python - Cheapest way to get a numpy array into C-contiguous order? -

the following produces c-contiguous numpy array:

import numpy  = numpy.ones((1024,1024,5))

now if slice it, result may not longer same. example:

bn = a[:, :, n]

with n 0 4. problem need bn c-contiguous, , need many instances of a. need each bn once, , want avoid doing

bn  = bn.copy(order='c')

i don't want rewrite code such that

a = numpy.ones((5,1024,1024))

is there faster, cheaper way bn doing copy?

background:

i want hash each slice of every a, using

import hashlib  hashlib.sha1(a[:, :, n]).hexdigest()

unfortunately, throw valueerror, complaining order. if there fast way hash want, i'd use it.

as things stand, attempt coerce slice bn c contiguous order going create copy.

if don't want change shapes you're starting (and don't need a in c order), 1 possible solution start array a in fortran order:

>>> = numpy.ones((1024, 1024, 5), order='f')

the slices f-contiguous:

>>> bn = a[:, :, 0] >>> bn.flags   c_contiguous : false   f_contiguous : true   owndata : false   ...

this means transpose of slice bn in c order , transposing not create copy:

>>> bn.t.flags   c_contiguous : true   f_contiguous : false   owndata : false   ...

and can hash slice:

>>> hashlib.sha1(bn.t).hexdigest() '01dfa447dafe16b9a2972ce05c79410e6a96840e'

Ruby Code