Python中pandas的应用

import pandas as pd
pd.set_option('display.unicode.east_asian_width',True)
name1='filmname,type,s_time,e_time,bor'.split(',')
"""读取film1.csv至df1中，跳过标题行，列名重新指定为filmname,type,s_time,e_time,bor ;s_time和e_time列解析为日期类型"""
df1=pd.read_csv('film1.csv',delimiter=',',skiprows=1,names=name1,usecols=[0,1,2,3,4],parse_dates=['s_time','e_time'])
print(df1)
print(df1['e_time']-df1['s_time'])
"""读取film2.csv至df2中，跳过标题行，列名重新指定为filmname,type,s_time,e_time,bor ，s_time和e_time列解析为日期类型"""
df2=pd.read_csv('film2.csv',delimiter=',',skiprows=1,names=name1,usecols=[0,1,2,3,4],parse_dates=['s_time','e_time'])
print(df2)
print(df2['e_time']-df1['s_time'])
"""将df1和df2串联为一个完整的dataframe，名称为df3"""
df3=pd.concat([df1,df2],ignore_index=True)
print(df3)
#""在df1的末尾附加df2,然后保存为df4"""
df4=df1.append(df2,ignore_index=True)
print(df4)
#"""对df3完成原地去重操作"""
df3.drop_duplicates(inplace=True)
print(df3)
"""对于票房为空的有该列的均值进行填充"""
# df3.dropna(inplace=True)
print(df3.bor.mean())
df3['bor'].fillna(df3.bor.mean(),inplace=True)
# df3['bor'].fillna(metod='ffillnplace=True)#前值填充
# df3['bor'].fillna(metod='bfill',inplace=True)#后值填充
print(df3)
"""统计出每部电影的放映天数和日平均票房，并作为新列“放映天数"和“日平均票房”插入df3的末尾，日平均票房计算的逻辑为  票房/放映天数"""
df3['days']=(df3.e_time-df3.s_time).apply(lambda x:x.days).astype('int')+1
df3['day_avg_bor']=df3.bor/df3['days']
print(df3)
"""统计出每种类型电影的总票房,并存入新的对象df3_2中"""
df3_2=df3.groupby(by='type')['bor'].sum()
print(df3_2)
"""读取score.txt至df5中，读取时指定列名为 filmname,userid,pf"""
df5=pd.read_csv('score.txt',delimiter=',',names='filmname,userid,pf'.split(','))
print(df5)
"""将df3和df5按filmname列进行连接成新的df6"""
df6=pd.merge(df3,df5,how='inner',on='filmname')
print(df6)
"""从df6中筛选出 亲密 这部电影的相关信息,并求所有观众对于这部电影的平均评分"""
df7=df6[df6['filmname']=='亲密']
print(df7)
print(df7.pf.mean())