*Hi all,*
*when I tune parameters in the hierarchical decoding,the following errors:*
*..*
*Finished loading phrase tables : [0.475] seconds*
*Created input-output object : [0.475] seconds*
*Start loading binary SCFG phrase table. : [0.483] seconds*
*Translating: china joined up to the intern
A pedantic point: they can be arranged any way you like but must be
accessed in multiples of 32 bits. The KenLM trie data structure does
not follow this alignment. However, NICT added code to KenLM for ARM
that accesses the data in an aligned way by rounding up/down and doing
multiple accesse
i think what you're doing is called 'sentence-level bleu'. google it to
see what other people are doing.
it's an approximation of the document bleu, but the exact number won't
be the same
On 22/11/2012 14:17, Cuong Hoang wrote:
Hi all,
I have a very naive question, as I guess so.
I'm curren
it was changed last week so that word alignment is always on during
training and decoder. For phrase-based and hiero/syntax models.
There's some documentation here:
http://www.statmt.org/moses/?n=Moses.AdvancedFeatures#ntoc3
On 27/11/2012 06:22, saeed farzi wrote:
Dear all,
I trained phra
ah sorry, i've found the rest of the code and checked it in.
https://github.com/moses-smt/mosesdecoder/tree/android-ui
Looks like a lot of people are interested in this so if you continue to
develop it openly on github, other people may be able to help you.
Some things i found when doing it:
Yes. In combination both methods can achieve a size reduction of two
orders of magnitude compared to the standard binary unfiltered phrase
table.
W dniu 27.11.2012 11:39, Germán Sanchis Trilles pisze:
There are also some works addressing phrase-table size reduction. With
very limited loss in t
There are also some works addressing phrase-table size reduction. With
very limited loss in translation quality (if any), it has been shown that
the phrase-table can be pruned by up to 97%. See for instance:
http://www.mt-archive.info/EAMT-2011-Sanchis-Trilles.pdf
http://aclweb.org/anthology-ne
Hi,
First thing you'll want is a moses executable compiled for ARM. There
are several guides on cross compiling; you'll want one based on gcc,
clang, or at least something that supports exceptions. My favorite
method was just building an entire gentoo system in a chroot with qemu,
bu
Phrase table shrinking is done :) See Compact Phrase Table in the
advanced section.
W dniu 27.11.2012 10:54, Miles Osborne pisze:
For a long time now I've wanted to see Moses on a small device. Apart
from all of the extra functionality that isn't needed, one would also
need to work on shrinkin
For a long time now I've wanted to see Moses on a small device. Apart from
all of the extra functionality that isn't needed, one would also need to
work on shrinking the phrase table and perhaps also the search graph.
KenLM / RandLM already deal with making the language model smaller.
An interest
Porting Moses to Android would be interesting. I think that the Android
NDK is improving so it could be more feasible nowadays. There are
several speech translation apps available on the market (e.g. Jibbigo)
that demonstrate that MT is certainly viable on Android (and iOS)
devices without a cl
Hi,
in the latest update to Moses, "-alignment-info" has become obsolete
and it is on by default. Make sure all your tools are up to date
(binary phrase table converter, Moses), then it should work.
-phi
On Tue, Nov 27, 2012 at 7:22 AM, saeed farzi wrote:
>
> Dear all,
> I trained phrase table
12 matches
Mail list logo