I have two different DataFrames and i wanna get a new DataFrame with the indexes that are different.
DataFrame A:
Name Color
Jony Blue
Mike Red
Joanna Green
DataFrame B:
Name Color
Jony Blue
Mike Red
DataFrame Output:
Name Color
Joanna Green
How can i do to get this DataFrame Output?
Using drop_duplicates
import pandas as pd
dataA = {'Name':['Jony', 'Mike', 'Joanna'], 'Color':['Blue', 'Red', 'Green']}
dataB = {'Name':['Jony', 'Mike'], 'Color':['Blue', 'Red']}
dfA = pd.DataFrame(dataA)
dfB = pd.DataFrame(dataB)
df = pd.concat([dfA, dfB]).drop_duplicates(keep=False, ignore_index=True)
drop_duplicates to the subset you want. The only drawback of this approach is if you have duplicates unique to one of the dataframe and want to keep them in the output.drop_duplicates?Maybe using 'symmetric difference' on a set that you convert to a Series of df ?
dfc=pd.DataFrame()
dfc['Name Color']=pd.Series(list(set(dfa['Name Color']).symmetric_difference(set(dfb['Name Color']))))
One option is to outer-merge with the indicator parameter set to True. Then the common rows will be flagged "both" and since you don't want the common rows, you filter them out:
out = df1.merge(df2, how='outer', indicator=True).query('_merge!="both"').drop(columns='_merge')
Output:
Name Color
2 Joanna Green
Assuming the Name Column is the index in both dataframes:
df_a = pd.DataFrame({'Color': {'Jony': 'Blue', 'Mike': 'Red', 'Joanna': 'Green'}})
df_a = df_a.rename_axis('Name')
df_b = pd.DataFrame({'Color': {'Jony': 'Blue', 'Mike': 'Red'}})
df_b = df_b.rename_axis('Name')
df = pd.concat([df_a[~df_a.index.isin(df_b.index)], df_b[~df_b.index.isin(df_a.index)]])
print(df)
Output:
Color
Name
Joanna Green
