数据比对去重

import pandas as pd
import numpy as np
from collections import Counter

df = pd.read_excel(r'E:\Anaconda3\work\csv.xlsx')

# 求出a列存在,b列不存在的数据。
ab = [i for i in set(df["a"]) - set(df["b"]) if i is not np.nan]
# 求出b列存在,a列不存在的数据。
ba = [i for i in set(df["b"]) - set(df["a"]) if i is not np.nan]

a = Counter(df["a"])
b = Counter(df["b"])
# 统计出现的个数,以字典形式返回

xa = []
xb = []
ji = a.keys() & b.keys()
for i in ji:
    ai = a.get(i)
    bi = b.get(i)
    if ai > bi:
        count = ai-bi
        for k in range(count):
            xa.append(i)
    elif bi > ai:
        count = bi-ai
        for k in range(count):
            xb.append(i)

for val in ab:
    sa = a.get(val)
    if sa !=None:
        for coun in range(sa):
            xa.append(val)
for val in ba:
    sb = b.get(val)
    if sb !=None:
        for coun in range(sb):
            xb.append(val)
ds = pd.DataFrame([xa,xb]).T
ds.columns = ['xa','xb']
ds.to_excel(r'E:\Anaconda3\work\result.xlsx',index=False)

来源之知乎

https://www.zhihu.com/question/385996087/answer/1139595963