Pandas数据清洗

数据清洗是数据分析前的一项预处理工作，想一想，无论从第三方接口获取的数据，还是网站爬取的数据、亦或是从文件、数据库读取的数据。其需要预先分析过滤一遍，比如错误数据、格式错误、数值异常、空缺等等。这一步的操作，才可以让后续的分析、可视化以及更下一步的操作更可靠。

下面，我为大家准备了一个有问题的天气数据，如下：

city,province,area_code,temp_max,temp_min,humidity,wind_power
beijing,beijing,10,31,22,58,3
shanghai,shanghai,21,32,25,72,3
tianjin,tianjin,22,30,NaN,60,3
chongqing,chongqing,23,34,26,70,2
guangzhou,guangdong,20,33,26,78,2
shenzhen,guangdong,755,32,26,80,3
zhuhai,guangdong,756,31,26,79,3
foshan,guangdong,,33,26,77,2
dongguan,guangdong,769,NaN,26,78,2
zhongshan,guangdong,760,32,26,79,2
huizhou,guangdong,752,32,25,77,3
shantou,guangdong,754,31,25,80,3
zhanjiang,guangdong,759,32,26,82,3
nanjing,jiangsu, ,32,24,68,3
suzhou,jiangsu,512,31,24,70,3
wuxi,jiangsu,510,31,24,69,3
changzhou,jiangsu,519,31,24,68,3
NULL,jiangsu,513,30,23,72,3
xuzhou,jiangsu,516,31,22,65,3
yangzhou,jiangsu,514,31,23,69,3
hangzhou,zhejiang,571,33,25,72,2
ningbo,zhejiang,574,31,25,75,3
wenzhou,zhejiang,577,31,25,78,3
jiaxing,zhejiang,573,31,24,74,3
shaoxing,zhejiang,575,32,25,73,2
jinhua,zhejiang,579,33,25,72,2
taizhou_zj,zhejiang,576,31,25,77,3
hefei,anhui,551,32,24,70,3
wuhu,anhui,553,32,24,72,3
bengbu,anhui,552,31,23,69,3
anqing,anhui,556,32,24,73,3
fuzhou,fujian,591,33,26,76,3
xiamen,fujian,592,32,26,79,3
quanzhou,fujian,595,32,26,78,3
zhangzhou,fujian,596,33,26,77,3
nanchang,jiangxi,791,33,25,72,2
ganzhou,jiangxi,797,33,25,74,2
jiujiang,jiangxi,792,32,25,73,3
jinan,shandong,531,31,22,62,3
qingdao,shandong,532,28,22,75,3
yantai,shandong,535,28,21,72,4
weifang,shandong,536,30,21,65,3
linyi,shandong,539,30,21,67,3
zhengzhou,henan,371,32,23,64,3
luoyang,henan,379,31,22,63,3
kaifeng,henan,378,32,23,65,3
xinxiang,henan,373,31,22,64,3
wuhan,hubei,27,33,25,72,2
yichang,hubei,717,32,24,73,2

该数据中故意包含一些错误数据，请大家先简单了解，比如有空缺、有NaN、有NULL值，大家可以先保存代码同级目录下并命名为dotcpp.csv 文件，我们后续会对其处理