b1 b2
0 0 1
1 2 3
2 4 5
For example, we have a dataframe a in python, is that possible to get value (b1,2) when you input 4, and get value (b2,2) when input value 5.
Also there's duplicate value in the dataframe, i want to get all their location.
Option 1
one-line
Not recommended!
a.unstack().eq(4).compress(lambda x: x).index.to_series().values[0]
('b1', 2)
Option 2
np.where
i, j = np.where(a == 4)
# or faster with
# i, j = np.where(a.values == 4)
(a.columns[j[0]], a.index[i[0]])
('b1', 2)
Option 3
s = a.unstack()
pd.Series(s.index.values, s.values).loc[4]
('b1', 2)
