我刚刚发现在查询方法中解析了关键字columns,但是正确的使用方法是什么呢?我没有在 doc 中找到解释.
>>> df = pd.DataFrame(np.arange(6).reshape(2,3))
>>> df
0 1 2
0 0 1 2
1 3 4 5
>>> df.query("columns >= 0")
0 1 2
0 0 1 2
1 3 4 5
>>> print df.query("columns >= 2")
Empty DataFrame
Columns: [0, 1, 2]
Index: []
#with unnamed columns, 'columns" seems to behave like "index"
>>> df2 = pd.DataFrame(np.arange(6).reshape(2,3), columns = ['a','b','c'])
>>> df2
a b c
0 0 1 2
1 3 4 5
>>> df2.query('columns >= 2')
IndexingError: Unalignable boolean Series key provided
请您参考如下方法:
DataFrame.query 相当于先调用DataFrame.eval,然后使用评估结果索引原始DataFrame。
In [9]: idx = df.eval('columns >= 1')
In [10]: idx
Out[10]:
0 False
1 True
2 True
dtype: bool
In [11]: df.loc[idx]
Out[11]:
0 1 2
1 3 4 5
这里的'columns' 等同于df.columns 并且eval 的结果是索引为df 的Serials。列,因此当列被命名时,eval的结果不能用作原始DataFrame的索引。
In [13]: idx2 = df2.eval('columns >= 1')
In [14]: idx2
Out[14]:
a True
b True
c True
dtype: bool
实际上,像这样在query 表达式中使用'columns' 是不好的。在第一个示例中,返回的 Series 可以用作索引只是碰巧,但它可能不适用于一般情况。
例如,DataFrame 形状的简单更改将导致错误。
In [15]: df3 = pd.DataFrame(np.arange(6).reshape(3,2))
In [16]: df3.query('columns >= 1')
---------------------------------------------------------------------------
IndexingError Traceback (most recent call last)
<ipython-input-26-75248672c992> in <module>()
1 df = pd.DataFrame(np.arange(6).reshape(3,2))
----> 2 df.query('columns >= 1')
/home/ubuntu/anaconda/lib/python2.7/site-packages/pandas/core/frame.pyc in query(self, expr, **kwargs)
1936
1937 try:
-> 1938 return self.loc[res]
1939 except ValueError:
1940 # when res is multi-dimensional loc raises, but this is sometimes a
/home/ubuntu/anaconda/lib/python2.7/site-packages/pandas/core/indexing.pyc in __getitem__(self, key)
1187 return self._getitem_tuple(key)
1188 else:
-> 1189 return self._getitem_axis(key, axis=0)
1190
1191 def _getitem_axis(self, key, axis=0):
/home/ubuntu/anaconda/lib/python2.7/site-packages/pandas/core/indexing.pyc in _getitem_axis(self, key, axis)
1304 return self._get_slice_axis(key, axis=axis)
1305 elif is_bool_indexer(key):
-> 1306 return self._getbool_axis(key, axis=axis)
1307 elif is_list_like_indexer(key):
1308
/home/ubuntu/anaconda/lib/python2.7/site-packages/pandas/core/indexing.pyc in _getbool_axis(self, key, axis)
1194 def _getbool_axis(self, key, axis=0):
1195 labels = self.obj._get_axis(axis)
-> 1196 key = check_bool_indexer(labels, key)
1197 inds, = key.nonzero()
1198 try:
/home/ubuntu/anaconda/lib/python2.7/site-packages/pandas/core/indexing.pyc in check_bool_indexer(ax, key)
1657 mask = com.isnull(result.values)
1658 if mask.any():
-> 1659 raise IndexingError('Unalignable boolean Series key provided')
1660
1661 result = result.astype(bool).values
IndexingError: Unalignable boolean Series key provided
