1. 选取Series数据

  1. #  读取college数据集，查看CITY的前5行
  2.  In[2]: college = pd.read_csv('data/college.csv', index_col='INSTNM')
  3.         city = college['CITY']
  4.         city.head()
  5. Out[2]: INSTNM
  6.         Alabama A & M University                   Normal
  7.         University of Alabama at Birmingham    Birmingham
  8.         Amridge University                     Montgomery
  9.         University of Alabama in Huntsville    Huntsville
  10.         Alabama State University               Montgomery
  11.         Name: CITY, dtype: object
  1. #  iloc可以通过整数选取
  2.  In[3]: city.iloc[3]
  3. Out[3]: 'Huntsville'
  1. #  iloc通过整数列表选取多行，返回结果是Series
  2.  In[4]: city.iloc[[10,20,30]]
  3. Out[4]: INSTNM
  4.         Birmingham Southern College                            Birmingham
  5.         George C Wallace State Community College-Hanceville    Hanceville
  6.         Judson College                                             Marion
  7.         Name: CITY, dtype: object
  1. #  选择等分的数据，可以使用切片语法
  2.  In[5]: city.iloc[4:50:10]
  3. Out[5]: INSTNM
  4.         Alabama State University              Montgomery
  5.         Enterprise State Community College    Enterprise
  6.         Heritage Christian University           Florence
  7.         Marion Military Institute                 Marion
  8.         Reid State Technical College           Evergreen
  9.         Name: CITY, dtype: object
  1. #  loc只接收行索引标签
  2.  In[6]: city.loc['Heritage Christian University']
  3. Out[6]: 'Florence'
  1. #  随机选择4个标签
  2.  In[7]: np.random.seed(1)
  3.         labels = list(np.random.choice(city.index, 4))
  4.         labels
  5. Out[7]: ['Northwest HVAC/R Training Center',
  6.          'California State University-Dominguez Hills',
  7.          'Lower Columbia College',
  8.          'Southwest Acupuncture College-Boulder']
  1. #  通过标签列表选择多行
  2.  In[8]: city.loc[labels]
  3. Out[8]: INSTNM
  4.         Northwest HVAC/R Training Center                Spokane
  5.         California State University-Dominguez Hills      Carson
  6.         Lower Columbia College                         Longview
  7.         Southwest Acupuncture College-Boulder           Boulder
  8.         Name: CITY, dtype: object
  1. #  也可以通过切片语法均匀选择多个
  2.  In[9]: city.loc['Alabama State University':'Reid State Technical College':10]
  3. Out[9]: INSTNM
  4.         Alabama State University              Montgomery
  5.         Enterprise State Community College    Enterprise
  6.         Heritage Christian University           Florence
  7.         Marion Military Institute                 Marion
  8.         Reid State Technical College           Evergreen
  9.         Name: CITY, dtype: object
  1. #  也可以不使用loc，直接使用类似Python的语法
  2.  In[10]: city['Alabama State University':'Reid State Technical College':10]
  3. Out[10]: INSTNM
  4.          Alabama State University              Montgomery
  5.          Enterprise State Community College    Enterprise
  6.          Heritage Christian University           Florence
  7.          Marion Military Institute                 Marion
  8.          Reid State Technical College           Evergreen
  9.          Name: CITY, dtype: object

更多

  1. #  要想只选取一项，并保留其Series类型，则传入一个只包含一项的列表
  2.  In[11]: city.iloc[[3]]
  3. Out[11]: INSTNM
  4.          University of Alabama in Huntsville    Huntsville
  5.          Name: CITY, dtype: object
  1. #  使用loc切片时要注意，如果start索引再stop索引之后，则会返回空，并且不会报警
  2.  In[12]: city.loc['Reid State Technical College':'Alabama State University':10]
  3. Out[12]: Series([], Name: CITY, dtype: object)
  1. #  也可以切片逆序选取
  2.  In[13]: city.loc['Reid State Technical College':'Alabama State University':-10]
  3. Out[13]: INSTNM
  4.          Reid State Technical College           Evergreen
  5.          Marion Military Institute                 Marion
  6.          Heritage Christian University           Florence
  7.          Enterprise State Community College    Enterprise
  8.          Alabama State University              Montgomery
  9.          Name: CITY, dtype: object