Fix BoxPlot KeyError when columns is not provided

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "validmind"
-version = "2.13.13"
+version = "2.13.14"
 description = "ValidMind Library"
 readme = "README.pypi.md"
 requires-python = ">=3.9,<3.15"

r/validmind/DESCRIPTION

@@ -1,7 +1,7 @@
 Package: validmind
 Type: Package
 Title: Interface to the 'ValidMind' Platform
-Version: 2.13.13
+Version: 2.13.14
 Authors@R: c(person("Andres", "Rodriguez", role = c("aut", "cre","cph"),
                       email = "andres@validmind.ai"))
 Maintainer: Andres Rodriguez <andres@validmind.ai>

tests/unit_tests/plots/test_BoxPlot.py

@@ -0,0 +1,45 @@
+import unittest
+
+import pandas as pd
+import validmind as vm
+from plotly.graph_objects import Figure
+
+from validmind.errors import SkipTestError
+from validmind.tests.plots.BoxPlot import BoxPlot
+
+
+class TestBoxPlot(unittest.TestCase):
+    def setUp(self):
+        df = pd.DataFrame(
+            {
+                "num1": [1.0, 2.0, 3.0, 4.0],
+                "num2": [10, 20, 30, 40],
+                "cat": ["a", "b", "a", "b"],
+                "target": [0, 1, 0, 1],
+            }
+        )
+        self.vm_dataset = vm.init_dataset(
+            input_id="dataset", dataset=df, target_column="target", __log=False
+        )
+
+    def test_default_columns_uses_numeric_features(self):
+        # Regression: columns=None used to raise KeyError("None of [Index([None]...")
+        fig = BoxPlot(self.vm_dataset)
+        self.assertIsInstance(fig, Figure)
+        self.assertEqual(sorted(t.name for t in fig.data), ["num1", "num2"])
+
+    def test_explicit_columns_filtered_to_numeric(self):
+        fig = BoxPlot(self.vm_dataset, columns=["num1", "cat", "missing"])
+        self.assertEqual([t.name for t in fig.data], ["num1"])
+
+    def test_group_by(self):
+        fig = BoxPlot(self.vm_dataset, columns=["num1"], group_by="cat")
+        self.assertEqual(len(fig.data), 2)
+
+    def test_no_numeric_columns_skips(self):
+        with self.assertRaises(SkipTestError):
+            BoxPlot(self.vm_dataset, columns=["cat"])
+
+
+if __name__ == "__main__":
+    unittest.main()

validmind/__version__.py

@@ -1 +1 @@
-__version__ = "2.13.13"
+__version__ = "2.13.14"

validmind/tests/plots/BoxPlot.py

@@ -4,7 +4,6 @@
 
 from typing import List, Optional
 
-import pandas as pd
 import plotly.graph_objects as go
 from plotly.subplots import make_subplots
 
@@ -18,21 +17,12 @@ def _validate_inputs(
 ):
     """Validate inputs and return validated columns."""
 
-    # Get dtypes without loading data into memory
-    if not isinstance(columns, list):
-        columns = [columns]
-
-    columns_dtypes = dataset._df[columns].dtypes
-
-    columns_numeric = []
-    columns_numeric = columns_dtypes[
-        columns_dtypes.apply(lambda x: pd.api.types.is_numeric_dtype(x))
-    ].index.tolist()
-
     if columns is None:
-        columns = columns_numeric
+        columns = list(dataset.feature_columns_numeric)
     else:
-        available_columns = set(columns_numeric)
+        if not isinstance(columns, list):
+            columns = [columns]
+        available_columns = set(dataset.feature_columns_numeric)
         columns = [col for col in columns if col in available_columns]
 
     if not columns: