feat(ComparisonReport[CVReport]): Add ROC curve

skore/src/skore/sklearn/_comparison/metrics_accessor.py

@@ -1263,9 +1263,6 @@ def _get_display(
         display : display_class
             The display.
         """
-        if self._parent._reports_type == "CrossValidationReport":
-            raise NotImplementedError()
-
         if "seed" in display_kwargs and display_kwargs["seed"] is None:
             cache_key = None
         else:
@@ -1288,55 +1285,121 @@ def _get_display(
             y_true: list[YPlotData] = []
             y_pred: list[YPlotData] = []
 
-            for report, report_name in zip(
-                self._parent.reports_, self._parent.report_names_
-            ):
-                report_X, report_y, _ = report.metrics._get_X_y_and_data_source_hash(
-                    data_source=data_source,
-                    X=X,
-                    y=y,
-                )
+            if self._parent._reports_type == "EstimatorReport":
+                for report, report_name in zip(
+                    self._parent.reports_, self._parent.report_names_
+                ):
+                    report_X, report_y, _ = (
+                        report.metrics._get_X_y_and_data_source_hash(
+                            data_source=data_source,
+                            X=X,
+                            y=y,
+                        )
+                    )
 
-                y_true.append(
-                    YPlotData(
-                        estimator_name=report_name,
-                        split_index=None,
-                        y=report_y,
+                    y_true.append(
+                        YPlotData(
+                            estimator_name=report_name,
+                            split_index=None,
+                            y=report_y,
+                        )
                     )
-                )
-                results = _get_cached_response_values(
-                    cache=report._cache,
-                    estimator_hash=report._hash,
-                    estimator=report._estimator,
-                    X=report_X,
-                    response_method=response_method,
+                    results = _get_cached_response_values(
+                        cache=report._cache,
+                        estimator_hash=report._hash,
+                        estimator=report._estimator,
+                        X=report_X,
+                        response_method=response_method,
+                        data_source=data_source,
+                        data_source_hash=None,
+                        pos_label=display_kwargs.get("pos_label"),
+                    )
+                    for key, value, is_cached in results:
+                        if not is_cached:
+                            report._cache[key] = value
+                        if key[-1] != "predict_time":
+                            y_pred.append(
+                                YPlotData(
+                                    estimator_name=report_name,
+                                    split_index=None,
+                                    y=value,
+                                )
+                            )
+
+                    progress.update(main_task, advance=1, refresh=True)
+
+                display = display_class._compute_data_for_display(
+                    y_true=y_true,
+                    y_pred=y_pred,
+                    report_type="comparison-estimator",
+                    estimators=[report.estimator_ for report in self._parent.reports_],
+                    estimator_names=self._parent.report_names_,
+                    ml_task=self._parent._ml_task,
                     data_source=data_source,
-                    data_source_hash=None,
-                    pos_label=display_kwargs.get("pos_label"),
+                    **display_kwargs,
                 )
-                for key, value, is_cached in results:
-                    if not is_cached:
-                        report._cache[key] = value
-                    if key[-1] != "predict_time":
-                        y_pred.append(
+
+            else:
+                for report, report_name in zip(
+                    self._parent.reports_, self._parent.report_names_
+                ):
+                    for split_index, estimator_report in enumerate(
+                        report.estimator_reports_
+                    ):
+                        report_X, report_y, _ = (
+                            estimator_report.metrics._get_X_y_and_data_source_hash(
+                                data_source=data_source,
+                                X=X,
+                                y=y,
+                            )
+                        )
+
+                        y_true.append(
                             YPlotData(
                                 estimator_name=report_name,
-                                split_index=None,
-                                y=value,
+                                split_index=split_index,
+                                y=report_y,
                             )
                         )
-                progress.update(main_task, advance=1, refresh=True)
 
-            display = display_class._compute_data_for_display(
-                y_true=y_true,
-                y_pred=y_pred,
-                report_type="comparison-estimator",
-                estimators=[report.estimator_ for report in self._parent.reports_],
-                estimator_names=self._parent.report_names_,
-                ml_task=self._parent._ml_task,
-                data_source=data_source,
-                **display_kwargs,
-            )
+                        results = _get_cached_response_values(
+                            cache=estimator_report._cache,
+                            estimator_hash=estimator_report._hash,
+                            estimator=estimator_report.estimator_,
+                            X=report_X,
+                            response_method=response_method,
+                            data_source=data_source,
+                            data_source_hash=None,
+                            pos_label=display_kwargs.get("pos_label"),
+                        )
+                        for key, value, is_cached in results:
+                            if not is_cached:
+                                report._cache[key] = value
+                            if key[-1] != "predict_time":
+                                y_pred.append(
+                                    YPlotData(
+                                        estimator_name=report_name,
+                                        split_index=split_index,
+                                        y=value,
+                                    )
+                                )
+
+                    progress.update(main_task, advance=1, refresh=True)
+
+                display = display_class._compute_data_for_display(
+                    y_true=y_true,
+                    y_pred=y_pred,
+                    report_type="comparison-cross-validation",
+                    estimators=[
+                        estimator_report.estimator_
+                        for report in self._parent.reports_
+                        for estimator_report in report.estimator_reports_
+                    ],
+                    estimator_names=self._parent.report_names_,
+                    ml_task=self._parent._ml_task,
+                    data_source=data_source,
+                    **display_kwargs,
+                )
 
             if cache_key is not None:
                 # Unless seed is an int (i.e. the call is deterministic),
@@ -1476,6 +1539,8 @@ def precision_recall(
         >>> display = comparison_report.metrics.precision_recall()
         >>> display.plot()
         """
+        if self._parent._reports_type == "CrossValidationReport":
+            raise NotImplementedError()
         response_method = ("predict_proba", "decision_function")
         display_kwargs = {"pos_label": pos_label}
         display = cast(
@@ -1560,6 +1625,8 @@ def prediction_error(
         >>> display = comparison_report.metrics.prediction_error()
         >>> display.plot(kind="actual_vs_predicted")
         """
+        if self._parent._reports_type == "CrossValidationReport":
+            raise NotImplementedError()
         display_kwargs = {"subsample": subsample, "seed": seed}
         display = cast(
             PredictionErrorDisplay,

skore/src/skore/sklearn/_plot/metrics/precision_recall_curve.py

@@ -21,7 +21,7 @@
     _validate_style_kwargs,
     sample_mpl_colormap,
 )
-from skore.sklearn.types import MLTask, PositiveLabel, YPlotData
+from skore.sklearn.types import MLTask, PositiveLabel, ReportType, YPlotData
 
 
 class PrecisionRecallCurveDisplay(
@@ -81,7 +81,8 @@ class PrecisionRecallCurveDisplay(
     ml_task : {"binary-classification", "multiclass-classification"}
         The machine learning task.
 
-    report_type : {"comparison-estimator", "cross-validation", "estimator"}
+    report_type : {"comparison-cross-validation", "comparison-estimator", \
+        "cross-validation", "estimator"}
         The type of report.
 
     Attributes
@@ -121,7 +122,7 @@ def __init__(
         pos_label: Optional[PositiveLabel],
         data_source: Literal["train", "test", "X_y"],
         ml_task: MLTask,
-        report_type: Literal["comparison-estimator", "cross-validation", "estimator"],
+        report_type: ReportType,
     ) -> None:
         self.precision = precision
         self.recall = recall
@@ -480,10 +481,15 @@ def plot(
         if pr_curve_kwargs is None:
             pr_curve_kwargs = self._default_pr_curve_kwargs
 
+        if self.ml_task == "binary-classification":
+            n_curves = len(self.average_precision[self.pos_label])
+        else:
+            n_curves = len(self.average_precision)
+
         pr_curve_kwargs = self._validate_curve_kwargs(
             curve_param_name="pr_curve_kwargs",
             curve_kwargs=pr_curve_kwargs,
-            metric=self.average_precision,
+            n_curves=n_curves,
             report_type=self.report_type,
         )
 
@@ -512,10 +518,13 @@ def plot(
                 estimator_names=self.estimator_names,
                 pr_curve_kwargs=pr_curve_kwargs,
             )
+        elif self.report_type == "comparison-cross-validation":
+            raise NotImplementedError()
         else:
             raise ValueError(
-                f"`report_type` should be one of 'estimator', 'cross-validation', "
-                f"or 'comparison-estimator'. Got '{self.report_type}' instead."
+                "`report_type` should be one of 'estimator', 'cross-validation', "
+                "'comparison-cross-validation' or 'comparison-estimator'. "
+                f"Got '{self.report_type}' instead."
             )
 
         xlabel = "Recall"
@@ -541,7 +550,7 @@ def _compute_data_for_display(
         y_true: Sequence[YPlotData],
         y_pred: Sequence[YPlotData],
         *,
-        report_type: Literal["comparison-estimator", "cross-validation", "estimator"],
+        report_type: ReportType,
         estimators: Sequence[BaseEstimator],
         estimator_names: list[str],
         ml_task: MLTask,
@@ -561,7 +570,8 @@ def _compute_data_for_display(
             confidence values, or non-thresholded measure of decisions (as returned by
             "decision_function" on some classifiers).
 
-        report_type : {"comparison-estimator", "cross-validation", "estimator"}
+        report_type : {"comparison-cross-validation", "comparison-estimator", \
+            "cross-validation", "estimator"}
             The type of report.
 
         estimators : list of estimator instances

skore/src/skore/sklearn/_plot/metrics/prediction_error.py

@@ -17,7 +17,7 @@
     _validate_style_kwargs,
     sample_mpl_colormap,
 )
-from skore.sklearn.types import MLTask, YPlotData
+from skore.sklearn.types import MLTask, ReportType, YPlotData
 
 RangeData = namedtuple("RangeData", ["min", "max"])
 
@@ -62,7 +62,8 @@ class PredictionErrorDisplay(StyleDisplayMixin, HelpDisplayMixin):
     ml_task : {"regression", "multioutput-regression"}
         The machine learning task.
 
-    report_type : {"cross-validation", "estimator", "comparison-estimator"}
+    report_type : {"comparison-cross-validation", "comparison-estimator", \
+        "cross-validation", "estimator"}
         The type of report.
 
     Attributes
@@ -113,7 +114,7 @@ def __init__(
         estimator_names: list[str],
         data_source: Literal["train", "test", "X_y"],
         ml_task: MLTask,
-        report_type: Literal["cross-validation", "estimator", "comparison-estimator"],
+        report_type: ReportType,
     ) -> None:
         self.y_true = y_true
         self.y_pred = y_pred
@@ -557,7 +558,7 @@ def _compute_data_for_display(
         y_true: list[YPlotData],
         y_pred: list[YPlotData],
         *,
-        report_type: Literal["cross-validation", "estimator", "comparison-estimator"],
+        report_type: ReportType,
         estimator_names: list[str],
         ml_task: MLTask,
         data_source: Literal["train", "test", "X_y"],
@@ -575,6 +576,10 @@ def _compute_data_for_display(
         y_pred : list of array-like of shape (n_samples,)
             Predicted target values.
 
+        report_type : {"comparison-cross-validation", "comparison-estimator", \
+            "cross-validation", "estimator"}
+            The type of report.
+
         estimators : list of estimator instances
             The estimators from which `y_pred` is obtained.