pandas做数据的清洗

#coding:utf8
import pandas as pd
import numpy as np
# import csv
df = pd.read_csv(r'D:\MobileFile\fangjia2.csv', encoding='gbk')#读取房价文件
df2 = pd.DataFrame(df)#数据传入
df3 = df2.replace(df2[df2['jiage']<2000],np.nan)#找出jiage里面2000以下的数据替换为空
df4 = df3.dropna(how='any')#删除为空值的行
df5 = df4.drop_duplicates()#去重
print df5.duplicated()#显示是否重复
df5.to_csv(r'D:\MobileFile\fangjia5.csv',encoding='utf-8')#存储新的csv文件