python里pandas里面groupby函数用法

import pandas as pd
import  numpy as np

df = pd.DataFrame({'key1':list('aabba'),
                  'key2': ['one','two','one','two','one'],
                  'data1': ['1','3','5','7','9'],
                  'data2': ['2','4','6','8','10']})
print (df)

out:
key1 key2 data1 data2
0    a  one     1     2
1    a  two     3     4
2    b  one     5     6
3    b  two     7     8
4    a  one     9    10

grouped = df.groupby(['key1']).size()
print(grouped)

out:
key1
a    3
b    2
dtype: int64

df["add"] =  1
print(df)

out:
key1 key2 data1 data2  add
0    a  one     1     2    1
1    a  two     3     4    1
2    b  one     5     6    1
3    b  two     7     8    1
4    a  one     9    10    1

grouped = df.groupby(["key1"]).agg('sum').reset_index()   #按key1的值分组，并统计个数
print (grouped)

out:
key1  add
0    a    3
1    b    2

grouped = df.groupby(["key1","data1"]).agg('sum').reset_index()   #按key1的值分组，并统计个数
print (grouped)

out:
key1 data1  add
0    a     1    1
1    a     3    1
2    a     9    1
3    b     5    1
4    b     7    1