Mailman 3 January 2021 - NumPy-Discussion

UPD: Rivest-Floyd selection algorithm
by Виктория Малясова Jan. 16, 2021

Jan. 16, 2021

1 0

Subclassing vs. dispatch
by Israel, Daniel M Jan. 15, 2021

Jan. 15, 2021

I hope this is the right place to post this. The numpy documentation talks about two methods for making ndarray-like objects, subclassing and dispatching, but it is not clear to me which one is most appropriate for which purpose. Can someone provide, or point me to, some guidance, about this? I’m particularly interested in what happens if there are multiple layers of subclassing. Can you subclass from a subclass? Dispatch from a dispatch? Subclass from a dispatch and vice versa? My specific application is a pair of classes, SpectralArray and PhysicalArray that uses numpy.fft to provides a to_physical() and to_spectral() method, respectively, to simplify writing pseudo-spectral codes. Initially this will be serial, but the implementation will eventually use a mechanism similar to mpi4py-fft to allow the arrays to be distributed. Further, it would be nice to be able to make the code interoperable with the cupy CUDA numpy implementation, so that the sub array on each MPI process could use GPU accelerated FFTs. Advice? Thanks. — Daniel M. Israel, Ph. D. XCP-4: Methods & Algorithms Mailstop F644 Los Alamos National Laboratory 505 665 5664<tel:505%20665%205664>

2 1

NumPy EOL Versions?
by Symphoni Bush - NOAA Affiliate Jan. 13, 2021

Jan. 13, 2021

I am trying to find out if there are any end-of-life versions for NumPy, and if so, when do these versions typically become EOL/unsupported? Thanks.

4 5

NumPy Development Meeting Wednesday - Triage Focus
by Sebastian Berg Jan. 13, 2021

Jan. 13, 2021

Hi all, Our bi-weekly triage-focused NumPy development meeting is tomorrow (Wednesday, Jan 13th) at 11 am Pacific Time (19:00 UTC). Everyone is invited to join in and edit the work-in-progress meeting topics and notes: https://hackmd.io/68i_JvOYQfy9ERiHgXMPvg I encourage everyone to notify us of issues or PRs that you feel should be prioritized, discussed, or reviewed. Best regards Sebastian

1 0

Re: [Numpy-discussion] numpy.linalg interference with matplot.lib
by Frederix96 Jan. 12, 2021

Jan. 12, 2021

Dear Kevin, Thanks for the suggestion. I do not have those Windows Versions. I have Microsoft Windows 10 Pro with Version 10.0.19041 Build 19041. Anyway I still tried upgrading to 1.19.5 with no changes. Thanks a lot! Best regards, Frederix96

1 0

using scalar input on np.PyArray_MultiIterNew2
by zoj613 Jan. 11, 2021

Jan. 11, 2021

Hi all, I am looking for a way to use `np.PyArray_MultiIterNew2` in Cython to broadcast parameters of a function. The requirement is that the two arguments can be scalar and/or sequences. Using the usual `np.broadcast` function works well but is slow when iterating over the broadcasted input in a tight loop. I want to achieve the same using the C API. Currently, if I used `(<double*>np.PyArray_MultiIter_DATA(bcast, i))[0]` to iterate over the input when one of them is a scalar, I get no errors, but I notice the output of the parent function returns an array of zeros, which implies this approach didn't work. After investigating, it seems that np.PyArray_MultiIter_DATA only accepts numpy arrays. I could write a function to handle all combinations of scalar/array/list/tuple, and create temporary arrays to store the input data, but that seems daunting and error prone. Is there a way I can achieve this and have scalar arguments passed to np.PyArray_MultiIter_DATA be converted to same-element arrays without writing my own code from scratch? Any suggestions are welcome. -- Sent from: http://numpy-discussion.10968.n7.nabble.com/

2 5

How to speed up array generating
by klark--kent＠yandex.ru Jan. 11, 2021

Jan. 11, 2021

6 8

numpy.linalg interference with matplot.lib
by Frederix96 Jan. 11, 2021

Jan. 11, 2021

Dear numpy community, I am having an issue with numpy.linalg where a matrix inversion of a complex matrix using numpy.linalg.inv results in containing nan values when a plot of any kind was performed beforehand by matplotlib. For the specific details on the problem and what I have tried already see the stackoverflow post I started a month ago: https://stackoverflow.com/questions/65183984/why-does-plotting-something-wi… I know that this problem has to do with some specifics of the maschine I am using, as it does not happen on other maschines I tried. However it still appears after rebuilding my entire python enviroment. Hence I ran out of options appart from setting up my whole PC from scratch again, which I would rather not do. Is some problem like this known in the community? Has anyone an idea at which other specifications of my system it might be worth taking a look at? Thanks a lot in advance! Best regards, Frederix96

2 1

Welcome Chunlin to the NumPy development team.
by Charles R Harris Jan. 9, 2021

Jan. 9, 2021

Hi All, Welcome Chunlin Fang, aka Qiyu8, to the NumPy development team. He has made substantial contributions to the SIMD infrastructure, both code and reviews, and we look forward to his continuing work. Chuck

3 2

Sayed Adel has joined the developer team.
by Charles R Harris Jan. 9, 2021

Jan. 9, 2021

Hi All, Sayed Adel is now part of the developer team. His SIMD work has been substantial and ongoing, may it long continue. Welcome Sayed. Chuck

4 3