skip to navigation
skip to content

numbyte 2009.12.24.py3k.cpp

numbyte - numerical bytearray - c++ numerical buffer interface extending bytearray into numpy-like, 2d array

Downloads ↓

numbyte - numerical bytearray - c++ numerical buffer interface extending bytearray into numpy-like, 2d array

REQUIRES PYTHON3.1

QUICK TEST: $ python3.1 setup.py build dev --quicktest

DESCRIPTION: numbyte - numerical bytearray - c++ numerical buffer interface extending bytearray into numpy-like, 2d array

SUMMARY: numbyte is a python3.1 extension module, efficiently implementing numerical 2d arrays. numbyte uses python bytearray as the underlying data structure, simplifying io operation with other python objects.

RECENT CHANGELOG: 20091224 - modularized package - fix install issues - added sdist check 20091209 - improved documentation 20091205 - moved source code to c++ 20091116 - package integrated

DEMO USAGE:
>>> from numbyte import *
>>> ## subclass numbyte
>>> class numbyte2(numbyte): pass

>>> ## create bytearray containing 3x4 array of longlong
>>> integers = numbyte2('i', range(12), shape0 = 3, shape1 = 4)
>>> print( integers.debug() )
<class 'numbyte.numbyte2'> i refcnt=4 tcode=i tsize=8 offset=0 shape=<3 4> stride=<4 1> transposed=0
[[          0           1           2           3]
[          4           5           6           7]
[          8           9          10          11]]

>>> ## modify underlying bytearray
>>> integers.bytes()[0] = 0xff; integers.bytes()[1] = 0xff
>>> print( integers.bytes() )
bytearray(b'\xff\xff\x00\x00\x00\x00\x00\x00\x01\x00\x00\x00\x00\x00\x00\x00\x02\x00\x00\x00\x00\x00\x00\x00\x03\x00\x00\x00\x00\x00\x00\x00\x04\x00\x00\x00\x00\x00\x00\x00\x05\x00\x00\x00\x00\x00\x00\x00\x06\x00\x00\x00\x00\x00\x00\x00\x07\x00\x00\x00\x00\x00\x00\x00\x08\x00\x00\x00\x00\x00\x00\x00\t\x00\x00\x00\x00\x00\x00\x00\n\x00\x00\x00\x00\x00\x00\x00\x0b\x00\x00\x00\x00\x00\x00\x00')
>>> print( integers.debug() )
<class 'numbyte.numbyte2'> i refcnt=4 tcode=i tsize=8 offset=0 shape=<3 4> stride=<4 1> transposed=0
[[      65535           1           2           3]
[          4           5           6           7]
[          8           9          10          11]]

>>> ## bytearray as sequence
>>> print( 2 in integers )
True
>>> print( integers.count(2) )
1
>>> print( integers.index(2) )
2
>>> for aa in integers.rows(): print( aa )
[[      65535           1           2           3]]
[[          4           5           6           7]]
[[          8           9          10          11]]
>>> ## slice
>>> print( integers[1:, 2:].debug() )
<class 'numbyte.numbyte2'> i refcnt=3 tcode=i tsize=8 offset=6 shape=<2 2> stride=<4 1> transposed=0
[[          6           7]
[         10          11]]
>>> ## transpose
>>> print( integers.T()[2:, 1:].debug() )
<class 'numbyte.numbyte2'> i refcnt=3 tcode=i tsize=8 offset=6 shape=<2 2> stride=<1 4> transposed=1
[[          6          10]
[          7          11]]
>>> ## reshape
>>> print( integers.reshape(2, -1).debug() )
<class 'numbyte.numbyte2'> i refcnt=3 tcode=i tsize=8 offset=0 shape=<2 6> stride=<6 1> transposed=0
[[      65535           1           2           3           4           5]
[          6           7           8           9          10          11]]
>>> ## setslice
>>> integers.T()[2:, 1:] = range(4); print( integers )
[[      65535           1           2           3]
[          4           5           0           2]
[          8           9           1           3]]

>>> ## almost all arithmetic operations are inplace - use copy to avoid side-effects
>>> ## recast to double
>>> floats = integers.recast('f') / 3; print( floats )
[[        21845      0.333333      0.666667             1]
[      1.33333       1.66667             0      0.666667]
[      2.66667             3      0.333333             1]]
>>> ## copy
>>> print( floats.copy() + integers[0, :] )
[[        87380       1.33333       2.66667             4]
[      65536.3       2.66667             2       3.66667]
[      65537.7             4       2.33333             4]]
>>> ## inplace
>>> print( floats + integers[:, 0] )
[[        87380       65535.3       65535.7         65536]
[      5.33333       5.66667             4       4.66667]
[      10.6667            11       8.33333             9]]
>>> ## inplace
>>> print( floats - integers[:, 0] )
[[        21845      0.333333      0.666667             1]
[      1.33333       1.66667             0      0.666667]
[      2.66667             3      0.333333             1]]
>>> ## inplace
>>> print( floats ** 2 )
[[  4.77204e+08      0.111111      0.444444             1]
[      1.77778       2.77778             0      0.444444]
[      7.11111             9      0.111111             1]]
>>> ## inplace
>>> print( floats.sqrt() )
[[        21845      0.333333      0.666667             1]
[      1.33333       1.66667             0      0.666667]
[      2.66667             3      0.333333             1]]

>>> ## the only inplace exception are logical comparisons, which return new char arrays
>>> print( floats )
[[        21845      0.333333      0.666667             1]
[      1.33333       1.66667             0      0.666667]
[      2.66667             3      0.333333             1]]
>>> ## copy as char
>>> print( floats == floats[:, 1] )
[[ 00  01  00  00]
[ 00  01  00  00]
[ 00  01  00  00]]
>>> ## copy as char
>>> print( floats > 1.5 )
[[ 01  00  00  00]
[ 01  01  00  00]
[ 01  01  00  00]]
 
File Type Py Version Uploaded on Size # downloads
numbyte-2009.12.24.py3k.cpp.tar.gz (md5) Source 2009-12-29 78KB 515