Re: JIT compilation with callbacks into Scheme

[Amirouche Boubekki]

On 2016-10-21 22:06, Jan Wedekind wrote:
> On Fri, 21 Oct 2016, Amirouche Boubekki wrote:
>
> > On 2016-10-21 14:13, Jan Wedekind wrote:
> > > On Fri, 21 Oct 2016, Amirouche Boubekki wrote:
> > >
> > > > Héllo Jan!
> > > >
> > > > On 2016-10-20 22:14, Jan Wedekind wrote:
> > > > > Hi,
> > > > >   I managed to implement JIT compilation of method calls into the
> > > > > Guile interpreter (AIscm 0.8.2 [1]):
> >
> > [...]
> >
> > > > > [1] http://wedesoft.github.io/aiscm/
> > > >
> > > > How does this compare with numpy?
> > > >
> > > > TIA!
> > > >
> > > > -- Amirouche ~ amz3 ~ http://www.hyperdev.fr
> > >
> > >   First of all numpy supports native floating point operations and
> > > this work does not yet do that ;)
> > >   In terms of performance I previously posted some benchmarks
> > > comparing with C programs [1].
> > >   Numpy is statically compiled and uses generic functions and 
> > > function
> > > pointers to implement binary operations for combinations of types 
> > > [2].
> > >
> > >     np.array([1,2,3],dtype=np.uint8) + 
> > > np.array([1,2,3],dtype=np.int16)
> > >     # array([2, 4, 6], dtype=int16)
> > >
> > >   When instead using a JIT (e.g. Python Theano) the resulting 
> > > software
> > > is more generic and composable in terms of operations (tensors,
> > > composite operations) and in terms of datatypes (RGB, complex values,
> > > arrays, hypercomplex values, ...).
> > >   Here's a technical report [3] with more details and motivation (in
> > > that case libJIT and Ruby was used).
> > >   Please let me know if you have some interesting applications in 
> > > mind :)
> >
> > It's a long shot, but I plan to port spacy and gensim and I read they
> > use numpy that's why I ask.
> >
> > > Regards
> > > Jan
> > >
> > > [1] 
> > > https://lists.gnu.org/archive/html/guile-user/2016-06/msg00019.html
> > > [2]
> > > https://github.com/numpy/numpy/blob/master/numpy/core/src/multiarray/calculation.c#L534
> > > [3] http://www.wedesoft.de/downloads/tensors_libjit.pdf
> >
> > -- Amirouche ~ amz3 ~ http://www.hyperdev.fr
>
> Not sure how they use NumPy. However you might be interested in the
> publication "Natural Language Processing (almost) from Scratch" which
> uses machine learning [4].
>
> [4] https://arxiv.org/abs/1103.0398

Thanks for that article it seems a compeling approach to do basic NLP 
tasks.
Sadly my ML-fu is too low to process the content of this article.

Anyway, thanks for the explanations.

--
Amirouche ~ amz3 ~ http://www.hyperdev.fr