python - Replace DataFrame column index with MultiIndex -
this question has answer here:
- convert dataframe columns multiindex 1 answer
i have pandas dataframe "flat" column index, ['sample 1', 'sample 2' ...], , i've constructed multiindex has 2 levels, first of having same elements dataframe's column index ('sample 1' ...). want replace column index of dataframe multiindex, each of elements of multiindex replacing column index element has matching name. so, column headed 'sample 1' should headed ('sample 1', 'group x'), column heading 'sample 2' should ('sample 2', 'group x'), , forth.
for example, dataframe might like:
b c row1 1 2 3 row2 4 5 6 : rown 7 8 9
and multiindex [('a', 'g1'), ('b', 'g1'), ('c', 'g2')]
with end result looking like:
g1 g1 g2 b c row1 1 2 3 row2 4 5 6 : rown 7 8 9
seems should simple join or matching concatenation, can't find method (and/or keywords) it.
just assign columns
instance attribute
in [10]: df = dataframe(randn(10, 3), columns=list('abc')) in [11]: df.columns = [list('abc'), df.columns] in [12]: df out[12]: b c b c 0 2.320 -1.670 0.511 1 0.350 -1.877 0.627 2 1.671 0.347 1.328 3 0.896 -0.061 0.083 4 1.168 1.212 -1.127 5 -0.015 0.402 0.249 6 -0.576 0.898 -0.464 7 -0.939 0.198 -2.698 8 -0.767 0.532 -0.369 9 0.365 0.293 -0.676 in [13]: df.columns out[13]: multiindex [(u'a', u'a'), (u'b', u'b'), (u'c', u'c')]
Comments
Post a Comment