Mailman 3 September 2013 - NumPy-Discussion

Unexpected casting result
by Warren Weckesser 16 Sep '13

16 Sep '13

An unexpected casting result was just reported on stackoverflow: http://stackoverflow.com/questions/18833639/attributeerror-in-python-numpy-… The following show the essence of the issue: In [1]: np.__version__ Out[1]: '1.9.0.dev-6ce65d8' In [2]: type(np.array(1.) * (2**64-1)) Out[2]: numpy.float64 In [3]: type(np.array(1.) * (2**64)) Out[3]: float Note that the result of `np.array(1.0) * 2**64` is a Python float, not a numpy float64. Is this intentional? (As pointed out in the stackoverflow question, the issue https://github.com/numpy/numpy/issues/3409 is at least tangentially related.) Warren

1 1

[ANN] OpenOpt suite v 0.51
by Dmitrey 15 Sep '13

15 Sep '13

Hi all, new OpenOpt suite v 0.51 has been released: Some improvements for FuncDesigner automatic differentiation and QP FuncDesigner now can model sparse (MI)(QC)QP Octave QP solver has been connected MATLAB solvers linprog ( LP ), quadprog ( QP ), lsqlin ( LLSP ), bintprog ( MILP ) New NLP solver: knitro Some elements of 2nd order interval analysis, mostly for interalg Some interalg improvements interalg can directly handle (MI)LP and (possibly nonconvex) (MI)(QC)QP New classes: knapsack problem ( KSP ), bin packing problem ( BPP ), dominating set problem ( DSP ) FuncDesigner can model SOCP SpaceFuncs has been adjusted for recent versions of Python and NumPy visit http://openopt.org for more details. Regards, D.

1 0

ANN: 1.8.0b2 release.
by Charles R Harris 14 Sep '13

14 Sep '13

Hi all, I'm happy to announce the second beta release of Numpy 1.8.0. This release should solve the Windows problems encountered in the first beta. Many thanks to Christolph Gohlke and Julian Taylor for their hard work in getting those issues settled. It would be good if folks running OS X could try out this release and report any issues on the numpy-dev mailing list. Unfortunately the files still need to be installed from source as dmg files are not avalable at this time. Source tarballs and release notes can be found at https://sourceforge.net/projects/numpy/files/NumPy/1.8.0b2/. The Windows and OS X installers will follow when the infrastructure issues are dealt with. Chuck

8 20

Re: [Numpy-discussion] can argmax be used to return row and column indices?
by Mark Bakker 13 Sep '13

13 Sep '13

Thanks, Gregorio. I would like it if argmax had a keyword option to return the row,column index automatically (or whatever the dimension of the array). Afterall, argmax already knows the shape of the array. Calling np.unravel_index( np.argmax( A ) ) seems unnecessarily long. But it works well though! I am not sure that such a PR would get much support.... Thanks again, Mark Hi Mark, > > You're looking for "np.unravel_index" function. > > Cheers, > Gregorio > Hello list, > > I am trying to find the indices of the maximum value in a 2D array. > argmax works fine, but returns the index in the flattened array. > That is often not very convenient. > Is there a function that returns the index of the row and column? > Or can the index of the flattened array easily be converted to the row can > column indices (I know, I can write a function for that, but I figure numpy > already has one). > > Thanks, > > Mark >

3 2

[ANN] Introductory Pytroll Workshop, Nov. 25-26 2013, at SMHI, Sweden
by Martin Raspaud 13 Sep '13

13 Sep '13

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 [sorry for crossposting] Hi all, After the success of last years workshop, we decided to hold the Pytroll workshop this year again at SMHI. Pytroll is a collection of free and open source python modules for the reading, interpretation, and writing of weather satellite data. The provided python packages are designed to be used both in R&D environments and in 24/7 operational production. More info at: www.pytroll.org This time, it will be a two days introductory workshop, during which we will present the different packages of Pytroll, and help you get started with the installation and the different applications you might have. Preliminary program: Day 1: - - Presentation of the different pytroll components - - Installation of pytroll on the participants' computer Day 2: - - Illustration with different test cases and/or user defined applications brought by the participant. Please tell us if you plan to attend, so we can plan effectively. For any more information, have a look at last year's workshop's webpage: http://www.pytroll.org/workshop2012.html or contact us with any questions directly or on the pytroll googlegroup. Best regards, Martin - -- You received this message because you are subscribed to the Google Groups "python meteorology" group. To unsubscribe from this group and stop receiving emails from it, send an email to python-meteorology+unsubscribe(a)googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.14 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJSMv1PAAoJEBdvyODiyJI4oiUIALqh3tdOov2Ykh6BLpfQRYdL BMw7BcHGZ81CXyldgQHQF4OROskce2od/FC18L3vgQrqFEI6uZ8hLqlhk3LvXrnP ToJsERolpcicAdcsMcMK2l/0ePUYfOS+Yu4pck4H1+skNlbBcBg98uW8ECXwe7Ks mhtQkcdNNSiEcCIe8/Yxb4Tbb6udM7CBBedJ7Xa5CmV5ZsC7lPKysOlNUZppTiBY +w5bfDRlPKEQHV6AkLZhyv6eYcd9Ncv5utcnRKICFbmXWtvpSC1c+zlB7vgAU7Um SU4f9UgsQw/Y+cd6OzxuQtsOrPO11NYXBFm0v8IcodGcgCYVhjVU6asQ32UFmTM= =XOzN -----END PGP SIGNATURE-----

1 0

[ANN] Introductory Pytroll Workshop, Nov. 25-26 2013, at SMHI, Sweden
by Martin Raspaud 13 Sep '13

13 Sep '13

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 [sorry for crossposting] Hi all, After the success of last years workshop, we decided to hold the Pytroll workshop this year again at SMHI. Pytroll is a collection of free and open source python modules for the reading, interpretation, and writing of weather satellite data. The provided python packages are designed to be used both in R&D environments and in 24/7 operational production. More info at: www.pytroll.org This time, it will be a two days introductory workshop, during which we will present the different packages of Pytroll, and help you get started with the installation and the different applications you might have. Preliminary program: Day 1: - - Presentation of the different pytroll components - - Installation of pytroll on the participants' computer Day 2: - - Illustration with different test cases and/or user defined applications brought by the participant. Please tell us if you plan to attend, so we can plan effectively. For any more information, have a look at last year's workshop's webpage: http://www.pytroll.org/workshop2012.html or contact us with any questions directly or on the pytroll googlegroup. Best regards, Martin -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.14 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJSMvosAAoJEBdvyODiyJI4LfkH/iisXrIEEIHhPgDdKPFtqOrJ q6pFq70RkgXwDVWC3AoRfeE9eggO0mE2RZOPL16I9a+1RE/HFKz6Az4sVI5UtC0Z nAU+pJd1KOzEBPBn59LQtU6+wX0rOI5UCA6yPMmR3DxkDDLhbrHLeebwLgVQb+gG gVMl4ADdRFbt5nRkAs2/I/bD3gs3AxMwrSuXgP8hgF8IDyE/w1ZmwHivhEHHFup8 /AB2411rPuNRxf1T121nj+st8i6+PDw43L+jIbqL4bRPflwXDA1yibX/WiN+aunt z3q6kCDfofnoM4vv0PlPcsYdVoqs0L3q7w3GwIz1e8tPm94sSrFMn/5zDWEya4c= =TOlz -----END PGP SIGNATURE-----

1 0

shape is (3,4) or (3L,4L) ?
by Mark Bakker 13 Sep '13

13 Sep '13

Hello List, I am teaching a Python programming class where students use their own computer. When I create an array with 3 rows and 4 columns, a = zeros((3,4)), and ask for the shape, shape(a), then some students get (3,4), while others get (3L,4L). I guess the 3L and 4L are long integers. I wonder why they don't all get the same thing. They all installed numpy through Canopy Express. Any thoughts? Is this a setting in printoptions maybe? Thanks, Mark

4 5

can argmax be used to return row and column indices?
by Mark Bakker 13 Sep '13

13 Sep '13

Hello list, I am trying to find the indices of the maximum value in a 2D array. argmax works fine, but returns the index in the flattened array. That is often not very convenient. Is there a function that returns the index of the row and column? Or can the index of the flattened array easily be converted to the row can column indices (I know, I can write a function for that, but I figure numpy already has one). Thanks, Mark

2 1

How to find number of bytes in a PyArrayObject without a segfault
by Alex Eftimiades 12 Sep '13

12 Sep '13

I ultimately want to access the total number of bytes in a PyArrayObject through a C++ program. I cannot seem to do it without sefaulting. All of the following yields a segfault: my_array->descr->subarray->shape my_array->dimensions PyArray_SHAPE(my_array) PyArray_DIMS(my_array) PyArray_ITEMSIZE(my_array) and of course... PyArray_NBYTES(my_array) I am working from within sage if that makes any difference. Is there a way to iterate over an array from within C++ without knowing the number of bytes in the array (and using strides--which oddly enough do not result in a segfault), or iterating over each dimension? Please point me in the right direction. There must be a way or else the entire software suite would not work. Thanks, Alex Eftimiades

1 0

Re: [Numpy-discussion] NumPy-Discussion Digest, Vol 84, Issue 19
by Graeme B. Bell 11 Sep '13

11 Sep '13

Hi again Julian, > its not np.any that is slow in this case its np.array([A, B, C]) Not really. The 4 unique alternative implementations provided all work quickly and correctly as-is, without a need to jump through a strange idiomatic hoop, so the problem does indeed lie with the implementation approach used in the current numpy.any(). It is unreasonable to expect users to know and always prepare their data using the idiom you propose if they might want to use the .any() function. If quirky idioms are needed to fix the performance of .any(), they should be placed in the implementation of .any(). > this has the same performance as your code: > a = empty([3] list(A.shape) How do you know? This isn't valid Python, it won't run. Please can you send me a working version I can benchmark and compare? Graeme. On Sep 5, 2013, at 8:52 PM, <numpy-discussion-request(a)scipy.org> wrote: > > hi, > its not np.any that is slow in this case its np.array([A, B, C]) > > np.dstack([A, B, C]) is better but writing it like this has the same > performance as your code: > a = empty([3] list(A.shape) > a[0] = A>5; a[1] = B<2; a[2] = A>10; > np.any(a, 0) > > I'll check if creating an array from a sequence can be improved for this > case.

1 1