BUG GH15150 crosstable normalize with multiindex

Author: Christoph Möhl

pandas/core/reshape/pivot.py

@@ -537,35 +537,46 @@ def _normalize(table, normalize, margins, margins_name='All'):
             raise ValueError("Not a valid normalize argument")
 
     if margins is False:
-
         # Actual Normalizations
         normalizers = {
             'all': lambda x: x / x.sum(axis=1).sum(axis=0),
             'columns': lambda x: x / x.sum(),
             'index': lambda x: x.div(x.sum(axis=1), axis=0)
         }
-    
+
     elif margins is True:
-        #skip margin rows and/or cols for normalization
+        # skip margin rows and/or cols for normalization
         normalizers = {
-            'all': lambda x: x / x.iloc[:-1,:-1].sum(axis=1).sum(axis=0),
-            'columns': lambda x: x.div(x.iloc[:-1,:].sum()).iloc[:-1,:],
-            'index': lambda x: (x.div(x.iloc[:,:-1].sum(axis=1), axis=0)).iloc[:,:-1]
+            'all': lambda x: x / x.iloc[:-1, :-1].sum(axis=1).sum(axis=0),
+            'columns': lambda x: x.div(x.iloc[:-1, :].sum()).iloc[:-1, :],
+            'index': lambda x: (x.div(x.iloc[:, :-1].sum(axis=1),
+                                axis=0)).iloc[:, :-1]
         }
 
     else:
-        raise ValueError("Not a valid margins argument")    
+        raise ValueError("Not a valid margins argument")
 
     normalizers[True] = normalizers['all']
 
     try:
         f = normalizers[normalize]
     except KeyError:
         raise ValueError("Not a valid normalize argument")
-      
+
     table = f(table)
     table = table.fillna(0)
-    
+
+    if margins is True:
+        # reset index to ensure default index dtype
+        if normalize == 'index':
+            colnames = table.columns.names
+            table.columns = Index(table.columns.tolist())
+            table.columns.names = colnames
+        if normalize == 'columns':
+            rownames = table.index.names
+            table.index = Index(table.index.tolist())
+            table.index.names = rownames
+
     return table
 
 

pandas/tests/reshape/test_pivot.py

@@ -1266,21 +1266,18 @@ def test_crosstab_normalize(self):
                                            [0.25, 0.75],
                                            [0.4, 0.6]],
                                           index=pd.Index([1, 2, 'All'],
-                                                         name='a',
-                                                         dtype='object'),
+                                                         name='a'),
                                           columns=pd.Index([3, 4], name='b'))
         col_normal_margins = pd.DataFrame([[0.5, 0, 0.2], [0.5, 1.0, 0.8]],
-                                          index=pd.Index([1, 2], name='a',
-                                                         dtype='object'),
+                                          index=pd.Index([1, 2], name='a'),
                                           columns=pd.Index([3, 4, 'All'],
                                                            name='b'))
 
         all_normal_margins = pd.DataFrame([[0.2, 0, 0.2],
                                            [0.2, 0.6, 0.8],
                                            [0.4, 0.6, 1]],
                                           index=pd.Index([1, 2, 'All'],
-                                                         name='a',
-                                                         dtype='object'),
+                                                         name='a'),
                                           columns=pd.Index([3, 4, 'All'],
                                                            name='b'))
         tm.assert_frame_equal(pd.crosstab(df.a, df.b, normalize='index',
@@ -1327,10 +1324,10 @@ def test_crosstab_normalize(self):
 
     def test_crosstab_norm_margins_with_multiindex(self):
         # GH 15150
-        a = np.array(['foo', 'bar', 'foo', 'bar','bar', 'foo'])
-        b = np.array(['one', 'one', 'two', 'one','two', 'two'])
-        c = np.array(['dull', 'shiny', 'dull', 'dull','dull', 'shiny'])
-        d = np.array(['a', 'a', 'b', 'a','b', 'b'])
+        a = np.array(['foo', 'bar', 'foo', 'bar', 'bar', 'foo'])
+        b = np.array(['one', 'one', 'two', 'one', 'two', 'two'])
+        c = np.array(['dull', 'shiny', 'dull', 'dull', 'dull', 'shiny'])
+        d = np.array(['a', 'a', 'b', 'a', 'b', 'b'])
         expected_col_colnorm = MultiIndex(levels=[['All', 'dull', 'shiny'],
                                                   ['', 'a', 'b']],
                                           labels=[[1, 1, 2, 2, 0],
@@ -1365,11 +1362,11 @@ def test_crosstab_norm_margins_with_multiindex(self):
                                             [.33333333, .33333333,
                                              .16666667, .16666667]])
         expected_indexnorm = pd.DataFrame(expected_data_indexnorm,
-                                 index=expected_index_indexnorm,
-                                 columns=expected_col_indexnorm)
+                                          index=expected_index_indexnorm,
+                                          columns=expected_col_indexnorm)
         expected_data_allnorm = np.array([[0.16666667, 0., .16666667,
                                            0., .33333333],
-                                          [0. ,.16666667, 0.,
+                                          [0., .16666667, 0.,
                                            0., .16666667],
                                           [.16666667, 0., 0.,
                                            0., .16666667],
@@ -1378,15 +1375,15 @@ def test_crosstab_norm_margins_with_multiindex(self):
                                           [0.33333333, .33333333, .16666667,
                                            .16666667, 1.]])
         expected_allnorm = pd.DataFrame(expected_data_allnorm,
-                                 index=expected_index_indexnorm,
-                                 columns=expected_col_colnorm)
+                                        index=expected_index_indexnorm,
+                                        columns=expected_col_colnorm)
 
-        result_colnorm = pd.crosstab([a, b], [c,d], normalize='columns',
+        result_colnorm = pd.crosstab([a, b], [c, d], normalize='columns',
                                      margins=True)
-        result_indexnorm = pd.crosstab([a, b], [c,d], normalize='index',
-                                       margins=True)
-        result_allnorm = pd.crosstab([a, b], [c,d], normalize='all',
+        result_indexnorm = pd.crosstab([a, b], [c, d], normalize='index',
                                        margins=True)
+        result_allnorm = pd.crosstab([a, b], [c, d], normalize='all',
+                                     margins=True)
 
         tm.assert_frame_equal(result_colnorm, expected_colnorm)
         tm.assert_frame_equal(result_indexnorm, expected_indexnorm)