DataFrame.groupby()所见的各种用法详解

import pandas as pd

df = pd.DataFrame({'Gender' : ['男', '女', '男', '男', '男', '男', '女', '女', '女'],
          'name' : ['周杰伦', '蔡依林', '林俊杰', '周杰伦', '林俊杰', '周杰伦', '田馥甄', '蔡依林', '田馥甄'],
          'income' : [4.5, 2.9, 3.8, 3.7, 4.0, 4.1, 1.9, 4.1, 3.2],
         'expenditure' : [1.5, 1.9, 2.8, 1.7, 4.1, 2.5, 1.1, 3.4, 1.2]
         })
#根据其中一列分组
df_expenditure_mean = df.groupby(['Gender']).mean()

#根据其中两列分组
df_expenditure_mean = df.groupby(['Gender', 'name']).mean()

#只对其中一列求均值
df_expenditure_mean = df.groupby(['Gender', 'name'])['income'].mean()

#不以组标签为索引，通过 as_index 来实现
df_expenditure_mean = df.groupby(['Gender', 'name'], as_index=False).mean()

# 使用了 as_index=False，但是从输出结果中可见没起到作用
df_apply = df.groupby(['Gender', 'name'], as_index=False).apply(lambda x: sum(x['income']-x['expenditure'])/sum(x['income']))
df_apply = pd.DataFrame(df_apply,columns=['存钱占比'])#转化成dataframe格式

# 加一句df_apply_index = df_apply.reset_index()
df_apply = df.groupby(['Gender', 'name'], as_index=False).apply(lambda x: sum(x['income']-x['expenditure'])/sum(x['income']))
df_apply = pd.DataFrame(df_apply,columns=['存钱占比'])#转化成dataframe格式
df_apply_index = df_apply.reset_index()

#只对其中一列求均值，并转化为 DataFrame
df_expenditure_mean = df.groupby(['Gender', 'name'], as_index=False)['income'].mean()
df_expenditure_mean = pd.DataFrame(df_expenditure_mean)#转化成dataframe格式
df_expenditure_mean.rename(columns={'income':'收入均值'}, inplace = True)