timings

ComparisonReport.metrics.timings(aggregate=('mean', 'std'))

Get all measured processing times related to the different estimators.

The index of the returned dataframe is the name of the processing time. When the estimators were not used to predict, no timings regarding the prediction will be present.

Parameters:

aggregate{“mean”, “std”}, list of such str or None, default=(“mean”, “std”)

Function to aggregate the scores across the cross-validation splits. None will return the scores for each split. Ignored when comparison is between EstimatorReport instances

Returns:

pd.DataFrame

A dataframe with the processing times.

Examples

>>> from sklearn.datasets import make_classification
>>> from sklearn.linear_model import LogisticRegression
>>> from skore import evaluate
>>> X, y = make_classification(random_state=42)
>>> estimator_1 = LogisticRegression()
>>> estimator_2 = LogisticRegression(C=2)  # Different regularization
>>> report = evaluate([estimator_1, estimator_2], X, y, splitter=0.2)
>>> report.metrics.timings()
                LogisticRegression_1    LogisticRegression_2
Fit time (s)                     ...                     ...
>>> report.cache_predictions(response_methods=["predict"])
>>> report.metrics.timings()
                        LogisticRegression_1    LogisticRegression_2
Fit time (s)                             ...                     ...
Predict time test (s)                    ...                     ...
Predict time train (s)                   ...                     ...