6. 惰性行切片

  1. #  读取college数据集；从行索引10到20，每隔一个取一行
  2.  In[50]: college = pd.read_csv('data/college.csv', index_col='INSTNM')
  3.          college[10:20:2]
  4. Out[50]:

6. 惰性行切片 - 图1

  1. #  Series也可以进行同样的切片
  2.  In[51]: city = college['CITY']
  3.          city[10:20:2]
  4. Out[51]: INSTNM
  5.          Birmingham Southern College              Birmingham
  6.          Concordia College Alabama                     Selma
  7.          Enterprise State Community College       Enterprise
  8.          Faulkner University                      Montgomery
  9.          New Beginning College of Cosmetology    Albertville
  10.          Name: CITY, dtype: object
  1. #  查看第4002个行索引标签
  2.  In[52]: college.index[4001]
  3. Out[52]: 'Spokane Community College'
  1. #  Series和DataFrame都可以用标签进行切片。下面是对DataFrame用标签切片
  2.  In[53]: start = 'Mesa Community College'
  3.          stop = 'Spokane Community College'
  4.          college[start:stop:1500]
  5. Out[53]:

6. 惰性行切片 - 图2

  1. #  下面是对Series用标签切片
  2.  In[54]: city[start:stop:1500]
  3. Out[54]: INSTNM
  4.          Mesa Community College                            Mesa
  5.          Hair Academy Inc-New Carrollton         New Carrollton
  6.          National College of Natural Medicine          Portland
  7.          Name: CITY, dtype: object

更多

惰性切片不能用于列，只能用于DataFrame的行和Series，也不能同时选取行和列。

  1. #  下面尝试选取两列，导致错误
  2.  In[55]: college[:10, ['CITY', 'STABBR']]
  3. ---------------------------------------------------------------------------
  4. TypeError                                 Traceback (most recent call last)
  5. <ipython-input-55-92538c61bdfa> in <module>()
  6. ----> 1 college[:10, ['CITY', 'STABBR']]
  8. /Users/Ted/anaconda/lib/python3.6/site-packages/pandas/core/frame.py in __getitem__(self, key)
  9.    1962             return self._getitem_multilevel(key)
  10.    1963         else:
  11. -> 1964             return self._getitem_column(key)
  12.    1965 
  13.    1966     def _getitem_column(self, key):
  15. /Users/Ted/anaconda/lib/python3.6/site-packages/pandas/core/frame.py in _getitem_column(self, key)
  16.    1969         #  get column
  17.    1970         if self.columns.is_unique:
  18. -> 1971             return self._get_item_cache(key)
  19.    1972 
  20.    1973         #  duplicate columns & possible reduce dimensionality
  22. /Users/Ted/anaconda/lib/python3.6/site-packages/pandas/core/generic.py in _get_item_cache(self, item)
  23.    1641         """Return the cached item, item represents a label indexer."""
  24.    1642         cache = self._item_cache
  25. -> 1643         res = cache.get(item)
  26.    1644         if res is None:
  27.    1645             values = self._data.get(item)
  29. TypeError: unhashable type: 'slice'
  1. #  只能用.loc和.iloc选取
  2.  In[56]: first_ten_instnm = college.index[:10]
  3.          college.loc[first_ten_instnm, ['CITY', 'STABBR']]
  4. Out[56]:

6. 惰性行切片 - 图3