Maybe I missed something, but I didn't find anything like this in numpy
a kind of apply along 2axis, iterates over all 2d arrays defined by
axes argument in a n-dim array with n>2
useful for functions that operate on a 2d array (chisquare_twoway,
np.corrcoef ? ...)
awful function name
Josef
def allbut2ax_iterator(table, axes=(-2,-1)):
axes = np.arange(table.ndim)[list(axes)]
othax = [i for i in np.arange(table.ndim) if not i in axes]
othshape = np.array(table.shape)[othax]
for othind in np.ndindex(*othshape):
index = np.array([slice(None)]*table.ndim)
index[othax] = othind
yield index.tolist()
>>> for i in allbut2ax_iterator(np.arange(3*2*4*4).reshape(3,2,4,4),
>>> axes=(-2,-1)): print i
[0, 0, slice(None, None, None), slice(None, None, None)]
[0, 1, slice(None, None, None), slice(None, None, None)]
[1, 0, slice(None, None, None), slice(None, None, None)]
[1, 1, slice(None, None, None), slice(None, None, None)]
[2, 0, slice(None, None, None), slice(None, None, None)]
[2, 1, slice(None, None, None), slice(None, None, None)]
>>> for i in allbut2ax_iterator(np.arange(3*2*4*4).reshape(3,2,4,4),
>>> axes=(0,3)): print i
[slice(None, None, None), 0, 0, slice(None, None, None)]
[slice(None, None, None), 0, 1, slice(None, None, None)]
[slice(None, None, None), 0, 2, slice(None, None, None)]
[slice(None, None, None), 0, 3, slice(None, None, None)]
[slice(None, None, None), 1, 0, slice(None, None, None)]
[slice(None, None, None), 1, 1, slice(None, None, None)]
[slice(None, None, None), 1, 2, slice(None, None, None)]
[slice(None, None, None), 1, 3, slice(None, None, None)]
>>> for i in allbut2ax_iterator(np.arange(3*2*4*4).reshape(3,2,4,4),
>>> axes=(1,2)): print i
[0, slice(None, None, None), slice(None, None, None), 0]
[0, slice(None, None, None), slice(None, None, None), 1]
[0, slice(None, None, None), slice(None, None, None), 2]
[0, slice(None, None, None), slice(None, None, None), 3]
[1, slice(None, None, None), slice(None, None, None), 0]
[1, slice(None, None, None), slice(None, None, None), 1]
[1, slice(None, None, None), slice(None, None, None), 2]
[1, slice(None, None, None), slice(None, None, None), 3]
[2, slice(None, None, None), slice(None, None, None), 0]
[2, slice(None, None, None), slice(None, None, None), 1]
[2, slice(None, None, None), slice(None, None, None), 2]
[2, slice(None, None, None), slice(None, None, None), 3]
>>> for i in allbut2ax_iterator(np.arange(2*4*4).reshape(2,4,4), axes=(1,2)):
>>> print i
[0, slice(None, None, None), slice(None, None, None)]
[1, slice(None, None, None), slice(None, None, None)]
>>> for i in allbut2ax_iterator(np.arange(2*4*4).reshape(2,4,4), axes=(0,2)):
>>> print i
[slice(None, None, None), 0, slice(None, None, None)]
[slice(None, None, None), 1, slice(None, None, None)]
[slice(None, None, None), 2, slice(None, None, None)]
[slice(None, None, None), 3, slice(None, None, None)]
>>> for i in allbut2ax_iterator(np.arange(2*4).reshape(2,4), axes=(0,1)): print
>>> i
[slice(None, None, None), slice(None, None, None)]
>>> table = np.arange(2*4*4).reshape(2,4,4)
>>> for i in allbut2ax_iterator(table, axes=(1,2)): print table[i]
[[ 0 1 2 3]
[ 4 5 6 7]
[ 8 9 10 11]
[12 13 14 15]]
[[16 17 18 19]
[20 21 22 23]
[24 25 26 27]
[28 29 30 31]]
_______________________________________________
NumPy-Discussion mailing list
NumPy-Discussion@scipy.org
http://mail.scipy.org/mailman/listinfo/numpy-discussion
