how to drop duplicates in pandas when entries are sets

Question

I have a series whose entries are sets. I want to remove all duplicate entries, using pandas.Series.drop_duplicates() but get an error. Here is an example:

import pandas as pd
ser = pd.Series([{1,2,3}, {4,5,6}, {4,5,6}])
ser.drop_duplicates()

The last line gives the following exception:

TypeError: unhashable type: 'set'

Whereas I would like to get:

0    {1, 2, 3}
1    {4, 5, 6}

Is this a bug? Or is there another way to acheive this?

Answer 1

Let us using astype(str) then duplicated

ser[~ser.astype(str).duplicated(keep='first')]
Out[170]: 
0    {1, 2, 3}
1    {4, 5, 6}
dtype: object

More Info :

ser.astype(str).duplicated(keep='first')
Out[171]: 
0    False
1    False
2     True
dtype: bool