pandas 设置索引

• pd.set_index()

• 将某一列作为索引

import pandas as pd
import numpy as np
df = pd.DataFrame({'country':['China','China', 'India', 'India', 'America', 'Japan', 'China', 'India'], 
 
                   'id':[10000, 10000, 5000, 5002, 40000, 50000, 8000, 5000],
 
                    'age':[50, 43, 34, 40, 25, 25, 45, 32]})
df

# 将country这一列作为索引
# inplace=True 表示当前操作对原数据生效
# append=False 表示原来的行索引被舍弃
df.set_index('country',drop=True, append=False, inplace=True)

• 设置复合索引

• 设置复合索引 源码

import pandas as pd
import numpy as np

# 以字典的的格式取创建DataFrane
a=pd.DataFrame({
    'a' : range(7),
    'b' : range(7, 0, -1),
    'c' : ['one', 'one', 'one', 'two',  'two', 'two', 'two'],
    'd' : list('hjklmmo')
})
a

# 设置c d两列为复合索引, 然后取出a这一列
x = a.set_index(['c','d'])['a']
x

# Series
type(x)

# 设置c d两列为复合索引, 然后取出a这一列
x = a.set_index(['c','d'])[['a']]
x

# DataFrame
type(x)