当行以不同顺序具有相同值时删除行

Question

I would like an efficient way to drop rows from a Pandas dataframe if rows have same values regardless of order

For example, this:

1    3
2    5
5    2
1    3
3    1

becomes:

1    3
5    2

Answer 1

You could sort the rows of a copy of the dataframe, drop_duplicates and use the resulting indices to index the dataframe:

df_ = df.copy()
df_.values.sort(1)
df = df.loc[df_.drop_duplicates().index, :]
# del df_

    col1  col2
0     1     3
1     2     5