chore(skore-hub-project): Turn pytest warnings into errors

Author: thomass-dev

skore-hub-project/pyproject.toml

@@ -59,6 +59,7 @@ addopts = [
   "--verbosity=2",
   "--dist=loadscope",
 ]
+filterwarnings = ["error"]
 
 [tool.coverage.run]
 branch = true

skore-hub-project/src/skore_hub_project/metric/precision.py

@@ -38,7 +38,6 @@ class PrecisionTest(Precision):  # noqa: D101
 
 class PrecisionMean(CrossValidationReportMetric):  # noqa: D101
     accessor: ClassVar[str] = "metrics.precision"
-    aggregate: ClassVar[Literal["mean"]] = "mean"
     name: str = "precision_mean"
     verbose_name: str = "Precision (macro) - MEAN"
     greater_is_better: bool = True
@@ -53,7 +52,7 @@ def value(self) -> float | None:  # noqa: D102
             return None
 
         dataframe = function(
-            data_source=self.data_source, aggregate=self.aggregate, average="macro"
+            data_source=self.data_source, aggregate="mean", average="macro"
         )
 
         return cast_to_float(dataframe.iloc[0, 0])
@@ -69,7 +68,6 @@ class PrecisionTestMean(PrecisionMean):  # noqa: D101
 
 class PrecisionStd(CrossValidationReportMetric):  # noqa: D101
     accessor: ClassVar[str] = "metrics.precision"
-    aggregate: ClassVar[Literal["std"]] = "std"
     name: str = "precision_std"
     verbose_name: str = "Precision (macro) - STD"
     greater_is_better: bool = False
@@ -84,7 +82,7 @@ def value(self) -> float | None:  # noqa: D102
             return None
 
         dataframe = function(
-            data_source=self.data_source, aggregate=self.aggregate, average="macro"
+            data_source=self.data_source, aggregate="std", average="macro"
         )
 
         return cast_to_float(dataframe.iloc[0, 0])

skore-hub-project/tests/unit/project/test_project.py

@@ -6,7 +6,7 @@
 
 import joblib
 from httpx import Client, Response
-from pytest import fixture, raises
+from pytest import fixture, mark, raises
 from skore import CrossValidationReport, EstimatorReport
 from skore_hub_project import Project
 from skore_hub_project.report import (
@@ -128,6 +128,9 @@ def test_put_estimator_report(self, monkeypatch, binary_classification, respx_mo
         # Compare content with the desired output
         assert content == desired
 
+    @mark.filterwarnings(
+        "ignore:Precision is ill-defined.*:sklearn.exceptions.UndefinedMetricWarning"
+    )
     def test_put_cross_validation_report(
         self, monkeypatch, small_cv_binary_classification, respx_mock
     ):

skore-hub-project/tests/unit/report/test_cross_validation_report.py

@@ -120,6 +120,9 @@ def test_splits_test_samples_density_many_rows(self):
     def test_class_names(self, payload):
         assert payload.class_names == ["1", "0"]
 
+    @mark.filterwarnings(
+        "ignore:The `disp` and `iprint` options.* will be removed.*:DeprecationWarning"
+    )
     def test_classes(self, payload):
         X, y = make_classification(
             random_state=42,
@@ -145,6 +148,9 @@ def test_classes(self, payload):
     def test_classes_many_rows(self, payload):
         assert payload.classes == [0, 0, 1, 1, 1, 0, 0, 1, 0, 1]
 
+    @mark.filterwarnings(
+        "ignore:Precision is ill-defined.*:sklearn.exceptions.UndefinedMetricWarning"
+    )
     @mark.usefixtures("monkeypatch_artifact_hub_client")
     @mark.usefixtures("monkeypatch_upload_routes")
     @mark.usefixtures("monkeypatch_upload_with_mock")
@@ -192,6 +198,9 @@ def test_pickle(
             "content_type": "application/octet-stream",
         }
 
+    @mark.filterwarnings(
+        "ignore:Precision is ill-defined.*:sklearn.exceptions.UndefinedMetricWarning"
+    )
     def test_metrics(self, payload):
         assert list(map(type, payload.metrics)) == [
             AccuracyTestMean,
@@ -234,6 +243,9 @@ def test_medias(self, payload):
             TableReport,
         ]
 
+    @mark.filterwarnings(
+        "ignore:Precision is ill-defined.*:sklearn.exceptions.UndefinedMetricWarning"
+    )
     @mark.usefixtures("monkeypatch_artifact_hub_client")
     @mark.usefixtures("monkeypatch_upload_routes")
     def test_model_dump(self, small_cv_binary_classification, payload):