BUG: Fix issue pandas-dev#14848 groupby().describe() on indices containing all tuples

Author: Dr-Irv

doc/source/whatsnew/v0.20.0.txt

@@ -344,6 +344,7 @@ Bug Fixes
 - Bug in converting object elements of array-like objects to unsigned 64-bit integers (:issue:`4471`, :issue:`14982`)
 - Bug in ``pd.pivot_table()`` where no error was raised when values argument was not in the columns (:issue:`14938`)
 
+- Bug in ``DataFrame.groupby().describe()`` when grouping on an ``Index`` containing tuples (:issue:`14848`)
 
 
 

pandas/tests/groupby/test_groupby.py

@@ -1490,6 +1490,19 @@ def test_frame_describe_multikey(self):
         for name, group in groupedT:
             assert_frame_equal(result[name], group.describe())
 
+    # GH #14848
+    def test_frame_describe_tupleindex(self):
+        df1 = DataFrame({'x': [1, 2, 3, 4, 5] * 3,
+                         'y': [10, 20, 30, 40, 50] * 3,
+                         'z': [100, 200, 300, 400, 500] * 3})
+        df1['k'] = [(0, 0, 1), (0, 1, 0), (1, 0, 0)] * 5
+        df2 = df1.rename(columns={'k': 'key'})
+        des1 = df1.groupby('k').describe()
+        des2 = df2.groupby('key').describe()
+        if len(des1) > 0:
+            des2.index.set_names(des1.index.names, inplace=True)
+        assert_frame_equal(des1, des2)
+
     def test_frame_groupby(self):
         grouped = self.tsframe.groupby(lambda x: x.weekday())
 

pandas/tools/merge.py

@@ -1626,7 +1626,14 @@ def __init__(self, objs, axis=0, join='outer', join_axes=None,
                 clean_objs.append(v)
             objs = clean_objs
             name = getattr(keys, 'name', None)
-            keys = Index(clean_keys, name=name)
+            # GH 14848
+            # If coming from groupby(), keys and levels are specified
+            # as Index type, so don't tupleize
+            tupleize_cols = (levels is None or
+                             (isinstance(levels, list) and
+                              not isinstance(keys, Index) and
+                              not isinstance(levels[0], Index)))
+            keys = Index(clean_keys, name=name, tupleize_cols=tupleize_cols)
 
         if len(objs) == 0:
             raise ValueError('All objects passed were None')