Report for a single estimator

The class EstimatorReport provides a report allowing to inspect and evaluate a scikit-learn estimator in an interactive way. The functionalities of the report are accessible through accessors.

EstimatorReport(estimator, *[, fit, ...])

Report for a fitted estimator.

Methods

EstimatorReport.help()	Display report help using rich or HTML.
EstimatorReport.diagnose(*[, ignore])	Run checks and return a diagnostic with detected issues.
EstimatorReport.add_checks(checks)	Register additional diagnostic checks for this report.
EstimatorReport.cache_predictions([data_source])	Cache estimator's predictions.
EstimatorReport.clear_cache()	Clear the cache.
EstimatorReport.get_predictions(*, data_source)	Get estimator's predictions.

Accessors

EstimatorReport.data	The data accessor helps you to get insights about the train and test datasets.
EstimatorReport.inspection	Accessor for model inspection related operations.
EstimatorReport.metrics	Accessor for metrics-related operations.

Data

The data accessor helps you to get insights about the dataset used to train and test your estimator.

EstimatorReport.data.help()	Display accessor help using rich or HTML.
EstimatorReport.data.analyze([data_source, ...])	Plot dataset statistics.

Metrics

The metrics accessor helps you to evaluate the statistical performance of your estimator.

EstimatorReport.metrics.help()	Display accessor help using rich or HTML.
EstimatorReport.metrics.summarize(*[, ...])	Report a set of metrics for our estimator.
EstimatorReport.metrics.accuracy(*[, ...])	Compute the accuracy score.
EstimatorReport.metrics.add(metric, *[, ...])	Add a custom metric to be included in summarize() by default.
EstimatorReport.metrics.available()	List available metric names in the registry.
EstimatorReport.metrics.brier_score(*[, ...])	Compute the Brier score.
EstimatorReport.metrics.confusion_matrix(*)	Plot the confusion matrix.
EstimatorReport.metrics.fit_time([cast])	Get time to fit the estimator.
EstimatorReport.metrics.log_loss(*[, ...])	Compute the log loss.
EstimatorReport.metrics.mae(*[, ...])	Compute the mean absolute error.
EstimatorReport.metrics.mape(*[, ...])	Compute the mean absolute percentage error.
EstimatorReport.metrics.precision(*[, ...])	Compute the precision score.
EstimatorReport.metrics.precision_recall(*)	Plot the precision-recall curve.
EstimatorReport.metrics.predict_time(*[, ...])	Get prediction time if it has been already measured.
EstimatorReport.metrics.prediction_error(*)	Plot the prediction error of a regression model.
EstimatorReport.metrics.r2(*[, data_source, ...])	Compute the R² score.
EstimatorReport.metrics.recall(*[, ...])	Compute the recall score.
EstimatorReport.metrics.remove(name)	Remove a metric from the registry.
EstimatorReport.metrics.rmse(*[, ...])	Compute the root mean squared error.
EstimatorReport.metrics.roc(*[, data_source])	Plot the ROC curve.
EstimatorReport.metrics.roc_auc(*[, ...])	Compute the ROC AUC score.
EstimatorReport.metrics.timings()	Get all measured processing times related to the estimator.

Inspection

The inspection accessor helps you inspect your model by e.g. evaluating the importance of the features in your model.

EstimatorReport.inspection.help()	Display accessor help using rich or HTML.
EstimatorReport.inspection.coefficients()	Retrieve the coefficients of a linear model, including the intercept.
EstimatorReport.inspection.impurity_decrease()	Retrieve the Mean Decrease in Impurity (MDI) of a tree-based model.
EstimatorReport.inspection.permutation_importance(*)	Display the permutation feature importance.