Handling return type when creating features from DatetimeTimeIndex

featuretools/primitives/aggregation_primitives.py

@@ -83,7 +83,7 @@ def pd_mode(s):
 Min = make_agg_primitive(
     np.min,
     [Numeric],
-    None,
+    Numeric,
     name="Min",
     stack_on_self=False,
     description="Finds the minimum non-null value of a numeric feature.")
@@ -93,7 +93,7 @@ class Max(AggregationPrimitive):
     """Finds the maximum non-null value of a numeric feature."""
     name = "max"
     input_types = [Numeric]
-    return_type = None
+    return_type = Numeric
     # max_stack_depth = 1
     stack_on_self = False
 
@@ -222,7 +222,7 @@ class Median(AggregationPrimitive):
     """Finds the median value of any feature with well-ordered values."""
     name = "median"
     input_types = [Numeric]
-    return_type = None
+    return_type = Numeric
     # max_stack_depth = 2
 
     def get_function(self):

featuretools/primitives/primitive_base.py

@@ -11,7 +11,7 @@
     _check_time_against_column,
     _check_timedelta
 )
-from featuretools.variable_types import Variable
+from featuretools.variable_types import Datetime, DatetimeTimeIndex, Variable
 
 logger = logging.getLogger('featuretools')
 
@@ -101,6 +101,11 @@ def variable_type(self):
             feature = feature.base_features[0]
             return_type = feature.return_type
 
+            # only the original time index should exist
+            # so make this feature's return type just a Datetime
+            if return_type == DatetimeTimeIndex:
+                return_type = Datetime
+
         return return_type
 
     @property

featuretools/tests/feature_function_tests/test_primitive_base.py

@@ -2,8 +2,9 @@
 
 from ..testing_utils import make_ecommerce_entityset
 
-from featuretools.primitives import Feature, IdentityFeature, Last, Sum
+from featuretools.primitives import Feature, IdentityFeature, Last, Mode, Sum
 from featuretools.utils.gen_utils import getsize
+from featuretools.variable_types import Datetime
 
 
 @pytest.fixture(scope='module')
@@ -79,3 +80,19 @@ def test_squared(es):
     squared = feature * feature
     assert len(squared.base_features) == 1
     assert squared.base_features[0].hash() == feature.hash()
+
+
+def test_return_type_inference(es):
+    mode = Mode(es["log"]["priority_level"], es["customers"])
+    assert mode.variable_type == es["log"]["priority_level"].__class__
+
+
+def test_return_type_inference_direct_feature(es):
+    mode = Mode(es["log"]["priority_level"], es["customers"])
+    mode_session = Feature(mode, es["sessions"])
+    assert mode_session.variable_type == es["log"]["priority_level"].__class__
+
+
+def test_return_type_inference_time_index(es):
+    last = Last(es["log"]["datetime"], es["customers"])
+    assert last.variable_type == Datetime