比较2 df的df1and df2

Question

比较2 df的df1有重复的1000+ _id的唯一行，但df2只有唯一的唯一id。我想要比较df1中的每一行，这样unique_id就在df2中，如果匹配，也可以从df1到df2比较相同的类别和子类别。输出:如果其中任何一个不匹配，则将该索引提取到一个数组中。

import pandas as pd
import numpy as np

data1 = {'unique_id': 
        ['Computer','iPhone','Printer','Desktop','Computer','iPhone','iphpne','Printer','Desktop','Computer','iPhone','Printer','Desktop'],
        'category': 
         ['movies','documentary','series','special','movies','documentary','series','special','series','special','movies','series','special'],
         'subcategory':
         ['drama','horror','comedy','reality','drama','documentary','comedy','reality','documentary','comedy','documentary','comedy','drama']
        }

df1 = pd.DataFrame(data1,columns= ['unique_id', 'category','subcategory'])


data2 = {'unique_id': ['Computer','iPhone','Printer','Desktop'],
         'category': ['movies','documentary','series','special'],
         'subcategory':['drama','horror','comedy','reality']
        }

df2 = pd.DataFrame(data2,columns= ['unique_id', 'category','subcategory'])

Answer 1

IIUC，这就是你需要的

pd.concat([df1,df2]).drop_duplicates(keep=False)

打印：

   unique_id     category  subcategory
5     iPhone  documentary  documentary
6     iphpne       series       comedy
7    Printer      special      reality
8    Desktop       series  documentary
9   Computer      special       comedy
10    iPhone       movies  documentary
12   Desktop      special        drama

获取索引的

pd.concat([df1,df2]).drop_duplicates(keep=False).index

打印：

   Int64Index([5, 6, 7, 8, 9, 10, 12], dtype='int64')