如何匹配熊猫数据框同一列中的两个值

Question

I have two data frames as below

PD  106352  00253   01-02-2018  0.73
PD  108181  00253   20-12-2017  13.91
PD  108222  00253   01-08-2017  -2,227.50
PD  108224  00253   01-08-2017  -4,455.00
PD  108848  00253   25-07-2017  -2,342.13
PD  108852  00253   25-06-2018  1,764.16
PD  108860  00253   12-07-2017  -3,144.81
PD  108871  00253   01-07-2017  -144.17
PD  109455  00253   01-07-2017  -271.25
PD  109472  00253   04-07-2017  -389.00

and

PV  73006   00253   01-09-2017  16,956.25
PV  73006   00253   01-09-2017  2,227.50
PV  73006   00253   01-09-2017  2,227.50
PV  75499   00253   01-07-2017  30,351.42
PV  75645   00253   03-07-2017  34,468.29
PV  82899   00253   12-12-2017  2,342.40

I tried making a list of of the fifth column of both dataframes, compare them, if match found take out the index and used loc to set the result column.but no successfull.

I want to compare the 5th column of both dataframes and match the absolute value ignoring the sign and if 1:1 match found i want to add a column and comment it as nill and if 1:n matches found i want to comment only 1:1 out of them as nill and leave others in the result column as blank

I want to do something like below

PD  108222  00253   01-08-2017  -2,227.50 Nill
PV  73006   00253   01-09-2017  2,227.50  Nill
PV  73006   00253   01-09-2017  2,227.50

Answer 1

Please look at the below code, this is something which i could come up really quick, i think it should solve your problem.

import pandas as pd

#creating data
data_a = pd.read_csv('data_a.csv', sep=',', header=None)
data_a[4]=data_a[4].abs()
data_b = pd.read_csv('data_b.csv', sep=',', header=None)

#converting to list
a=data_a[4].tolist()
b=data_b[4].tolist()


#Removing duplicates and preserving the order so you get 1:1 and not 1:N
b1=[el for i, el in enumerate(b) if el not in b[:i]]

#getting indices of matching values in tow datasets
abc=[i for i, item in enumerate(a) if item in b1]
deg=[i for i, item in enumerate(b1) if item in a]

#Creating blank new column
data_a[5]=''
data_b[5]=''

#Filling matching locations with Nill
data_a.iloc[abc,5] = 'Nill'
data_b.iloc[deg,5] = 'Nill'