'index' and 'columns' added to fixture and related changes.

Author: tp

doc/source/whatsnew/v0.24.0.txt

@@ -475,11 +475,9 @@ Numeric
 - Bug in :class:`Series` ``__rmatmul__`` doesn't support matrix vector multiplication (:issue:`21530`)
 - Bug in :func:`factorize` fails with read-only array (:issue:`12813`)
 - Fixed bug in :func:`unique` handled signed zeros inconsistently: for some inputs 0.0 and -0.0 were treated as equal and for some inputs as different. Now they are treated as equal for all inputs (:issue:`21866`)
-- Bug in :meth:`DataFrame.agg`, :meth:`DataFrame.transform` and :meth:`DataFrame.apply` when ``axis=1``.
-  Using ``apply`` with a list of functions and axis=1 (e.g. ``df.apply(['abs'], axis=1)``)
-  previously gave a TypeError. This fixes that issue.
-  As ``agg`` and ``transform`` in some cases delegate to ``apply``, this also
-  fixed this issue for them (:issue:`16679`).
+- Bug in :meth:`DataFrame.agg`, :meth:`DataFrame.transform` and :meth:`DataFrame.apply` where,
+  when supplied with a list of functions and ``axis=1`` (e.g. ``df.apply(['sum', 'mean'], axis=1)``),
+  a ``TypeError`` was wrongly raised. For all three methods such calculation are now done correctly. (:issue:`16679`).
 -
 
 Strings

pandas/conftest.py

@@ -60,16 +60,16 @@ def spmatrix(request):
     return getattr(sparse, request.param + '_matrix')
 
 
-@pytest.fixture(params=[0, 1],
-                ids=lambda x: "axis {}".format(x))
+@pytest.fixture(params=[0, 1, 'index', 'columns'],
+                ids=lambda x: "axis {!r}".format(x))
 def axis(request):
     """
      Fixture for returning the axis numbers of a dataframe.
      """
     return request.param
 
 
-@pytest.fixture(params=[0], ids=lambda x: "axis {}".format(x))
+@pytest.fixture(params=[0, 'index'], ids=lambda x: "axis {!r}".format(x))
 def axis_series(request):
     """
      Fixture for returning the axis numbers of a series.

pandas/core/apply.py

@@ -5,6 +5,8 @@
 from pandas.core.dtypes.generic import ABCSeries
 from pandas.core.dtypes.common import (
     is_extension_type,
+    is_dict_like,
+    is_list_like,
     is_sequence)
 from pandas.util._decorators import cache_readonly
 
@@ -106,7 +108,7 @@ def get_result(self):
         """ compute the results """
 
         # dispatch to agg
-        if isinstance(self.f, (list, dict)):
+        if is_list_like(self.f) or is_dict_like(self.f):
             return self.obj.aggregate(self.f, axis=self.axis,
                                       *self.args, **self.kwds)
 

pandas/core/frame.py

@@ -6079,7 +6079,7 @@ def aggregate(self, func, axis=0, *args, **kwargs):
         return result
 
     def _aggregate(self, arg, axis=0, *args, **kwargs):
-        if axis == 1:
+        if axis in {1, 'columns'}:
             result, how = (super(DataFrame, self.T)
                            ._aggregate(arg, *args, **kwargs))
             result = result.T if result is not None else result
@@ -6089,7 +6089,7 @@ def _aggregate(self, arg, axis=0, *args, **kwargs):
     agg = aggregate
 
     def transform(self, func, axis=0, *args, **kwargs):
-        if axis == 1:
+        if axis in {1, 'columns'}:
             return super(DataFrame, self.T).transform(func, *args, **kwargs).T
         return super(DataFrame, self).transform(func, *args, **kwargs)
 

pandas/tests/frame/test_apply.py

@@ -869,7 +869,7 @@ def zip_frames(frames, axis=1):
 class TestDataFrameAggregate(TestData):
 
     def test_agg_transform(self, axis):
-        other_axis = abs(axis - 1)
+        other_axis = 1 if axis in {0, 'index'} else 0
 
         with np.errstate(all='ignore'):
 
@@ -890,7 +890,7 @@ def test_agg_transform(self, axis):
             # list-like
             result = self.frame.apply([np.sqrt], axis=axis)
             expected = f_sqrt.copy()
-            if axis == 0:
+            if axis in {0, 'index'}:
                 expected.columns = pd.MultiIndex.from_product(
                     [self.frame.columns, ['sqrt']])
             else:
@@ -906,7 +906,7 @@ def test_agg_transform(self, axis):
             # functions per series and then concatting
             result = self.frame.apply([np.abs, np.sqrt], axis=axis)
             expected = zip_frames([f_abs, f_sqrt], axis=other_axis)
-            if axis == 0:
+            if axis in {0, 'index'}:
                 expected.columns = pd.MultiIndex.from_product(
                     [self.frame.columns, ['absolute', 'sqrt']])
             else:
@@ -1001,7 +1001,7 @@ def test_agg_dict_nested_renaming_depr(self):
                     'B': {'bar': 'max'}})
 
     def test_agg_reduce(self, axis):
-        other_axis = abs(axis - 1)
+        other_axis = 1 if axis in {0, 'index'} else 0
         name1, name2 = self.frame.axes[other_axis].unique()[:2].sort_values()
 
         # all reducers
@@ -1010,7 +1010,7 @@ def test_agg_reduce(self, axis):
                               self.frame.sum(axis=axis),
                               ], axis=1)
         expected.columns = ['mean', 'max', 'sum']
-        expected = expected.T if axis == 0 else expected
+        expected = expected.T if axis in {0, 'index'} else expected
 
         result = self.frame.agg(['mean', 'max', 'sum'], axis=axis)
         assert_frame_equal(result, expected)
@@ -1031,7 +1031,7 @@ def test_agg_reduce(self, axis):
                           index=['mean']),
             name2: Series([self.frame.loc(other_axis)[name2].sum()],
                           index=['sum'])})
-        expected = expected.T if axis == 1 else expected
+        expected = expected.T if axis in {1, 'columns'} else expected
         assert_frame_equal(result, expected)
 
         # dict input with lists with multiple
@@ -1045,7 +1045,7 @@ def test_agg_reduce(self, axis):
                            self.frame.loc(other_axis)[name2].max()],
                            index=['sum', 'max'])),
         ]))
-        expected = expected.T if axis == 1 else expected
+        expected = expected.T if axis in {1, 'columns'} else expected
         assert_frame_equal(result, expected)
 
     def test_nuiscance_columns(self):

pandas/tests/generic/test_label_or_level_utils.py

@@ -82,7 +82,7 @@ def test_is_level_or_label_reference_df_simple(df_levels, axis):
     expected_labels, expected_levels = get_labels_levels(df_levels)
 
     # Transpose frame if axis == 1
-    if axis == 1:
+    if axis in {1, 'columns'}:
         df_levels = df_levels.T
 
     # Perform checks
@@ -93,7 +93,7 @@ def test_is_level_or_label_reference_df_simple(df_levels, axis):
 def test_is_level_reference_df_ambig(df_ambig, axis):
 
     # Transpose frame if axis == 1
-    if axis == 1:
+    if axis in {1, 'columns'}:
         df_ambig = df_ambig.T
 
     # df has both an on-axis level and off-axis label named L1
@@ -166,7 +166,7 @@ def test_is_label_or_level_reference_panel_error(panel):
 def test_check_label_or_level_ambiguity_df(df_ambig, axis):
 
     # Transpose frame if axis == 1
-    if axis == 1:
+    if axis in {1, 'columns'}:
         df_ambig = df_ambig.T
 
     # df_ambig has both an on-axis level and off-axis label named L1
@@ -176,7 +176,7 @@ def test_check_label_or_level_ambiguity_df(df_ambig, axis):
 
         assert df_ambig._check_label_or_level_ambiguity('L1', axis=axis)
         warning_msg = w[0].message.args[0]
-        if axis == 0:
+        if axis in {0, 'index'}:
             assert warning_msg.startswith("'L1' is both an index level "
                                           "and a column label")
         else:
@@ -236,7 +236,7 @@ def test_check_label_or_level_ambiguity_panel_error(panel):
 # ===============================
 def assert_label_values(frame, labels, axis):
     for label in labels:
-        if axis == 0:
+        if axis in {0, 'index'}:
             expected = frame[label]._values
         else:
             expected = frame.loc[label]._values
@@ -248,7 +248,7 @@ def assert_label_values(frame, labels, axis):
 
 def assert_level_values(frame, levels, axis):
     for level in levels:
-        if axis == 0:
+        if axis in {0, 'index'}:
             expected = frame.index.get_level_values(level=level)._values
         else:
             expected = (frame.columns
@@ -267,7 +267,7 @@ def test_get_label_or_level_values_df_simple(df_levels, axis):
     expected_labels, expected_levels = get_labels_levels(df_levels)
 
     # Transpose frame if axis == 1
-    if axis == 1:
+    if axis in {1, 'columns'}:
         df_levels = df_levels.T
 
     # Perform checks
@@ -278,7 +278,7 @@ def test_get_label_or_level_values_df_simple(df_levels, axis):
 def test_get_label_or_level_values_df_ambig(df_ambig, axis):
 
     # Transpose frame if axis == 1
-    if axis == 1:
+    if axis in {1, 'columns'}:
         df_ambig = df_ambig.T
 
     # df has both an on-axis level and off-axis label named L1
@@ -298,7 +298,7 @@ def test_get_label_or_level_values_df_ambig(df_ambig, axis):
 def test_get_label_or_level_values_df_duplabels(df_duplabels, axis):
 
     # Transpose frame if axis == 1
-    if axis == 1:
+    if axis in {1, 'columns'}:
         df_duplabels = df_duplabels.T
 
     # df has unambiguous level 'L1'
@@ -308,7 +308,7 @@ def test_get_label_or_level_values_df_duplabels(df_duplabels, axis):
     assert_label_values(df_duplabels, ['L3'], axis=axis)
 
     # df has duplicate labels 'L2'
-    if axis == 0:
+    if axis in {0, 'index'}:
         expected_msg = "The column label 'L2' is not unique"
     else:
         expected_msg = "The index label 'L2' is not unique"
@@ -355,7 +355,7 @@ def assert_labels_dropped(frame, labels, axis):
     for label in labels:
         df_dropped = frame._drop_labels_or_levels(label, axis=axis)
 
-        if axis == 0:
+        if axis in {0, 'index'}:
             assert label in frame.columns
             assert label not in df_dropped.columns
         else:
@@ -367,7 +367,7 @@ def assert_levels_dropped(frame, levels, axis):
     for level in levels:
         df_dropped = frame._drop_labels_or_levels(level, axis=axis)
 
-        if axis == 0:
+        if axis in {0, 'index'}:
             assert level in frame.index.names
             assert level not in df_dropped.index.names
         else:
@@ -383,7 +383,7 @@ def test_drop_labels_or_levels_df(df_levels, axis):
     expected_labels, expected_levels = get_labels_levels(df_levels)
 
     # Transpose frame if axis == 1
-    if axis == 1:
+    if axis in {1, 'columns'}:
         df_levels = df_levels.T
 
     # Perform checks