guile-user
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Re: Matrix or array operations library


From: Zelphir Kaltstahl
Subject: Re: Re: Matrix or array operations library
Date: Wed, 26 Dec 2018 12:38:36 +0100
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.2.1

On 25.12.18 18:00, address@hidden wrote:
>> Hello Guile Users,
>>
>> Is there some library, that enables high performance matrix operations
>> or even n-dimensional array operations? I am thinking of something like
>> NumPy in the Python ecosystem. I think NumPy in turn also uses some
>> lower level thing to do what it does. I think OpenBLAS or MKL, depending
>> on the architecture. I wonder if there is any wrapper around OpenBLAS
>> for Guile or something similar.
>>
>> I am writing a little code for matrix operations and currently I am
>> using Guile arrays, as they are made of vectors and have constant access
>> time, which is already great. My guess is, that this would be the right
>> choice if using pure Guile. I am writing data abstraction procedures, so
>> that later on I could exchange what is used to represent the data.
>>
>> Maybe, if there is something like NumPy or lower level, I should use
>> that instead? (Would I have to learn how to use FFI first?)
>>
>> Or maybe Guile's implementation is already so fast, that it would not
>> make that much difference to use a lower level thing?
>>
>> Currently I have only a little experimental program, started today, so
>> no huge plan. OK, one can fantasize about stuff like Pandas data frames
>> in Guile, but I have no illusion, that it is a work of a few days or
>> even weeks. It would be nice to learn, how to use a low level thing or
>> maybe even Pandas, if there are any such bindings for Guile. I could
>> make the implementation use different representations, depending on a
>> parameter or something like that.
>>
>> Regards,
>>
>> Zelphir
>>
> IMO, Guile probably needs something like ndarray, the base for numpy,
> if matrix operations are going to go anywhere.
>
> As far as using blas or others, I have developed a package for helping
> to use the FFI API for C libraries.  It may be of interest to you.
>
> ffi-helper: https://www.nongnu.org/nyacc/ffi-help.html
>
> Matt

Hi!

Thanks for the link Matt. Now I have a bunch of questions :-)

I started reading the guide you linked to. I've never done any FFI stuff
in any programming language so far, so this is completely new territory
for me.

(Q1) Can you explain what the advantage of the method you linked to is,
in comparison to doing something like
https://www.gnu.org/software/guile/manual/html_node/Dynamic-FFI.html ?

(Q2) What is that `guild` command in the guide? (sounds like guildhall
another Guile project, but I forgot what that does)

(Q3) How would I set everything up, so that I can follow the tutorial at
https://www.nongnu.org/nyacc/ffi-help.html?

iirc NYACC is a minimalistic compiler that was mentioned here on the
mailing list several times in the context of bootstrapping Guile.

(Q4) How is NYACC connected to doing FFI?

(Q5) Do I need NYACC installed on my system to do the stuff in the
tutorial you linked?

I guess, since I don't know things about FFI in general (except that I
somehow get to magically call functions of another programming
language), things like NumPy are out of my league for now and I should
start with simple learning examples, to get an understanding of how FFI
usually works and what things one touches, when using FFI for a library
in another language and how much understanding of how the library works
one needs to have.

With regard to NumPy, especially the stuff about the "strides" (whatever
that is) in the data buffer of a NumPy array and how what changes when
creating a view and changing indices (For example I cannot imagine a
single reason, why the ordering of how one writes the indices would
matter, except for human readability?) … If I need to keep track of that
… Let's just say I dislike working on a low level with a lot of
pointers, addresses and segfaults ; )

(Q6) Maybe I do not need to pay attention to how NumPy does things,
except for procedure names, since NumPy handles it and I am just calling
its procedures somehow? That's what I would be hoping for.

Maybe those are a lot of questions and maybe most of them would be
answered, if I ever did any FFI stuff before. If anyone knows a good
tutorial on FFI in Guile that explains this stuff, that would be great
too. Or maybe someone knows some easy-to-ffi library to try things with?

Best regards,

Zelphir



reply via email to

[Prev in Thread] Current Thread [Next in Thread]