I usually merge the two data frames and define an indicator, then filter out the values where the indicator says both meaning data is in both data frames.
import pandas as pd
mergedData= pd.merge(df1,df2, how='left' ,left_on='Key1', right_on='Key2', indicator ='Exists')
filteredData = mergedData[mergedData[Exists]='both']
与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…