Re: [Numpy-discussion] floor with dtype

2017-09-25 Thread Mark Bakker
> On 2017-09-25 10:59, Renato Fabbri wrote: > > > """ > > > In [3]: n.floor(n.linspace(0,5,7), dtype=n.int ) > > > > > --- > > > TypeError Traceback (most recent call > > last) >

Re: [Numpy-discussion] Proposal - change to OpenBLAS for Windows wheels

2017-09-25 Thread Ralf Gommers
On Tue, Sep 26, 2017 at 6:48 AM, Nathaniel Smith wrote: > Makes sense to me. > > On Sep 25, 2017 05:54, "Matthew Brett" wrote: > >> Hi, >> >> I suggest we switch from ATLAS to OpenBLAS for our Windows wheels: >> >> * OpenBLAS is much faster, at least

Re: [Numpy-discussion] Proposal - change to OpenBLAS for Windows wheels

2017-09-25 Thread Nathaniel Smith
Makes sense to me. On Sep 25, 2017 05:54, "Matthew Brett" wrote: > Hi, > > I suggest we switch from ATLAS to OpenBLAS for our Windows wheels: > > * OpenBLAS is much faster, at least when Tony Kelman tested it last year > [1]; > * We now have an automated Appveyor build

[Numpy-discussion] nditer and updateifcopy semantics - advice needed

2017-09-25 Thread Matti Picus
I filed issue 9714 trying to get some feedback on what to do with updateifcopy semantics and user-exposed nditer. For those who are unfamiliar with the issue see below for a short summary, issue 7054 for a lengthy discussion, or pull request 9639 (which is still not merged). As I mention in

Re: [Numpy-discussion] Proposal - change to OpenBLAS for Windows wheels

2017-09-25 Thread Olivier Grisel
+1 for the change. -- Olivier ​ ___ NumPy-Discussion mailing list NumPy-Discussion@python.org https://mail.python.org/mailman/listinfo/numpy-discussion

[Numpy-discussion] Proposal - change to OpenBLAS for Windows wheels

2017-09-25 Thread Matthew Brett
Hi, I suggest we switch from ATLAS to OpenBLAS for our Windows wheels: * OpenBLAS is much faster, at least when Tony Kelman tested it last year [1]; * We now have an automated Appveyor build for OpenBLAS [2, 3]; * Tests are passing with 32-bit and 64-bit wheels [4]; * The next Scipy release will

Re: [Numpy-discussion] floor with dtype

2017-09-25 Thread Renato Fabbri
I have an array of floats and want their floor values as integers. (e.g. to use them as indexes for a table lookup) It seems reasonable to assume this is a frequent use of floor. Anyway, you gave me a better vay to do it: >>> myints = n.floor(myarray).astype(n.int) On Mon, Sep 25, 2017 at 7:23

Re: [Numpy-discussion] floor with dtype

2017-09-25 Thread Thomas Jollans
On 2017-09-25 10:59, Renato Fabbri wrote: > """ > In [3]: n.floor(n.linspace(0,5,7), dtype=n.int ) > --- > TypeError                                 Traceback (most recent call last) > in () > > 1

[Numpy-discussion] floor with dtype

2017-09-25 Thread Renato Fabbri
""" In [3]: n.floor(n.linspace(0,5,7), dtype=n.int) --- TypeError Traceback (most recent call last) in () > 1 n.floor(n.linspace(0,5,7), dtype=n.int) TypeError: No loop matching the