Python Pandas的简单使用教程

df['price']=df['price'].str.replace('人均','') # 删除多余文字
df['price']=df['price'].str.split("￥").str[-1] # 分割文本串
df['price']=df['price'].str.replace('-','0') # 替换文本
df['price']=df['price'].astype(int) # 文本转整型

df['kw']=df['commentlist'].str.split().str[0].str.replace("口味",'')
df['hj']=df['commentlist'].str.split().str[1].str.replace("环境",'')
df['fw']=df['commentlist'].str.split().str[2].str.replace("服务",'')

del df['commentlist']

df.sort_values(by=['kw','price'],axis=0,ascending=[False,True],inplace=True) 

df.columns=['类型','店铺名称','点评数量','星级','人均消费','店铺地址','口味','环境','服务']

print(df.loc[:,['店铺名称','口味','人均消费']].head(6))
# 或者 # print(df.iloc[0:6,[1,6,4]]) # 前6行（整数）
# 但不能是 # print(df.loc[0:6,['店铺名称','口味','人均消费']]) # 从索引0到索引6的行（对象）

import pandas as pd 
df=pd.read_csv('spdata.csv',encoding='gbk')  #读入文件，编码为gbk # 注意编码，重要

#对数据进行清洗
df['comment']=df['comment'].str.replace('条点评','')
df['price']=df['price'].str.replace('人均','')
df['price']=df['price'].str.split("￥").str[-1]
df['price']=df['price'].str.replace('-','0')
df['price']=df['price'].astype(int)
df['kw']=df['commentlist'].str.split().str[0].str.replace("口味",'')
df['hj']=df['commentlist'].str.split().str[1].str.replace("环境",'')
df['fw']=df['commentlist'].str.split().str[2].str.replace("服务",'')
del df['commentlist']

#按口味降序，人均消费升序进行排序
df.sort_values(by=['kw','price'],axis=0,ascending=[False,True],inplace=True) 
#重新设置列索引标签
df.columns=['类型','店铺名称','点评数量','星级','人均消费','店铺地址','口味','环境','服务']

print(df.loc[:,['店铺名称','口味','人均消费']].head(6))

import pandas as pd
df=pd.read_csv('spdata.csv',encoding='gbk')

df['comment']=df['comment'].str.replace('条点评','')
df['price']=df['price'].str.replace('人均','').str.replace('￥','').str.replace('-','0').str.replace(' ','').astype(int)
df[['kw','hj','fw']]=df['commentlist'].str.replace('口味','').str.replace('环境','').str.replace('服务','').str.split(expand=True).astype(float) # expand将普通的列表转为DataFrame对象
del df['commentlist']

df.sort_values(by=['kw','price'],axis=0,ascending=[False,True],inplace=True) # 注意inplace=True
df.columns=['类型','店铺名称','点评数量','星级','人均消费','店铺地址','口味','环境','服务']

print(df[['店铺名称','口味','人均消费']].head(6))