")

set_style(**kwargs)

Set the style parameters for the display.

Parameters:

**kwargsdict

Style parameters to set. Each parameter name should correspond to a a style attribute passed to the plot method of the display.

Returns:

selfobject

Returns the instance itself.

Raises:

ValueError

If a style parameter is unknown.

static style_plot(plot_func)

Apply consistent style to skore displays.

This decorator: 1. Applies default style settings 2. Executes plot_func 3. Applies tight_layout

Parameters:

plot_funccallable

The plot function to be decorated.

Returns:

callable

The decorated plot function.