1. 选取多个DataFrame列

  1. #  用列表选取多个列
  2.  In[2]: movie = pd.read_csv('data/movie.csv')
  3.         movie_actor_director = movie[['actor_1_name', 'actor_2_name', 'actor_3_name', 'director_name']]
  4.         movie_actor_director.head()
  5. Out[2]:

1. 选取多个DataFrame列 - 图1

  1. #  选取单列
  2.  In[3]: movie[['director_name']].head()
  3. Out[3]:

1. 选取多个DataFrame列 - 图2

  1. #  错误的选取多列的方式
  2.  In[4]: movie['actor_1_name', 'actor_2_name', 'actor_3_name', 'director_name']
  3. ---------------------------------------------------------------------------
  4. KeyError                                  Traceback (most recent call last)
  5. /Users/Ted/anaconda/lib/python3.6/site-packages/pandas/core/indexes/base.py in get_loc(self, key, method, tolerance)
  6.    2441             try:
  7. -> 2442                 return self._engine.get_loc(key)
  8.    2443             except KeyError:
  10. pandas/_libs/index.pyx in pandas._libs.index.IndexEngine.get_loc (pandas/_libs/index.c:5280)()
  12. pandas/_libs/index.pyx in pandas._libs.index.IndexEngine.get_loc (pandas/_libs/index.c:5126)()
  14. pandas/_libs/hashtable_class_helper.pxi in pandas._libs.hashtable.PyObjectHashTable.get_item (pandas/_libs/hashtable.c:20523)()
  16. pandas/_libs/hashtable_class_helper.pxi in pandas._libs.hashtable.PyObjectHashTable.get_item (pandas/_libs/hashtable.c:20477)()
  18. KeyError: ('actor_1_name', 'actor_2_name', 'actor_3_name', 'director_name')
  20. During handling of the above exception, another exception occurred:
  22. KeyError                                  Traceback (most recent call last)
  23. <ipython-input-4-954222273e42> in <module>()
  24. ----> 1 movie['actor_1_name', 'actor_2_name', 'actor_3_name', 'director_name']
  26. /Users/Ted/anaconda/lib/python3.6/site-packages/pandas/core/frame.py in __getitem__(self, key)
  27.    1962             return self._getitem_multilevel(key)
  28.    1963         else:
  29. -> 1964             return self._getitem_column(key)
  30.    1965 
  31.    1966     def _getitem_column(self, key):
  33. /Users/Ted/anaconda/lib/python3.6/site-packages/pandas/core/frame.py in _getitem_column(self, key)
  34.    1969         #  get column
  35.    1970         if self.columns.is_unique:
  36. -> 1971             return self._get_item_cache(key)
  37.    1972 
  38.    1973         #  duplicate columns & possible reduce dimensionality
  40. /Users/Ted/anaconda/lib/python3.6/site-packages/pandas/core/generic.py in _get_item_cache(self, item)
  41.    1643         res = cache.get(item)
  42.    1644         if res is None:
  43. -> 1645             values = self._data.get(item)
  44.    1646             res = self._box_item_values(item, values)
  45.    1647             cache[item] = res
  47. /Users/Ted/anaconda/lib/python3.6/site-packages/pandas/core/internals.py in get(self, item, fastpath)
  48.    3588 
  49.    3589             if not isnull(item):
  50. -> 3590                 loc = self.items.get_loc(item)
  51.    3591             else:
  52.    3592                 indexer = np.arange(len(self.items))[isnull(self.items)]
  54. /Users/Ted/anaconda/lib/python3.6/site-packages/pandas/core/indexes/base.py in get_loc(self, key, method, tolerance)
  55.    2442                 return self._engine.get_loc(key)
  56.    2443             except KeyError:
  57. -> 2444                 return self._engine.get_loc(self._maybe_cast_indexer(key))
  58.    2445 
  59.    2446         indexer = self.get_indexer([key], method=method, tolerance=tolerance)
  61. pandas/_libs/index.pyx in pandas._libs.index.IndexEngine.get_loc (pandas/_libs/index.c:5280)()
  63. pandas/_libs/index.pyx in pandas._libs.index.IndexEngine.get_loc (pandas/_libs/index.c:5126)()
  65. pandas/_libs/hashtable_class_helper.pxi in pandas._libs.hashtable.PyObjectHashTable.get_item (pandas/_libs/hashtable.c:20523)()
  67. pandas/_libs/hashtable_class_helper.pxi in pandas._libs.hashtable.PyObjectHashTable.get_item (pandas/_libs/hashtable.c:20477)()
  69. KeyError: ('actor_1_name', 'actor_2_name', 'actor_3_name', 'director_name')

更多

  1. #  将列表赋值给一个变量，便于多选
  2.  In[6]: cols =['actor_1_name', 'actor_2_name', 'actor_3_name', 'director_name']
  3.         movie_actor_director = movie[cols]
  4. Out[6]: float64    13
  5.         int64       3
  6.         object     11
  7.         dtype: int64
  1. #  使用select_dtypes()，选取整数列
  2.  In[7]: movie.select_dtypes(include=['int']).head()
  3. Out[7]:

1. 选取多个DataFrame列 - 图3

  1. #  选取所有的数值列
  2.  In[8]: movie.select_dtypes(include=['number']).head()
  3. Out[8]:

1. 选取多个DataFrame列 - 图4

  1. #  通过filter()函数过滤选取多列
  2.  In[9]: movie.filter(like='facebook').head()
  3. Out[9]:

1. 选取多个DataFrame列 - 图5

  1. #  通过正则表达式选取多列
  2.  In[10]: movie.filter(regex='\d').head()
  3. Out[10]:

1. 选取多个DataFrame列 - 图6

  1. #  filter()函数，传递列表到参数items，选取多列
  2.  In[11]: movie.filter(items=['actor_1_name', 'asdf']).head()
  3. Out[11]:

1. 选取多个DataFrame列 - 图7