Compare distribution of differences in InterRowMSAS instead of average (#805)

Author: sarahmish

sdmetrics/column_pairs/statistical/inter_row_msas.py

@@ -85,13 +85,11 @@ def diff_func(group):
             if len(group) <= n_rows_diff:
                 return np.nan
             group = group.to_numpy()
-            with warnings.catch_warnings():
-                warnings.filterwarnings('ignore', message='Mean of empty slice')
-                return np.nanmean(group[n_rows_diff:] - group[:-n_rows_diff])
+            return (group[n_rows_diff:] - group[:-n_rows_diff]).flatten()
 
         with warnings.catch_warnings():
             warnings.filterwarnings('ignore', message='invalid value encountered in.*')
-            return grouped.apply(diff_func)
+            return grouped.apply(diff_func).explode().astype(float)
 
     @classmethod
     def compute(cls, real_data, synthetic_data, n_rows_diff=1, apply_log=False):

tests/unit/column_pairs/statistical/test_inter_row_msas.py

@@ -24,7 +24,7 @@ def test_compute_breakdown(self):
         )
 
         # Assert
-        assert result == {'score': 0.5}
+        assert result == {'score': 0.25}
 
     def test_compute(self):
         """Test it runs."""
@@ -40,7 +40,7 @@ def test_compute(self):
         )
 
         # Assert
-        assert score == 0.5
+        assert score == 0.25
 
     def test_compute_nans(self):
         """Test it runs with nans."""