pandas中groupby分组

import numpy as np
import pandas as pd

• 导入模块

df1 = pd.DataFrame({'fruit':['apple','banana','orange','apple','banana'],
                    'color':['red','yellow','yellow','cyan','cyan'],
                    'price':[8.5,6.8,5.6,7.8,6.4]})
df1

• 1.求苹果的平均价格

df2 = df1[df1['fruit'] == 'apple']
df2

• 2.geoupby求苹果平均价格

dict(list(df1.groupby(by='fruit')))['apple'].mean()

• 3.语法糖（每种水果的平均价格）多加一个中括号

df1.groupby(by='fruit')[['price']].mean()

• 4.利用语法糖生成水果的平均值

df1.groupby(by=['fruit','color'])[['price']].mean()

• 转置

df1.groupby(by=['fruit','color'])['price'].mean().unstack()

• 5.统计出现水果的数量

df1['fruit'].value_counts()

df1.groupby(by='fruit')['color'].count()

• 6.每种水果的差价
  

def func(a):
    return a.max() - a.min()
    
df1.groupby('fruit')['price'].apply(func)

df1.groupby('fruit')['price'].agg(func)

df1.groupby('fruit')['price'].aggregate(func)