[Feature Store] Fix Imputer's None types check (mlrun#2941)

mlrun/feature_store/steps.py

@@ -272,12 +272,12 @@ def __init__(
         :param kwargs:        optional kwargs (for storey)
         """
         super().__init__(**kwargs)
-        self.mapping = mapping
+        self.mapping = mapping or {}
         self.method = method
         self.default_value = default_value
 
-    def _impute(self, feature: str, value):
-        if value is None:
+    def _impute(self, feature: str, value: Any):
+        if pd.isna(value):
             return self.mapping.get(feature, self.default_value)
         return value
 

tests/feature-store/test_steps.py

@@ -16,6 +16,7 @@
 import time
 import unittest.mock
 
+import numpy as np
 import pandas as pd
 import pytest
 
@@ -616,6 +617,41 @@ def test_pandas_step_drop_feature(rundb_mock, entities, set_index_before):
     )
 
 
+@pytest.mark.parametrize("engine", ["storey", "pandas"])
+def test_imputer_default_value(rundb_mock, engine):
+    data_with_nones = pd.DataFrame(
+        {
+            "id": [1, 2, 3, 4],
+            "height": [None, 160, pd.NA, np.nan],
+            "age": [20, pd.NaT, 19, 18],
+        }
+    )
+    # Building graph with Imputer:
+    feature_set = fstore.FeatureSet(
+        "fs-default-value",
+        entities=["id"],
+        description="feature set with nones",
+        engine=engine,
+    )
+    feature_set.graph.to(Imputer(default_value=1))
+
+    # Mocking
+    output_path = tempfile.TemporaryDirectory()
+    feature_set._run_db = rundb_mock
+    feature_set.reload = unittest.mock.Mock()
+    feature_set.save = unittest.mock.Mock()
+    feature_set.purge_targets = unittest.mock.Mock()
+
+    imputed_df = fstore.ingest(
+        featureset=feature_set,
+        source=data_with_nones,
+        targets=[ParquetTarget(path=f"{output_path.name}/temp.parquet")],
+    )
+
+    # Checking that the ingested dataframe is none-free:
+    assert not imputed_df.isnull().values.any()
+
+
 def get_data(with_none=False):
     names = ["A", "B", "C", "D", "E"]
     ages = [33, 4, 76, 90, 24]