Mailman 3 March 2010 - NumPy-Discussion

Re: [Numpy-discussion] lists of zeros and ones
by Joe Kington March 19, 2010

March 19, 2010

I just realized that permutations isn't quite what you want, as swapping the first "1" for the second "1" gives the same thing. You can use set to get the unique permutations. e.g. In [4]: set(itertools.permutations([1,1,0,0])) Out[4]: set([(0, 0, 1, 1), (0, 1, 0, 1), (0, 1, 1, 0), (1, 0, 0, 1), (1, 0, 1, 0), (1, 1, 0, 0)]) On Fri, Mar 19, 2010 at 10:17 AM, Joe Kington <jkington(a)wisc.edu> wrote: > See itertools.permutations (python standard library) > > e.g. > In [3]: list(itertools.permutations([1,1,0,0])) > Out[3]: > [(1, 1, 0, 0), > (1, 1, 0, 0), > (1, 0, 1, 0), > (1, 0, 0, 1), > (1, 0, 1, 0), > (1, 0, 0, 1), > (1, 1, 0, 0), > (1, 1, 0, 0), > (1, 0, 1, 0), > (1, 0, 0, 1), > (1, 0, 1, 0), > (1, 0, 0, 1), > (0, 1, 1, 0), > (0, 1, 0, 1), > (0, 1, 1, 0), > (0, 1, 0, 1), > (0, 0, 1, 1), > (0, 0, 1, 1), > (0, 1, 1, 0), > (0, 1, 0, 1), > (0, 1, 1, 0), > (0, 1, 0, 1), > > > > (0, 0, 1, 1), > (0, 0, 1, 1)] > > Hope that helps, > -Joe > > On Fri, Mar 19, 2010 at 9:53 AM, gerardob <gberbeglia(a)gmail.com> wrote: > >> >> Hello, i would like to produce lists of lists 1's and 0's. >> >> For example, to produce the list composed of: >> >> L = [[1,0,0,0],[0,1,0,0],[0,0,1,0],[0,0,0,1]] >> >> I just need to do the following: >> >> n=4 >> numpy.eye(n,dtype=int).tolist() >> >> I would like to know a simple way to generate a list containing all the >> lists having two 1's at each element. >> >> Example, n = 4 >> L2 = [[1,1,0,0],[1,0,1,0],[1,0,0,1],[0,1,1,0],[0,1,0,1],[0,0,1,1]] >> >> Any ideas? >> Thanks. >> >> >> >> >> >> >> >> -- >> View this message in context: >> http://old.nabble.com/lists-of-zeros-and-ones-tp27950978p27950978.html >> Sent from the Numpy-discussion mailing list archive at Nabble.com. >> >> _______________________________________________ >> NumPy-Discussion mailing list >> NumPy-Discussion(a)scipy.org >> http://mail.scipy.org/mailman/listinfo/numpy-discussion >> > > > _______________________________________________ > NumPy-Discussion mailing list > NumPy-Discussion(a)scipy.org > http://mail.scipy.org/mailman/listinfo/numpy-discussion > >

1 0

evaluating a function in a matrix element???
by Frank Horowitz March 19, 2010

March 19, 2010

Dear All, I'm working on a piece of optimisation code where it turns out to be mathematically convenient to have a matrix where a few pre-chosen elements must be computed at evaluation time for the dot product (i.e. matrix multiplication) of a matrix with a vector. As I see the problem, there are two basic approaches to accomplishing this. First (and perhaps conceptually simplest, not to mention apparently faster) might be to stash appropriate functions at their corresponding locations in the matrix (with the rest of the matrix being constants, as usual). I mucked around with this for a little while in iPython, and it appears that having dtype == object_ works for stashing the references to the functions, but fails to allow me to actually evaluate the function(s) when the matrix is used in the dot product. Does anyone have any experience with making such a beast work within numpy? If so, how? The second basic approach is to build a ufunc that implements the switching logic, and returns the constants and evaluated functions in the appropriate locations. This seems to be the more do-able approach, but it requires the ufunc to be aware of both the position of each element (via index, or somesuch) as well as the arguments to the functions themselves being evaluated at the matrix elements. It appears that frompyfunc() nearly does what I want, but I am currently failing to see how to actually *use* it for anything more elaborate than the octal example code (i.e. one value in, and one value out). Does anyone have any other more elaborate examples they can point me towards? Thanks in advance for any help you might be able to provide! Frank Horowitz frank(a)horow.net

3 2

evaluating a function in a matrix element???
by Frank Horowitz March 19, 2010

March 19, 2010

> >>> > class X: > > ... def __mul__(self, other): > ... return numpy.random.random() * other > ... def __rmul__(self, other): > ... return other * numpy.random.random() Thanks for that Friedrich! I had forgotten about __mul__, __rmul__, __float__ et al. I think that using a class method instead of a function and overriding those __*__ methods is the cleaner way to do the task I need. Cheers, Frank Horowitz frank(a)horow.net

1 0

peak finding approach
by Davide Cittaro March 18, 2010

March 18, 2010

Hi all, Is there a fast numpy way to find the peak boundaries in a (looong, millions of points) smoothed signal? I've found some approaches, like this: z = data[1:-1] l = data[:-2] r = data[2:] f = np.greater(z, l) f *= np.greater(z, r) boundaries = np.nonzero(f) but it is too sensitive... it detects any small variations in slope on the shoulders of a bigger peak... Any hint? thanks d

2 1

numpy.gradient() does not return ndarray subclasses
by Ryan May March 18, 2010

March 18, 2010

Hi, Can I get someone to look at: http://projects.scipy.org/numpy/ticket/1435 Basically, numpy.gradient() uses numpy.zeros() to create an output array. This breaks the use of any ndarray subclasses, like masked arrays, since the function will only return ndarrays. I've attached a patch that fixes that problem and has a simple test checking output types. With this patch, I can use gradient on masked arrays and get appropriately masked output. If we could, it'd be nice to get this in for 2.0 so that I (and my coworkers who found the bug) don't have to use a custom patched gradient until the next release after that. Thanks, Ryan -- Ryan May Graduate Research Assistant School of Meteorology University of Oklahoma

1 0

ask.scipy.org is online
by Gökhan Sever March 18, 2010

March 18, 2010

Hello, Please tolerate my impatience for being the first announcing the new discussion platform :) and my cross-posting over the lists. The new site is beating at ask.scipy.org . David Warde-Farley is moving the questions from the old-site at advice.mechanicalkern.com (announced at SciPy09 by Robert Kern) We welcome you to join the discussions there. I have kicked-off the new questions chain by http://ask.scipy.org/en/topic/11-what-is-your-favorite-numpy-feature (Also cross-posted at http://stackoverflow.com/questions/2471872/what-is-your-favorite-numpy-feat… pull more attention to ask.scipy) Please share your questions and comments, and have fun with your discussions. -- Gökhan

1 0

Daniel Wright
by Gustavo Mercier March 18, 2010

March 18, 2010

http://carshowcolombia.com/John.html

1 0

Int bitsize in python and c
by Martin Raspaud March 18, 2010

March 18, 2010

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hello, I work on a 64bit machine with 64bits enable fedora on it. I just discovered that numpy.int on the python part are 64bits ints, while npy_int in the C api are 32bits ints. I can live with it, but it seems to be different on 32bit machines, hence I wonder what is the right way to do when retrieving an array from python to C. Here is what I use now: data_pyarray = (PyArrayObject *)PyArray_ContiguousFromObject(data_list, PyArray_INT, 1, 2); but that implies that I send np.int32 arrays to the C part. Should I use longs instead ? Regards, Martin -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org/ iQEcBAEBAgAGBQJLoiuqAAoJEBdvyODiyJI4TikIAIUpnsIxxeYMlz8qEeZL/UUB 3UTGOCcrcIICPVRW/CLbOss5W4xe8BTxPslRXZfckSuMMgHHiD3rGC302gZgfvsb mS6fcDzTOboJ1da1xoczpJYVCwvC9aWAPEjEDa6jyI331pDAXABurmjzIQqjowDw 1cWX5swt9MeSn0yOa/a2EYQP8Xj+n0RQlSIutEDR5jktlK3yyHX8LAtZd0tAPgrd hr9RGwO09Hwcn7ke4B9SwHF7Zg/mBrHgdTdaufW+kjPleZ479lyMO8r/LsWbehVo usQ5wefnmnzhDhOoxff8aKUo8D+Ne8gqxI4BR5EOAdHfQ2uUPpBA91pJ0cNbzZI= =E0XH -----END PGP SIGNATURE-----

3 3

f2py compiler version errors
by Peter Brady March 17, 2010

March 17, 2010

Hello all, The version of f2py that's installed on our system doesn't appear to handle version numbers correctly. I've attached the relevant output of f2py below: customize IntelFCompiler > Couldn't match compiler version for 'Intel(R) Fortran Intel(R) 64 Compiler > Professional for applications running on Intel(R) 64, Version 11.0 Build > 20090318 \nCopyright (C) 1985-2009 Intel Corporation. All rights > reserved.\nFOR NON-COMMERCIAL USE ONLY\n\n Intel Fortran 11.0-1578' > IntelFCompiler instance properties: > archiver = ['ar', '-cr'] > compile_switch = '-c' > compiler_f77 = ['/opt/intel/Compiler/11.0/083/bin/intel64/ifort', '- > 72', '-w90', '-w95', '-KPIC', '-cm', '-O3', '-unroll', > '- > tpp7', '-xW', '-arch SSE2'] > compiler_f90 = ['/opt/intel/Compiler/11.0/083/bin/intel64/ifort', '- > FR', '-KPIC', '-cm', '-O3', '-unroll', '-tpp7', '-xW', > '- > arch SSE2'] > compiler_fix = ['/opt/intel/Compiler/11.0/083/bin/intel64/ifort', '- > FI', '-KPIC', '-cm', '-O3', '-unroll', '-tpp7', '-xW', > '- > arch SSE2'] > libraries = [] > library_dirs = [] > linker_so = ['/opt/intel/Compiler/11.0/083/bin/intel64/ifort', '- > shared', '-tpp7', '-xW', '-arch SSE2'] > object_switch = '-o ' > ranlib = ['ranlib'] > version = None > version_cmd = ['/opt/intel/Compiler/11.0/083/bin/intel64/ifort', '-FI > -V -c /tmp/tmpx6aZa8__dummy.f -o > /tmp/tmpx6aZa8__dummy.o'] > The output of f2py is: Version: 2_3473 > numpy Version: 1.0.1 > Requires: Python 2.3 or higher. > License: NumPy license (see LICENSE.txt in the NumPy source code) > Copyright 1999 - 2005 Pearu Peterson all rights reserved. > http://cens.ioc.ee/projects/f2py2e/ > We're running 64bit linux with python 2.4. How do I make this work? thanks, Peter.

2 1

Setting small numbers to zero.
by gerardob March 17, 2010

March 17, 2010

How can i modified all the values of a numpy array whose value is smaller than a given epsilon to zero? Example epsilon=0.01 a = [[0.003,2][23,0.0001]] output: [[0,2][23,0]] -- View this message in context: http://old.nabble.com/Setting-small-numbers-to-zero.-tp27933569p27933569.ht… Sent from the Numpy-discussion mailing list archive at Nabble.com.

7 14