For the groupby aggregation function of the dataframe, we properly understand the syntactic sugar, which will have a multiplier effect on data analysis.

Iterate over the group

First look at the type of each field

import numpy as np
import pandas as pd
import pymysql
conn = pymysql.connect(host='localhost', user='root', passwd='123456', db='test', port=3306, charset='utf8')
jianshu = pd.read_sql('select * from jianshu1',conn)
jianshu.dtypes

It can be seen that these fields of view are integer type data, but here is object data, so we need to modify the data type, take view as an example.

jianshu['view'] = jianshu['view'].astype('int64')
jianshu.dtypes

Use the user column as an index.

jianshu.set_index('user',inplace=True)
jianshu

The grouped data type is a groupby object, which can be iterated.

jianshu.groupby(jianshu.index)

for name,group in jianshu.groupby(jianshu.index):
    print(name,group)

Syntactic sugar one: select one or more columns

jianshu.groupby(jianshu.index)[['view']].sum()

Syntactic sugar 2: data aggregation through aggregate or agg method

jianshu.groupby(jianshu.index)[['view']].agg(['mean','sum'])