ENH: interleave_columns function

[raybellwaves]

ENH: interleave_columns function

[asishm]

what about df.T.melt()? you can then either set_index the variable column or drop it altogether.

In [10]: df = pd.DataFrame({0: ['A1', 'A2', 'A3'], 1: ['B1', 'B2', 'B3'], 2: ['C1', 'C2', 'C3']})

In [11]: df
Out[11]:
    0   1   2
0  A1  B1  C1
1  A2  B2  C2
2  A3  B3  C3

In [12]: df.T.melt()
Out[12]:
   variable value
0         0    A1
1         0    B1
2         0    C1
3         1    A2
4         1    B2
5         1    C2
6         2    A3
7         2    B3
8         2    C3

[phofl]

Yep agree that this should solve the problem

[raybellwaves]

I ended up finding out how to do this in the cudf test suite https://github.com/rapidsai/cudf/blob/branch-24.06/python/cudf/cudf/tests/test_reshape.py#L305 pd.Series(np.vstack(pdf.to_numpy()).reshape((-1,))) which is much faster than df.T.melt()["value"]. Thanks for tip though!