"id","date","amount"
"1","2019-02-08","6214.23"
"1","2019-02-08","6247.32"
"1","2019-02-09","85.63"
"2","2019-02-14","943.18"
"2","2019-02-15","369.76"
"2","2019-02-18","795.15"
"2","2019-02-19","715.65"
"2","2019-02-21","537.71"
"2","2019-02-24","1037.71"
"3","2019-02-09","967.36"
"3","2019-02-10","85.69"
"3","2019-02-12","769.85"
"3","2019-02-13","943.86"
"3","2019-02-19","843.86"
"3","2019-02-11","85.69"
"3","2019-02-14","843.86"
"1","2019-02-10","985.63"
"1","2019-02-09","285.63"
"1","2019-02-11","1285.63"

第一種常見(jiàn)排序：將上面數(shù)據(jù)按照amount字段進(jìn)行排序

import pandas as pd
filename="data.csv"
df=pd.read_csv(filename)
#增加一個(gè)rank排序字段
df['rank']=df['amount'].rank(ascending=0, method='first')

說(shuō)明：ascending ：1 表示升序，0表示降序

method：此參數(shù)的作用是，當(dāng)遇到兩個(gè)值相同時(shí)，排序處理的方式。可以取的值有 first、max、min、dense

（1）method =‘first’ 時(shí)

表示排序時(shí)，序號(hào)不會(huì)重復(fù)且是連續(xù)的，遇到相同的值時(shí)，會(huì)按照數(shù)據(jù)的先后順序標(biāo)序號(hào)，如下圖：

df['rank']=df['amount'].rank(ascending=0, method='first')
print(df)

（2）method='min’時(shí)

表示排序時(shí)，遇到相同的值時(shí)，后面數(shù)的序號(hào)與最先出現(xiàn)的數(shù)的序號(hào)保持一致，如下圖，843.86值重復(fù)兩次，排名均為9，且排序中沒(méi)有序號(hào)10（序號(hào)不連續(xù)）

df['rank']=df['amount'].rank(ascending=0, method='min')
print(df)

（3）method='max’時(shí)

表示排序時(shí)，遇到相同的值時(shí)，后面數(shù)的序號(hào)與最后出現(xiàn)的數(shù)的序號(hào)保持一致，如下圖，843.86值重復(fù)兩次，排名均為10，且排序中沒(méi)有序號(hào)9（序號(hào)不連續(xù)）

df['rank']=df['amount'].rank(ascending=0, method='max')
print(df)

（4）method='dense’時(shí)

表示排序時(shí)，遇到相同的值時(shí)，重復(fù)值序號(hào)保持一致，如下圖，843.86值重復(fù)兩次，排名均為9，且下一個(gè)數(shù)序號(hào)為10，序號(hào)保持連續(xù)

df['rank']=df['amount'].rank(ascending=0, method='dense')
print(df)

第二種常見(jiàn)排序：組內(nèi)排序 ，將上面數(shù)據(jù)根據(jù)id分組，并按照amount字段進(jìn)行組內(nèi)排序

df['rank']=df['amount'].groupby(df['id']).rank(ascending=0, method='dense')
#對(duì)結(jié)果按照id和rank進(jìn)行升序排列
data=df.sort_values(by=['id','rank'],ascending=(1,1))