Re: [Pytables-users] Use of recarrays as representation for Tables in memory

[Alvaro T. C. <al...@mi...>]

Thank you Josh, that is representative enough. In my system the
speedup of structured arrays is ~30x. A copy of the whole array is
still ~6x faster.

-á.


On Thu, Jun 28, 2012 at 10:13 PM, Josh Ayers <jos...@gm...> wrote:
> import time
> import numpy as np
>
> dtype = np.format_parser(['i4', 'i4'], [], [])
> N = 100000
> rec = np.recarray((N, ), dtype=dtype)
> struc = np.zeros((N, ), dtype=dtype)
>
> t1 = time.clock()
> for row in rec:
>     pass
> print time.clock() - t1
>
> t1 = time.clock()
> for row in struc:
>     pass
> print time.clock() - t1