Merge branch 'main' into followup_1759

Author: GaetandeCast

.github/dependabot.yml

@@ -1,17 +1,20 @@
 version: 2
 updates:
-
   - package-ecosystem: "github-actions"
     directory: "/"
     schedule:
       interval: "monthly"
+    cooldown:
+      default-days: 14
     commit-message:
       prefix: "chore(dependencies): GITHUB-ACTIONS"
-
   - package-ecosystem: "pip"
     directory: "/"
     schedule:
       interval: "weekly"
+    cooldown:
+      default-days: 14
+      semver-patch-days: 7
     commit-message:
       prefix: "chore(dependencies): PIP"
     groups:

skore-hub-project/src/skore_hub_project/artifact/media/feature_importance.py

@@ -45,7 +45,7 @@ class Permutation(FeatureImportance[EstimatorReport], ABC):  # noqa: D101
     name: Literal["permutation"] = "permutation"
 
     def content_to_upload(self) -> bytes | None:  # noqa: D102
-        for key, obj in reversed(self.report._cache.items()):
+        for key, obj in reversed(list(self.report._cache.items())):
             if len(key) < 7:
                 continue
 

skore-hub-project/tests/conftest.py

@@ -1,4 +1,6 @@
 from datetime import datetime, timezone
+from functools import partial
+from importlib import reload
 from unittest.mock import Mock
 from urllib.parse import urljoin
 
@@ -141,11 +143,10 @@ def monkeypatch_tmpdir(monkeypatch, tmp_path):
 
     https://docs.python.org/3/library/tempfile.html#tempfile.gettempdir
     """
-    import importlib
     import tempfile
 
     monkeypatch.setenv("TMPDIR", str(tmp_path))
-    importlib.reload(tempfile)
+    reload(tempfile)
 
 
 @fixture
@@ -155,7 +156,10 @@ def monkeypatch_skrub(monkeypatch):
 
     https://github.com/skrub-data/skrub/blob/35f573ce586fe61ef2c72f4c0c4b188ebf2e664b/skrub/_reporting/_html.py#L153
     """
+    from skrub._dataframe import sample
+
     monkeypatch.setattr("secrets.token_hex", lambda: "<token>")
+    monkeypatch.setattr("skrub._dataframe.sample", partial(sample, seed=42))
 
 
 @fixture
@@ -210,11 +214,22 @@ def monkeypatch_sklearn_estimator_html_repr(monkeypatch):
     monkeypatch.setattr(importpath, lambda self: "<id>")
 
 
+@fixture
+def monkeypatch_rich(monkeypatch):
+    """Make `rich` silent."""
+    from rich.console import Console
+
+    monkeypatch.setattr("rich.console.Console", partial(Console, quiet=True))
+    monkeypatch.setattr("skore.console.quiet", True)
+    monkeypatch.setattr("skore_hub_project.console.quiet", True)
+
+
 @fixture(autouse=True)
 def setup(
     monkeypatch_tmpdir,
     monkeypatch_matplotlib,
     monkeypatch_skrub,
     monkeypatch_skore_hub_envars,
     monkeypatch_sklearn_estimator_html_repr,
+    monkeypatch_rich,
 ): ...

skore/src/skore/_sklearn/_comparison/metrics_accessor.py

@@ -67,12 +67,14 @@ def summarize(
 
         Parameters
         ----------
-        data_source : {"test", "train", "X_y"}, default="test"
+        data_source : {"test", "train", "X_y", "both"}, default="test"
             The data source to use.
 
             - "test" : use the test set provided when creating the report.
             - "train" : use the train set provided when creating the report.
             - "X_y" : use the provided `X` and `y` to compute the metric.
+            - "both" : use both the train and test sets to compute the metrics and
+              present them side-by-side.
 
         X : array-like of shape (n_samples, n_features), default=None
             New data on which to compute the metric. By default, we use the validation
@@ -254,6 +256,7 @@ def _compute_metric_scores(
                     indicator_favorability=metric_kwargs.get(
                         "indicator_favorability", False
                     ),
+                    data_source=data_source,
                 )
             else:  # "CrossValidationReport"
                 results = _combine_cross_validation_results(

skore/src/skore/_sklearn/_comparison/utils.py

@@ -1,14 +1,16 @@
 import copy
+from typing import Literal
 
 import pandas as pd
 
-from skore._sklearn.types import Aggregate
+from skore._sklearn.types import Aggregate, DataSource
 
 
 def _combine_estimator_results(
     individual_results: list[pd.DataFrame],
     estimator_names: list[str],
     indicator_favorability: bool,
+    data_source: DataSource | Literal["both"],
 ) -> pd.DataFrame:
     """Combine a list of dataframes provided by `EstimatorReport`s.
 
@@ -24,6 +26,9 @@ def _combine_estimator_results(
     indicator_favorability : bool
         Whether to keep the Favorability column.
 
+    data_source : {"test", "train", "X_y", "both"}
+        The data source.
+
     Examples
     --------
     >>> from skore._sklearn._comparison.utils import _combine_estimator_results
@@ -55,6 +60,7 @@ def _combine_estimator_results(
     ...     individual_results,
     ...     estimator_names,
     ...     indicator_favorability=False,
+    ...     data_source="test",
     ... )
     Estimator    LogisticRegression_1  LogisticRegression_2
     Metric
@@ -73,7 +79,17 @@ def _combine_estimator_results(
     else:
         favorability = None
 
-    results.columns = pd.Index(estimator_names, name="Estimator")
+    if data_source == "both":
+        results.columns = pd.Index(
+            [
+                x
+                for name in estimator_names
+                for x in [f"{name} (train)", f"{name} (test)"]
+            ],
+            name="Estimator",
+        )
+    else:
+        results.columns = pd.Index(estimator_names, name="Estimator")
 
     if favorability is not None:
         results["Favorability"] = favorability

skore/src/skore/_sklearn/_estimator/metrics_accessor.py

@@ -1976,7 +1976,7 @@ class is set to the one provided when creating the report. If `None`,
     def prediction_error(
         self,
         *,
-        data_source: DataSource = "test",
+        data_source: DataSource | Literal["both"] = "test",
         X: ArrayLike | None = None,
         y: ArrayLike | None = None,
         subsample: float | int | None = 1_000,
@@ -1988,12 +1988,14 @@ def prediction_error(
 
         Parameters
         ----------
-        data_source : {"test", "train", "X_y"}, default="test"
+        data_source : {"test", "train", "X_y", "both"}, default="test"
             The data source to use.
 
             - "test" : use the test set provided when creating the report.
             - "train" : use the train set provided when creating the report.
             - "X_y" : use the provided `X` and `y` to compute the metric.
+            - "both" : use both the train and test sets to compute the metrics and
+              present them side-by-side.
 
         X : array-like of shape (n_samples, n_features), default=None
             New data on which to compute the metric. By default, we use the validation