多级索引

· 多级index的一级内容或全部内容转到数据内

以index为2个级的数据为例

将0级的数据加入到数据内容

import pandas as pd
import numpy as np
index = pd.MultiIndex.from_tuples([('bird', 'falcon'), ('bird', 'parrot'), ('mammal', 'lion'), ('mammal', 'monkey')], names=['class', 'name'])
columns = pd.MultiIndex.from_tuples([('speed', 'max'), ('species', 'type')])
df = pd.DataFrame([(389.0, 'fly'), ( 24.0, 'fly'), ( 80.5, 'run'), (np.nan, 'jump')], index=index, columns=columns)
print(df)
print('\n')
df1 = df.reset_index(level=0, col_level=0)
print(df1)
print('\n')
df2 = df.reset_index(level=0, col_level=1)
print(df2)
print('\n')

import pandas as pd
import numpy as np
index = pd.MultiIndex.from_tuples([('bird', 'falcon'), ('bird', 'parrot'), ('mammal', 'lion'), ('mammal', 'monkey')], names=['class', 'name'])
columns = pd.MultiIndex.from_tuples([('speed', 'max'), ('species', 'type')])
df = pd.DataFrame([(389.0, 'fly'), ( 24.0, 'fly'), ( 80.5, 'run'), (np.nan, 'jump')], index=index, columns=columns)
print(df)
print('\n')
df0 = df.reset_index(level=0, col_level=0)
print(df0)
print('\n')
df1 = df.reset_index(level=0, col_level=0, col_fill=None)
print(df1)
print('\n')
df2 = df.reset_index(level=0, col_level=1, col_fill='species')
print(df2)
print('\n')
df3 = df.reset_index(level=0, col_level=0, col_fill='genus')
print(df3)
print('\n')

将1级的数据加入到数据内容

import pandas as pd
import numpy as np
index = pd.MultiIndex.from_tuples([('bird', 'falcon'), ('bird', 'parrot'), ('mammal', 'lion'), ('mammal', 'monkey')], names=['class', 'name'])
columns = pd.MultiIndex.from_tuples([('speed', 'max'), ('species', 'type')])
df = pd.DataFrame([(389.0, 'fly'), ( 24.0, 'fly'), ( 80.5, 'run'), (np.nan, 'jump')], index=index, columns=columns)
print(df)
print('\n')
df0 = df.reset_index()
print(df0)
print('\n')
df1 = df.reset_index(level=1)
print(df1)
print('\n')
df2 = df.reset_index(level='name')
print(df2)

import pandas as pd
import numpy as np
index = pd.MultiIndex.from_tuples([('bird', 'falcon'), ('bird', 'parrot'), ('mammal', 'lion'), ('mammal', 'monkey')], names=['class', 'name'])
columns = pd.MultiIndex.from_tuples([('speed', 'max'), ('species', 'type')])
df = pd.DataFrame([(389.0, 'fly'), ( 24.0, 'fly'), ( 80.5, 'run'), (np.nan, 'jump')], index=index, columns=columns)
print(df)
print('\n')
df0 = df.reset_index(level=1, col_level=0)
print(df0)
print('\n')
df1 = df.reset_index(level=1, col_level=0, col_fill=None)
print(df1)
print('\n')
df2 = df.reset_index(level=1, col_level=1, col_fill='species')
print(df2)
print('\n')
df3 = df.reset_index(level=1, col_level=0, col_fill='genus')
print(df3)
print('\n')

· 使用列表添加第二级索引对列进行手动分组

import pandas as pd
data = {"apple": [7,5,6,4,7,5,8,6],
       "strawberry": [3,5,2,1,3,0,4,2],
       "banana": [1,2,1,2,2,2,1,3],
        "chocolate" : [5,8,4,2,1,6,4,5],
        "cake":[4,4,5,1,3,0,0,3]
       }
data = pd.DataFrame(data)
food_cat = ["fv","fv","fv","j","j"]
data.columns = pd.MultiIndex.from_arrays([food_cat, data.columns])
data

参考资料：https://www.5axxw.com/questions/content/nnnrx3