Yeah, sorry, this is the output:
In [3]: import numpy
In [4]: import rpy2
In [5]: import rpy2.robjects as robjects
In [7]: marr = robjects.r.array(robjects.IntVector(range(54)),dim=
robjects.IntVector([9,2,3]))
In [8]: marrpy = numpy.array(marr)
In [9]: print marrpy.flatten()
------> print(marrpy.flatten())
[ 0 2 4 9 11 13 1 3 5 10 12 14 2 4 6 11 13 15 3 5 7 12 14 16
4
6 8 13 15 17 5 7 9 14 16 18 6 8 10 15 17 19 7 9 11 16 18 20 8
10
12 17 19 21]
In [10]: print robjects.r['as.vector'](marr)
-------> print(robjects.r['as.vector'](marr))
[1] 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
[26] 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46
47 48 49
[51] 50 51 52 53
In [15]: print marrpy.flatten('F')
-------> print(marrpy.flatten('F'))
[ 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 2 3 4 5 6 7
8
9 10 11 12 13 14 15 16 17 18 19 4 5 6 7 8 9 10 11 12 13 14 15 16
17
18 19 20 21]
I was prepared for differences due to column/row ordering, but this
doesn't seem to be the problem.
Works fine on matrices:
In [11]: mat = robjects.r.matrix(robjects.IntVector(range(54)),nrow=9)
In [12]: matpy = numpy.array(mat)
In [13]: print mat
-------> print(mat)
[,1] [,2] [,3] [,4] [,5] [,6]
[1,] 0 9 18 27 36 45
[2,] 1 10 19 28 37 46
[3,] 2 11 20 29 38 47
[4,] 3 12 21 30 39 48
[5,] 4 13 22 31 40 49
[6,] 5 14 23 32 41 50
[7,] 6 15 24 33 42 51
[8,] 7 16 25 34 43 52
[9,] 8 17 26 35 44 53
In [14]: print matpy
-------> print(matpy)
[[ 0 9 18 27 36 45]
[ 1 10 19 28 37 46]
[ 2 11 20 29 38 47]
[ 3 12 21 30 39 48]
[ 4 13 22 31 40 49]
[ 5 14 23 32 41 50]
[ 6 15 24 33 42 51]
[ 7 16 25 34 43 52]
[ 8 17 26 35 44 53]]
Regards, Robert
-----Original Message-----
From: Laurent Gautier [mailto:lgaut...@gmail.com]
Sent: Monday, 1 November 2010 4:58 PM
To: Denham Robert
Cc: RPy help, support and design discussion list
Subject: Re: [Rpy] converting r array to numpy array
On 01/11/10 07:28, Denham Robert wrote:
> I got caught out converting a 3d array in R to a numpy array. I did
> something like:
>
> import numpy
> import rpy2
> import rpy2.robjects as robjects
>
> marr = robjects.r.array(robjects.IntVector(range(54)),dim=
> robjects.IntVector([9,2,3]))
> marrpy = numpy.array(marr)
> print marrpy.flatten()
> print robjects.r['as.vector'](marr)
>
> Which clearly doesn't work (at least for me, using rpy2.1.7, R-10.0 on
> 64bit suse linux).
Could you be more specific about "does not work" ?
If you are after getting the same output for the two print calls, this
is not happening because arrays are being stored "column-major" by R by
while numpy has "row-major" as a default.
Try:
print marrpy.flatten('F')
> The matrix version does work, eg
>
> mat = robjects.r.matrix(robjects.IntVector(range(54)),nrow=9)
> matpy = numpy.array(mat)
> print mat
> print matpy
>
> I checked the documentation, and found examples for vectors and
> matrices, but not for arrays, but also no warnings. So, I was just
> wondering, is this supposed to work? If it is, a hint on why it
> doesn't work for me would be good. If it isn't supposed to work, I was
> wondering if it were possible to throw an error or something.
It is supposed to work, but not claim about an absence of bugs is made.
Best,
Laurent
> Regards,
> Robert
>
>
>
> Robert Denham
> Environmental Statistician, Remote Sensing Centre Environment and
> Resource Sciences Phone 07 3896 9899 (ext 69899)
> Email: robert.den...@derm.qld.gov.au
> www.derm.qld.gov.au<http://www.derm.qld.gov.au/>
>
> Department of Environment and Resource Management QCCE Building, 80
> Meiers Rd, Indooroopilly QLD 4068
>
>
> +----------------------------------------------------------------+
> Think B4U Print
> 1 ream of paper = 6% of a tree and 5.4kg CO2 in the atmosphere
> 3 sheets of A4 paper = 1 litre of water
> +----------------------------------------------------------------+
>
>
> ----------------------------------------------------------------------
> -------- Nokia and AT&T present the 2010 Calling All Innovators-North
> America contest Create new apps& games for the Nokia N8 for consumers
> in U.S. and Canada $10 million total in prizes - $4M cash, 500
> devices, nearly $6M in marketing Develop with Nokia Qt SDK, Web
> Runtime, or Java and Publish to Ovi Store
> http://p.sf.net/sfu/nokia-dev2dev
> _______________________________________________
> rpy-list mailing list
> rpy-list@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/rpy-list
------------------------------------------------------------------------------
Nokia and AT&T present the 2010 Calling All Innovators-North America contest
Create new apps & games for the Nokia N8 for consumers in U.S. and Canada
$10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing
Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store
http://p.sf.net/sfu/nokia-dev2dev
_______________________________________________
rpy-list mailing list
rpy-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rpy-list
