17-pandas处理数据的简单功能

import numpy as np
import pandas as pd
import tushare as ts
 #1.
df=pd.DataFrame({"num":[1,2,3,4,3,2,1,4],
                 "id":["A","B","C","D","C","B","A","D"]})
print(df.duplicated())#列出重复
df.drop_duplicates()#删除重复的副本
df.replace({"C":"Cred","D":"Drop"})#替换副本

#2.
s=pd.Series([1,np.nan,2,np.nan])
s=s.replace({np.nan:0})#替换数据
print(s)

#3.
stock=ts.get_hist_data('600848')
litestock=stock[["high","low"]].head(5)
newline=litestock["low"].map({21.38:1,21.50:2,20.97:3,21.10:4,20.52:5})
litestock["high"]=newline

df=pd.DataFrame({'item':['ball','mug','pen'],
                 'color':['white','rosso','verde']})
price={'ball':5.56,'mug':4.20,'pen':1.30}
df["color"]=df["item"].map(price)#改变一个序列
df["color2"]=df["item"].map(price)#原来有就覆盖，原来没有就新增

#4.
ddd=pd.DataFrame({'zs':[129,130,34],'ls':[136,98,8]})
def mapscore(score):
    if score<60:
        return "不及格"
    elif score>120:
        return "不错"
    else:
        return "得过且过"
ddd["zsisOk"]=ddd["zs"].map(mapscore)#map作用于一维

def mapscore(score):
    return 1
ddd["zstimes"]=ddd["zs"].map(mapscore)
print(ddd)

ddd.rename({0:"one",1:"two",2:"three"})#替换索引