我想需要
MultiIndex.from_tuples
从列表理解:
L1 = list('abc')
L2 = list('ghi')
tups = ([('h/a','means', x) for x in L1] +
[('h/a','percentage', x) for x in L2] +
[('totals','means', x) for x in L1] +
[('totals','percentage', x) for x in L2])
columnas=pd.MultiIndex.from_tuples(tups, names=['data level 1','data level 2','data level 3'])
print (columnas)
MultiIndex(levels=[['h/a', 'totals'],
['means', 'percentage'],
['a', 'b', 'c', 'g', 'h', 'i']],
labels=[[0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1],
[0, 0, 0, 1, 1, 1, 0, 0, 0, 1, 1, 1],
[0, 1, 2, 3, 4, 5, 0, 1, 2, 3, 4, 5]],
names=['data level 1', 'data level 2', 'data level 3'])
#some random data
np.random.seed(785)
data = np.random.randint(10, size=(3, 12))
print (data)
[[8 0 4 1 2 5 4 1 4 1 1 8]
[1 5 0 7 4 8 4 1 3 8 0 2]
[5 9 4 9 4 6 3 7 0 5 2 1]]
newframe=pd.DataFrame(data,columns=columnas)
print (newframe)
data level 1 h/a totals
data level 2 means percentage means percentage
data level 3 a b c g h i a b c g h i
0 8 0 4 1 2 5 4 1 4 1 1 8
1 1 5 0 7 4 8 4 1 3 8 0 2
2 5 9 4 9 4 6 3 7 0 5 2 1
