diff --git a/src/whylogs/core/types/typeddataconverter.py b/src/whylogs/core/types/typeddataconverter.py
index 48d8d8c17a..4de1197cb1 100644
--- a/src/whylogs/core/types/typeddataconverter.py
+++ b/src/whylogs/core/types/typeddataconverter.py
@@ -13,7 +13,7 @@
 # Dictionary mapping from type Number to type name
 TYPENUM_TO_NAME = {k: v for v, k in InferredType.Type.items()}
 INTEGRAL_TYPES = (int, np.integer)
-FLOAT_TYPES = (float, np.float)
+FLOAT_TYPES = (float, np.floating)
 
 
 class TypedDataConverter:
diff --git a/tests/unit/core/statistics/datatypes/test_floattracker.py b/tests/unit/core/statistics/datatypes/test_floattracker.py
index 33fad8e840..31ad4134d3 100644
--- a/tests/unit/core/statistics/datatypes/test_floattracker.py
+++ b/tests/unit/core/statistics/datatypes/test_floattracker.py
@@ -1,45 +1,75 @@
+import numpy as np
+
 from whylogs.core.statistics.datatypes import FloatTracker
 
 
+def _test_tracker_vs_array(tracker_of_array, array):
+    for val in array:
+        tracker_of_array.update(val)
+    assert tracker_of_array.count == len(array)
+    assert tracker_of_array.max == max(array)
+    assert tracker_of_array.min == min(array)
+    assert tracker_of_array.sum == sum(array)
+
+
 def test_values_are_min_max():
-    first = FloatTracker()
+    tracker = FloatTracker()
     vals1 = [1.0, 2.0, 3.0]
-    for val in vals1:
-        first.update(val)
+    _test_tracker_vs_array(tracker, vals1)
+
+
+def test_np_float():
+    float32_tracker = FloatTracker()
+    float32_array = np.array([1.0, 2.0, 3.0], dtype=np.float32)
+    _test_tracker_vs_array(float32_tracker, float32_array)
 
-    assert first.count == len(vals1)
-    assert first.max == max(vals1)
-    assert first.min == min(vals1)
-    assert first.sum == sum(vals1)
+    float64_tracker = FloatTracker()
+    float64_array = np.array([1.0, 2.0, 3.0], dtype=np.float64)
+    _test_tracker_vs_array(float64_tracker, float64_array)
 
 
 def test_merge_floattrackers_should_addup():
-    first = FloatTracker()
+    first_tracker = FloatTracker()
     vals1 = [1.0, 2.0, 3.0]
-    for val in vals1:
-        first.update(val)
 
-    assert first.count == len(vals1)
-    assert first.max == max(vals1)
-    assert first.min == min(vals1)
-    assert first.sum == sum(vals1)
+    second_tracker = FloatTracker()
+    vals2 = [4.0, 5.0, 6.0]
+
+    all_vals = vals1 + vals2
+    _test_merged_tracker_vs_arrays(all_vals, first_tracker, second_tracker)
+
 
-    second = FloatTracker()
+def _test_merged_tracker_vs_arrays(combined_arrays, first_tracker, second_tracker):
+    merge_first = first_tracker.merge(second_tracker)
+    assert merge_first.count == len(combined_arrays)
+    assert merge_first.max == max(combined_arrays)
+    assert merge_first.min == min(combined_arrays)
+    assert merge_first.sum == sum(combined_arrays)
+    merge_second = second_tracker.merge(first_tracker)
+    assert merge_second.__dict__ == merge_first.__dict__
+
+
+def test_merge_floattrackers_should_addup():
+    float32_tracker = FloatTracker()
+    float32_array = np.array([1.0, 2.0, 3.0], dtype=np.float32)
+    for val in float32_array:
+        float32_tracker.update(val)
+
+    float64_tracker = FloatTracker()
+    float64_array = np.array([1.0, 2.0, 3.0], dtype=np.float64)
+    for val in float64_array:
+        float64_tracker.update(val)
+
+    simple_tracker = FloatTracker()
     vals2 = [4.0, 5.0, 6.0]
     for val in vals2:
-        second.update(val)
+        simple_tracker.update(val)
 
-    assert second.count == len(vals2)
-    assert second.max == max(vals2)
-    assert second.min == min(vals2)
-    assert second.sum == sum(vals2)
+    merge_32_and_simple = float32_array.tolist() + vals2
+    _test_merged_tracker_vs_arrays(merge_32_and_simple, float32_tracker, simple_tracker)
 
-    all_vals = vals1 + vals2
-    merge_first = first.merge(second)
-    assert merge_first.count == len(all_vals)
-    assert merge_first.max == max(all_vals)
-    assert merge_first.min == min(all_vals)
-    assert merge_first.sum == sum(all_vals)
+    merge_64_and_simple = float64_array.tolist() + vals2
+    _test_merged_tracker_vs_arrays(merge_64_and_simple, float64_tracker, simple_tracker)
 
-    merge_second = second.merge(first)
-    assert merge_second.__dict__ == merge_first.__dict__
+    merge_64_and_32 = np.concatenate((float64_array, float32_array))
+    _test_merged_tracker_vs_arrays(merge_64_and_32, float32_tracker, float64_tracker)