python - Error when indexing with 2 dimensions in NumPy -

why work:

>>> (tf[:,[91,1063]])[[0,3,4],:] array([[ 0.04480133,  0.01079433],        [ 0.11145042,  0.        ],        [ 0.01177578,  0.01418614]])

but not:

>>> tf[[0,3,4],[91,1063]] indexerror: shape mismatch: indexing arrays not broadcast shapes (3,) (2,)

what doing wrong?

tf[:,[91,1063]])[[0,3,4],:]

operates in 2 steps, first selecting 2 columns, , 3 rows result

tf[[0,3,4],[91,1063]]

tries select tf[0,91], tf[3,1063] , ft[4, oops].

tf[[[0],[3],[4]], [91,1063]]

should work, giving same result first expression. think of 1st list being column, selecting rows.

tf[np.array([0,3,4])[:,newaxis], [91,1063]]

is way of generating column index array

tf[np.ix_([0,3,4],[91,1063])]

np.ix_ can generate these index arrays.

in [140]: np.ix_([0,3,4],[91,1063]) out[140]:  (array([[0],         [3],         [4]]), array([[  91, 1063]]))

these column , row arrays broadcast produce 2d array of coordinates

[[(0,91), (0,1063)]  [(3,91), ...     ]  ....             ]]

Braziel