分组及统计

import math
def  add_label(data,k):
    label = [math.ceil(i/k) for i in range(1,len(df)+1)]
    return label


df = df.sort_values(by=['age'],ascending=False)

df['group'] = add_label(df,k=10)

#df['group'] = df['group'].apply(str)#不需要改成字符，也能groupby
#看一下每组是不是10个
print(df.head(1))


t = pd.DataFrame(df.groupby(['group']).age.sum())#group是index，直接产生字段age
t['avg'] = pd.DataFrame(df.groupby(['group']).age.mean())#生成新的字段
t = t.rename(columns={'age': 'age_sum'})#改个字段名
print(t.columns)
print(t)