Mailman 3 September 2013 - SciPy-Dev

matrix_rank scipy version ?
by josef.pktd＠gmail.com 03 Sep '13

03 Sep '13

random question, while catching up with some linalg (matrix decomposition and least squares with reduced rank) scipy has rank-revealing (pivoting) QR for a few versions now. Would it be worth it to add a scipy version of matrix_rank that uses QR, instead of SVD as in numpy (IIRC) What's the cutoff for a QR version? Josef

1 0

fortran docstrings
by josef.pktd＠gmail.com 03 Sep '13

03 Sep '13

playing with parsing fortran documentation in special cdflib https://gist.github.com/josef-pkt/6426618 some fortran files don't work example ################# cumpoi(s, xlam, cum, ccum) CUMulative POIsson distribution Returns the probability of S or fewer events in a Poisson distribution with mean XLAM. Parameters ---------- s : array_like Upper limit of cumulation of the Poisson. xlam : array_like Mean of the Poisson distribution. Returns ------- cum : array_like Cumulative poisson distribution. ccum : array_like Compliment of Cumulative poisson distribution. Notes ----- Uses formula 26.4.21 of Abramowitz and Stegun, Handbook of Mathematical Functions to reduce the cumulative Poisson to the cumulative chi-square distribution. ################# cumt(t, df, cum, ccum) CUMulative T-distribution Computes the integral from -infinity to T of the t-density. Parameters ---------- t : array_like Upper limit of integration of the t-density. df : array_like Degrees of freedom of the t-distribution. Returns ------- cum : array_like Cumulative t-distribution. ccum : array_like Compliment of Cumulative t-distribution. Notes ----- Formula 26.5.27 of Abramowitz and Stegun, Handbook of Mathematical Functions is used to reduce the t-distribution to an incomplete beta. ################# cumtnc(t, df, pnonc, cum, ccum) CUMulative Non-Central T-distribution Computes the integral from -infinity to T of the non-central t-density. Parameters ---------- t : array_like Upper limit of integration of the non-central t-density. df : array_like Degrees of freedom of the non-central t-distribution. pnonc : array_like Non-centrality parameter of the non-central t distibutio Returns ------- cum : array_like Cumulative t-distribution. ccum : array_like Compliment of Cumulative t-distribution. Notes ----- Upper tail of the cumulative noncentral t using formulae from page 532 of Johnson, Kotz, Balakrishnan, Coninuous Univariate Distributions, Vol 2, 2nd Edition. Wiley (1995) This implementation starts the calculation at i = lambda, which is near the largest Di. It then sums forward and backward.

1 0

Re: [SciPy-Dev] [Numpy-discussion] ANN: Scipy 0.13.0 beta 1 release
by David Cournapeau 03 Sep '13

03 Sep '13

On Mon, Sep 2, 2013 at 5:46 PM, Pauli Virtanen <pav(a)iki.fi> wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hi, > > 02.09.2013 18:30, David Cournapeau kirjoitti: > [clip] > > Ok, I think the commit 5935030f8cced33e433804a21bdb15572d1d38e8 is > > quite wrong. > > > > It conflates the issue of dealing with Accelerate brokenness and > > using g77 ABI. I would suggest reverting it for 0.13.0 (and > > re-disable single precision), as fixing this correctly may require > > quite some time/testing. > > I'm -1 on returning to the previous situation where many routines on > OSX are simply broken (which was the situation previously --- > "disabling single precision" left several things still broken). I'd > rather just postpone the 0.13.0 release until this issue is solved > properly. > I see, I missed that it was more than just reverting to slower versions. > > Can you say what exactly is wrong -- as far as I know, Accelerate on > OSX simply uses g77 ABI. There were some bugs previously where it in > places did things differently, but on recent OSX releases this is no > longer the case? > > I guess you are trying to link with MKL? In that case, I would rather > propose restoring the previous MKL wrappers (with trivial extensions > for the missing w* symbols). > The pb is specific to MKL, yes, but I've found a workaround: in the case of MKL, you just need to workaround the g77 ABI of the MKL vs gfortran ABI, but the LAPACK interface is the actual LAPACK, not CLAPACK. So for MKL, you need non dummy wrappers iff the function returns a complex. A dirty patch seems to confirm that this fixes the issue, I will prepare an actual patch tomorrow, David > > - -- > Pauli Virtanen > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.12 (GNU/Linux) > > iEYEARECAAYFAlIkwPQACgkQ6BQxb7O0pWBnkwCfeKj32CYYCPdEWVcYYMtq/OZM > 32wAoIn9yG/HNtUeh+XwqAm2uAS9sVQ5 > =CKBN > -----END PGP SIGNATURE----- > > _______________________________________________ > NumPy-Discussion mailing list > NumPy-Discussion(a)scipy.org > http://mail.scipy.org/mailman/listinfo/numpy-discussion >

2 2

ANN: Scipy 0.13.0 beta 1 release
by Ralf Gommers 02 Sep '13

02 Sep '13

Hi all, I'm happy to announce the availability of the first beta release of Scipy 0.13.0. Please try this beta and report any issues on the scipy-dev mailing list. Source tarballs and release notes can be found at https://sourceforge.net/projects/scipy/files/scipy/0.13.0b1/. Windows and OS X installers will follow later (we have a minor infrastructure issue to solve, and I'm at EuroScipy now). Cheers, Ralf

6 14

ANN: Numpy 1.8.0 beta 1 release
by Charles R Harris 01 Sep '13

01 Sep '13

Hi all, I'm happy to announce the first beta release of Numpy 1.8.0. Please try this beta and report any issues on the numpy-dev mailing list. Source tarballs and release notes can be found at https://sourceforge.net/projects/numpy/files/NumPy/1.8.0b1/. The Windows and OS X installers will follow when the infrastructure issues are dealt with. Chuck

1 0