d = pd.DataFrame([[1,2,3],[4,5,6]], columns=[‘a‘,‘b‘,‘c‘])
print(d)

import pymysql
import numpy as np
import pandas as pd
from pandas import DataFrame
import datetime
import sys
import pymysql
import csv


db = pymysql.connect(‘localhost‘, ‘root‘, ‘123456‘, ‘languid‘)

if __name__ == ‘__main__‘:
    filepath= sys.argv[1]
    f = pd.read_csv(filepath,engine=‘python‘,sep=‘,‘,index_col=[0],header=[0])#去掉索引
    q=f.drop([‘abc‘,‘deg‘],axis=1)#去掉列abc deg
    qq= q[‘APZWMC‘].str.split(‘-‘,expand=True)#拆分
    qqq=pd.merge(q,qq,on=‘APMAC‘,how=‘right‘)#拼接dataframe 以APMAC列为关键组合列
    qqq[‘sum‘]=qqq[‘APGLDMC‘]+qqq[‘APJWD‘]#这两列都是数字 把他们相加并生成新的一列负在后面
    new_qqq=qqq.rename(columns={0: ‘A‘, 1: ‘B‘}, inplace=False)
    data=[]
    for row in new_qqq.itertuples():#按顺序遍历，将DataFrame迭代为元祖
        row_tuple= [row.Index, row.APMC, row.APZWMC, row.APGLDDM, row.APGLDMC, row.APJWD, row.A,row.B,row.sum]
        data.append(row_tuple)
        print(row)
    cursor = db.cursor()
    sql = ‘INSERT INTO SJJH_APXXB2 (APMAC,APMC,APZWMC,APGLDDM,APGLDMC,APJWD,A,B,sum) values (%s,%s,%s,%s,%s,%s,%s,%s,%s)‘#表在mysql中已经建好了 直接插入
    cursor.executemany(sql,data)
    db.commit()
    db.close()

第二种
import numpy as np
import pandas as pd
from pandas import DataFrame
import datetime
import sys
import pymysql
import csv

from sqlalchemy import create_engine
from sqlalchemy.orm import scoped_session, sessionmaker


engine = create_engine(‘mysql+pymysql://root:123456@localhost/languid?charset=utf8‘)
db = scoped_session(sessionmaker(bind=engine))

aa=[]

if __name__ == ‘__main__‘:
    f = pd.read_csv(‘c://百度网盘//xingqiu.csv‘,engine=‘python‘,sep=‘,‘,index_col=[0],header=[0])
    q=f.drop([‘abc‘,‘deg‘],axis=1)
    qq= q[‘APZWMC‘].str.split(‘-‘,expand=True)
    qqq=pd.merge(q,qq,on=‘APMAC‘,how=‘right‘)
    qqq[‘sum‘]=qqq[‘APGLDMC‘]+qqq[‘APJWD‘]
    new_qqq=qqq.rename(columns={0: ‘A‘, 1: ‘B‘}, inplace=False)
    pd.pivot_table(qqq,index=[‘APMAC‘],values=[‘APZWMC‘],aggfunc=‘count‘)
    for iii in pd.pivot_table:
        print(iii)
   
    pd.io.sql.to_sql(b, ‘sjjh_apxxb‘, engine, schema=‘languid‘, if_exists=‘append‘)#append：如果表存在，则将数据添加到这个表的后面fail：如果表存在就不操作replace：如果存在表，删了，重建
    db.commit()
    db.close()

import pandas as pd
data = pd.read_csv(‘c://百度网盘//movie_metadata.csv‘,engine=‘python‘,sep=‘,‘)
# print(data.head())
print(data.genres.describe())
# print(data[‘genres‘])

#删除不完整的行
data.country=data.country.fillna(‘‘)#用空字符串代替NAN值
data.dropna()#删除任何包含NA值的行
data.dropna(how=‘all‘)#删除一整行的值都为NA
data.drop(theresh=5)#在行数据有至少5个为非na的数据
data.dropna(subset=[‘title_year‘])#去掉上映时间这个数据


#删除不完整的列
data.drop(axis=1,how=‘all‘)#删除一整列为na的列
data.drop(axis=1,how=‘any‘)#删除任何包含空值的列

#必要的变换
data[‘movie_title‘].str.upper() #将title大写
data[‘movei_title‘].str.strip()#去掉末尾空格

#重命名列名
data.rename(columns={‘title_year‘:‘release_dae‘,‘move_facebook_likes‘:‘facebook_likes‘})

#保存结果
data.to_csv(‘cleanfile.csv‘,encoding=‘utf-8‘)

import pandas as pd
# data = pd.read_csv(‘c://百度网盘//movie_metadata.csv‘,engine=‘python‘,sep=‘,‘)

#没有列头
column_names= [‘id‘, ‘name‘, ‘age‘, ‘weight‘,‘m0006‘,‘m0612‘,‘m1218‘,‘f0006‘,‘f0612‘,‘f1218‘]
df = pd.read_csv(‘../data/patient_heart_rate.csv‘, names = column_names)

#一个列有多个参数
df[‘first_name‘,‘last_name‘]=df[‘name‘].str.split(expand=True)
df.drop(‘name‘,axis=1,inplace=True)

#列数据的单位不统一
rows_with_lbs = df[‘weight‘].str.contains(‘lbs‘).fillna(False)
df[rows_with_lbs]
for i,lbs_row in df[rows_with_lbs].iterrows():
    weight=int(float(lbs_row[‘wieght‘][:-3])/2.2)
df.at[i,‘weight‘]=‘{}kgs‘.format(weight)

#空行
df.dropna(how=‘all‘,inplace=True)#删除全空的行

#重复数据
df.drop_duplicates([‘first_name‘,‘last_name‘],inpalec=True)

#非ASCII字符
df[‘first_name‘].replace({r‘[^\x00-\x7F]+‘:‘‘}, regex=True, inplace=True)

#有些列头应该是数据，而不是列名参数

import pandas as pd


df =pd.read_csv(‘C://百度网盘//Artworks.csv‘).head(100)
#时间范围变成时间
row_with_dashes= df[‘Data‘].str.contains(‘-‘).fillna(False)
for i,dash in df[row_with_dashes].iterrows():
    df.at[i,‘data‘] = dash[‘Data‘][0:4]
df[‘Date‘].value_counts

#c.1971变成时间 后四位
row_with_cs=df[‘Data‘].str.contains(‘c‘).fillna(False)
for i,row in df[row_with_cs].iterrrows():
    df.at[i,‘Data‘] = row[‘Date‘][-4:]
df[row_with_cs]

#问题数据赋值为0
df[‘Data‘] = df[‘Date‘].replace(‘Unknown‘,‘0‘,regex=True)
df[‘Data‘] = df[‘Date‘].replace(‘n.d‘,‘0‘,regex=True)

#选择指定单元格
#选取 windspeed 整列数据
print(type(df.loc[:,[‘windspeed‘]]))
#选区列索引为1，2的整行数据
print(type(df.iloc[:,[1,2]]))
#选区索引为1和2的整行数据
print(type(df.loc[1,2]))
#选择windspeed 整列数据
print(type(df[‘windspeed‘]))