in Moses would involve more effort to
implement for phrase-based than for syntax-based.
Phil
On 11 Mar 2015, at 20:03, Chris Dyer cd...@cs.cmu.edu wrote:
Hi all, can anyone tell me if moses can be configured to generate
unique k-best lists using the Huang et al. (2006) dynamic
programming
Hi all, can anyone tell me if moses can be configured to generate
unique k-best lists using the Huang et al. (2006) dynamic
programming algorithm (Section 5.2 of
http://www.cis.upenn.edu/~lhuang3/amta06-sdtedl.pdf)?
Thanks--
Chris
___
Moses-support
We've put together a small corpus (15k sentences training, 2k each of
dev/test) of Swahili-English which you can get here:
http://demo.clab.cs.cmu.edu/cdyer/gv.sw-en.tar.gz
It's roughly equivalent to the data used for the experiments reported
in this paper:
It's useful to have epsilons since it simplifies the creation of
lattices in some cases. Yes, you can convert them to a deterministic
equivalent, but that involves implementing FSA determinatization (or
using a tool like https://pypi.python.org/pypi/pyfst), which may not
be convenient.
Btw, I've
http://i0.kym-cdn.com/photos/images/original/000/002/299/ze_house_.jpg
On Wed, Sep 18, 2013 at 1:15 PM, Lane Schwartz dowob...@gmail.com wrote:
Thanks! What happens if the reference isn't reachable?
On Wed, Sep 18, 2013 at 4:50 AM, Hieu Hoang hieu.ho...@ed.ac.uk wrote:
For anyone who's
Yes, there definitely should be a few checks in various places. I've
got a list of recommendations to make lattice decoding a bit easier to
get started with. We'll discuss this next week.
-C
On Fri, Sep 6, 2013 at 5:47 PM, Hieu Hoang hieuho...@gmail.com wrote:
Good to know. I don't think it's
demo,
The id of node of 1 and 2 should be exchanged. But anyway, it doesn't
matter right?
I will just try feeding the input to moses without checking the format.
Many thanks
Best,
Wei
On Thu, Jun 6, 2013 at 5:19 PM, Chris Dyer cd...@cs.cmu.edu wrote:
I think you've converted the lattice
The JHU summer workshop final report had some experiments on this:
http://www.learningace.com/doc/3098660/be148017730f3f3a7b45d656276b482a/jhu-summer-workshop-final-report
(See Fig. 6.7 and surrounding)
In general:
1) MERT works on so few features that you don't need much dev data to learn them
Giza? Who needs Giza? Consider this:
http://www.ark.cs.cmu.edu/cdyer/fast_valign.pdf
On Mon, Apr 15, 2013 at 7:58 AM, Hieu Hoang hieuho...@gmail.com wrote:
You're unlikely to get a response to difficult questions on giza++/mgiza
these days. IMO, the knowledge about them is being slowly lost as
If you're interested in statistical significant testing, you really
ought to read the Clark et al. (2011) paper
(http://www.cs.cmu.edu/~jhclark/pubs/significance.pdf). We showed that
the Koehn technique and related methods can indicate significance for
reasons that have little to do with the
cdec has one in https://github.com/redpony/cdec/tree/master/mteval
It depends only on the utils subfolder, so it's a bit easier to
build than the whole decoder. The bleu implementation is in ns.cc, and
there is also an implementation of TER in ns_ter.cc.
-Chris
On Thu, Sep 13, 2012 at 4:15 PM,
, I would set this to 0 if i2real i1 (or perhaps i2real
i1real ?). That may be all you need to do.
-Chris
On Mon, Aug 27, 2012 at 5:38 PM, Dario Ernst mo...@kanojo.de wrote:
Hi Chris,
On 08/24/2012 04:50 PM, Chris Dyer wrote:
Although forcing monotone alignments sounds like a fairly large
(in this case, a
reverse alignment), the posterior distribution will also be 0. For
this reason, if you initialize carefully, you'll be fine.
On 08/24/2012 04:10 AM, Chris Dyer wrote:
I think adding this would be have tremendous value.
Yay! I'm not alone ;P. Especially in conjunction with PISA
It should be possible to adapt Giza's HMM implementation to produce
monotone alignments. These are the changes that would be necessary
(and which should be fairly easy, if you can figure out the code):
1) alignment distribution initialization. by default Giza initializes
the HMM transition
Hi Marcin,
mkcls is best understood as implementing the Brown et al (1992)
clustering model (i.e., a bigram HMM with some extra hard
constraints), although it uses a different algorithm for parameter
learning than the algorithm proposed by Brown.
Its performance has been analyzed and compared to
Hi all, is there a particular citation that should be used for the
moses MERT implementation? I seem to recall that there might have been
a paper when the old c-mert got reimplemented, but I can't seem to
find it.
-C
___
Moses-support mailing list
},
url = {prague-mert.pdf}
}
On Sat, Apr 23, 2011 at 4:12 PM, Lane Schwartz dowob...@gmail.com wrote:
Check the Prague Bulletin. I thought that's where they published the
description.
On Sat, Apr 23, 2011 at 2:35 PM, Chris Dyer cd...@cs.cmu.edu wrote:
Hi all, is there a particular citation
cdec (https://github.com/redpony/cdec) includes an implementation,
called vest. But someone needs to write code that will cause moses to
export its search lattices in the right format (which is a funny
crappy json-based encoding).
On Fri, Mar 25, 2011 at 2:59 PM, Lane Schwartz dowob...@gmail.com
On Mon, Mar 21, 2011 at 3:19 AM, Alex Fraser alexfra...@gmail.com wrote:
2) there seems to be some evidence that some translations in the
phrase table are so bad that having leaving some words untranslated
is better than using what's in the phrase table. I can see an
argument that says that
I allow pass through of all words, with a penalty that is also learned
by MERT.
Interesting stuff. Do you have results published on this?
This was easiest to implement when I wrote cdec, and the results
seemed good enough, so I never did a proper comparison. I will
describe the newer innovation
I've started using an OOV feature (fires for each LM-OOV) together
with an open-vocabulary LM, and found that this improves the BLEU
score. Typically, the weight learned on the OOV feature (by MERT) is
quite a bit more negative than the default amount estimated during LM
training, but it is still
Hi Sylvain,
I've gone ahead and added the relevant function to WordLattice.h/cpp
that should make it a bit easier to construct lattices
programmatically. You'll need to encode them in the data type defined
in PCNTools.h, which is basically a programmatic representation of the
PLF format described
You can escape it with a backslash:
((('\'',1,1),),)
On Sat, Dec 18, 2010 at 7:47 AM, Mehmet Tatlıcıoğlu
tatlicio...@gmail.com wrote:
Hi,
How can I put ' character as a label on an edge in word lattices?
eg. if the label is test, then the lattice component is the form
of ((('test', 1.0, 1),
Would it be possible to have some kind of flag that turns this on or
off? For a variety of reasons I've been working with the same software
in a bunch of different environments that are similar (but just
different enough) that I found it useful to make the parts that deal
with the cluster sort of
I had a query with regard to use of lattice input in moses.
There is a little difference in the translations generated when I run moses
using the 'normal' input format and when I run it with 'lattice input'
format.
The translations weren't radically different - only a few phrases were
I have some software that will generate splits from German language compounds:
https://github.com/redpony/cdec/tree/master/compound-split/
It can produce either lattices of high probability splits or just a
1-best split. The model used is a conditional random field trained on
a small amount
--factorDelimiter=|
There is such a flag. I implemented this about 4 years ago, but AFAIK
I'm the only one who ever uses (and I still use it).
-C
etc.
Miles
On 15 November 2010 21:30, Hieu Hoang hieuho...@gmail.com wrote:
That's a good idea. In the decoder, there's 4 places that has to
Hi Mehmet,
The following lattice will do what you are asking for, I think:
((('x',0.5,1),('xy',0.5,2)),(('yz',1,2),),(('z',1,1),),)
The trick is ot use the last element of the tuples to indicate what
node the edge ends up in. The first two nodes, have single edges
leaving, but the edges don't
I don't know if moses's chart decoder supports lattices, but two other
chart decoders, Joshua and cdec, do.
On Tue, Aug 24, 2010 at 8:27 PM, Hwidong Na le...@postech.ac.kr wrote:
Hi all,
I want to decode an input lattice with moses_chart. When I switch the
decoder from moses to moses_chart,
Hi moses users,
This message is only of interest if you use Moses's word lattice
translation features. Since Moses is hardly graceful when it
encounters malformed lattice input, I've added a simple binary
(moses-cmd/src/checkplf) that you can use to verify that your lattice
inputs are both
Yes, that will do it.
On Mon, Aug 9, 2010 at 4:13 PM, Sylvain Raybaud
sylvain.rayb...@loria.fr wrote:
On Monday 09 August 2010 20:43:53 Chris Dyer wrote:
I'm sorry I haven't had an opportunity to look into this yet
(hopefully later this evening). But, one thing that you need to do is
make
Moses interprets the string *EPS* as an epsilon transition in the
lattice, which means it can take the transition (and use any
associated features), but the translation model will ignore the
transition.
-Chris
On Fri, Aug 6, 2010 at 11:02 AM, Sylvain Raybaud
sylvain.rayb...@loria.fr wrote:
There
If you do put together a script to convert to PLF (using the Moses
documentation), this would be a valuable contribution to the moses
code base. I'll be happy to answer questions about the lattice format
as they come up, although I'm starting a new job this week and may be
delayed in responding
That doesn't happen too often...Can you send along your moses.ini file?
On Sun, May 16, 2010 at 7:17 PM, David Edelstein dedelst...@ucdavis.edu wrote:
Hello,
I have trained, tuned, and prepared an eval set, trying to get Moses
to decode Arabic to English. However, trying with either filtered
Jie's assessment is correct, your moses.ini is missing a value under
the [weight-t] section. There should be 5 values there, but instead
there are 4. I'm not familiar with the MERT implementation that is
bundled with moses these days, so I can't really tell you where to
look, but it should
Moses uses features to discriminate between alternative translations
of individual sentences, so if the value is constant for all possible
translations (for example, because it is a function of the input), the
model won't be able to take advantage of it. It sounds like you might
be proposing
any
suggestions you have.
Suzy
On 26/03/2010, at 11:32 AM, Chris Dyer wrote:
Moses uses features to discriminate between alternative translations
of individual sentences, so if the value is constant for all possible
translations (for example, because it is a function of the input
Oh right, I had completely forgotten about this. With non-lattice
input, there is some logic that looks for phrases only up to a certain
max phrase size. However, this does not work with lattices and must
be disabled. I usually set the max phrase size to be 10 or
something like that, which
produce who is the bill which is
not necessarily an option ...
Thanks a lot for clarifying this to me!
Jörg
Chris Dyer wrote:
As long as you're just splitting, keeping the weights consistent
isn't
too hard- just keep all the weight in one segment
or directory.
in
/usr/lib/gcc/x86_64-linux-gnu/4.1.2/../../../../include/c++/4.1.2/ext/new_allocator.h
Indeed, the header file does not exist on my system.
Do I need to install some additional packages and re-compile Moses in a
certain way to get rid of this error?
Jörg
Chris Dyer
on the sentence that gave you the problems to see if you
can reproduce the problem.
regards
Barry
On Thursday 04 March 2010 15:24, Chris Dyer wrote:
I'm not certain what's causing this. From the part of the stack trace
you're showing, it looks like it's probably when translations options
to each part of the split ?
Maybe it has not any real impact in the end, or has it ?
Loïc
2010/3/1 Chris Dyer redp...@umd.edu
I guess word-graph doesn't split phrases either (I was just guessing).
It appears to be in SLF format, which is used by a number of tools
(like HTK and the SRI tools
For French-English translation, hierarchical models will probably not
do any better than phrase-based models. The rule-of-thumb seems to be
that language which large-scale reordering fare better with
hierarchical models, but language pairs with only relatively local
reordering are better with
I think the issue here has to do with the MERT configuration. When
you use lattice input, the weights on the lattice edges are included
in the translation model as a feature that MERT can optimize. I'm not
familiar with the MERT optimizers that are included with Moses, but it
sounds like it is
Stack smashing- that's a new one for Giza! Are you using the version
of giza++ from google code (http://code.google.com/p/giza-pp/)? Some
older versions had a few uninitialized variables that could
conceivably cause crashes on some architectures.
2010/1/21 Guillem Massó Sanabre
Hi John-
The first label is the orientation of the phrase pair with respect to
its left context (on the source side), and the second is the
orientation with respect to its right context. That's why you have to
have swap other or other swap, since a phrase can only be inverted
on one side.
Hope
This functionality is broken in the tip of the trunk. There was a
project last january to check change the way hypothesis scoring was
done to be more flexible that broke this. It needs to be fixed. One
alternative is to roll back to the version of the code that was at the
tip of the trunk in
results on the benefits of
modeling reordering.
http://aclweb.org/anthology-new/J/J03/J03-1002.pdf
On Sat, Oct 31, 2009 at 7:56 PM, Chris Dyer redp...@umd.edu wrote:
Modeling reordering is usually helpful, even during alignment. This
is especially true for lexical translation models (where words
Is it possible that you have a sentence of length zero? size2 is, I
believe, one of the dimensions of the trellis, which in one direction
is the source sentence length and in the other is the target sentence
length.
On Mon, Sep 14, 2009 at 1:28 AM, John Kolen johnfko...@gmail.com wrote:
I'm
This was recently announced on the corpora list:
http://www.uncorpora.org/
-Chris
On Sun, Sep 6, 2009 at 1:36 PM, Catalin Braescucata...@braescu.com wrote:
Thanks, Miles! From your link I got http://www.statmt.org/europarl/
Any other such goodies?
Catalin
--
Omlulu.com
On Sun, Sep
This is probably a problem with the regression test. The two
conditions ought to be identical, as you expect. However, keep in
mind that the distortion model is incredibly weak, and the heuristic
distance definition used in lattice decoding is also just an
approximation, so an off-by-one error
The alignment models are going to struggle quite a bit when the source
to target length ratio is so skewed. I would recommend finding a way
to retokenize/resegment the source and/or target language so as to
induce a more even ratio. If this isn't possible, you may need to
look into custom
Hi Thang,
The ranges that are specified for moses are just suggestions for random
starting points for the MERT algorithm. However, it may (and often does)
find weights that end up outside of these ranges.
-Chris
On Thu, Jun 4, 2009 at 2:32 AM, Thang Luong Minh luong.m.th...@gmail.comwrote:
These warnings aren't too serious. The alignments should be fine.
Chris
On Thu, May 14, 2009 at 8:56 PM, Tom Hoar tah...@gmail.com wrote:
During the train-factored-phrase-model.perl script of my
character-by-character aligment, giza-- reported the errors below.
Will this trained model be
There's actually an option for this in
train-factored-phrase-model.perl. Just specify --hmm and it will
automatically set the Giza++ options appropriately.
-Chris
2009/4/18 Manoj C (మనోజ్ చిన్నకోట్ల) manoj.chinnako...@gmail.com:
Hi All,
I am training a standard translation model using moses
You need to add a -weight-i flag to the command line which specifies
how much weighting to apply to the arc feature.
e.g.:
moses ... -weight-i 0.5
-Chris
On Thu, Apr 16, 2009 at 9:58 AM, Nguyen Manh Hung
manhh...@cl.ics.tut.ac.jp wrote:
Hi,
I'm using Moese to decode with lattices format as
Can you send me a stack trace for where the SEGV is happening? Once
the phrase table has been binarized, there's no need to have any
special temporary space.
On Tue, Apr 28, 2009 at 10:46 AM, Nguyen Manh Hung
manhh...@cl.ics.tut.ac.jp wrote:
Chris Dyer さんは書きました:
You need to add a -weight-i
2009-04-16 (木) の 11:34 -0400 に Chris Dyer さんは書きました:
Can you send me a stack trace for where the SEGV is happening? Once
the phrase table has been binarized, there's no need to have any
special temporary space.
On Tue, Apr 28, 2009 at 10:46 AM, Nguyen Manh Hung
manhh...@cl.ics.tut.ac.jp
Hi all-
There's a new release of GIZA++ available from
http://code.google.com/p/giza-pp/ . The changes address build issues
on a variety of platforms and compilers, including:
- better adherence to c++ header naming conventions (fixes build
problems on gcc 3.4)
- autodetection of MacOSX
Yeah, sorry about this- I broke moses, at least for certain compilers.
I'll fix it shortly.
-Chris
On Wed, Mar 4, 2009 at 12:17 PM, Miles Osborne mi...@inf.ed.ac.uk wrote:
ok, it seems that the most recent version of Moses had a bad commit
and broke the language model interface. so, this is
Another architecture to consider is storing/distributing the ttable
from a single central repository. Most of the ttable is full of crap,
and for each sentence, you know exactly what parameters will be
required in advance of running your E step. However, by not
distributing stuff that you don't
One way to do it is to just set a really high number for the
threshold. The maximum ceiling used by moses for a feature value is
100, and then pick the largest total sum that your feature weights can
have, double it (since you may have negative feature values), and set
that... There may be an
Hi Ondrej,
See below.
And one additional question: when extracting phrases, phrase-extract actually
extracts all phrases that *are not incompatible* with the alignment. I'm
thinking about a different method: just phrases that *are 'strictly'
compatible*, which means I would extract:
a=A
Hi,
There's a known bug with certain versions of g++ that GIZA++ hits. If
feasible, you might switch to a different version (4.0 works for
sure). If not, send your machine architecture/OS type and perhaps
someone will have a binary they can provide.
Thanks,
Chris
On Fri, Oct 17, 2008 at 1:43
normalizeTable
I use a more straightforward method (below) which yields in different
output, could someone elaborate?
Mycode:
Probability(f|e) = Count(e,f) / count (e)
The code that normalizes the counts to probabilities is in
TTable::normalizeTable. But, for the initialization, it
There's been a recent release of GIZA (July 8) that fixes some
potential sources of non-determinism, specifically relating to how
distortion models (model 2 or the HMM) get initialized.
When did you download it from http://code.google.com/p/giza-pp/ ?
--Chris
On Wed, Jul 16, 2008 at 6:35 PM,
Some versions of the g++ compiler (3.4.x I think) have a bug that
prevents this code from compiling properly. Working around the bug is
not straightforward, so trying a different version of the compiler is
probably your best bet.
Chris
On Mon, Jul 7, 2008 at 10:00 PM, Vineet Kashyap
[EMAIL
Hi Hu-
What tool are you using to generate the confusion networks? If you
use SRILM's lattice-tool, you should make sure to set the LM weight
and AM weights to something appropriate.
Chris
On Sun, May 11, 2008 at 11:33 PM, Hu Xiaoguang
[EMAIL PROTECTED] wrote:
hi, all
I did some experiments
:02 AM, John D. Burger [EMAIL PROTECTED] wrote:
Chris Dyer wrote:
I haven't looked into what's causing the particular problem on this
corpus, but another known problem with the GIZA HMM model is that it
doesn't do a fairly standard kind of normalization in the
forward-backward training
There have been some advocates of preserving case information as you
describe, although I've only seen them discussed in the context of
small-coverage systems, such as in the IWSLT task. See, for example,
the system description of the Carnegie Mellon Univ system from 2006's
IWSLT entry:
Faced with improper input, would it not make more sense to try and fix
it in the source language before translation, rather than distorting
the translation with the induced errors, then trying to fix the
translation ?
That could be an interesting experiment. The transformation one
I haven't looked into what's causing the particular problem on this
corpus, but another known problem with the GIZA HMM model is that it
doesn't do a fairly standard kind of normalization in the
forward-backward training, which causes underflow errors in some
sentences (especially quite long
/column: nan
avg. cols/sent:nan
Let me know if I made mistake somewhere.
Thanks
Linh
Chris Dyer wrote:
I am still confused about the lattice format,
In your examples:
1 ((('A',1.0,1),),(('B',1.0,1),),)
2 ((('A',1.0,1),('Z',1.0,2),),(('B',1.0,1),),(('C',1.0,1
The lattice format isn't documented yet on the webpage, but you can
see some examples of it in the lattice-distortion test directory Hieu
mentions. It should be fairly straightforward to decipher. Since
this format encodes a single lattice/CN per line of text, it can be
used easily with MER
Hi Qin-
Thanks for letting me know about this problem. I'll submit your
recommended fix. I'm not completely familiar with the GIZA
implementation of the HMM model, but this seems reasonable enough.
Chris
On Fri, Feb 15, 2008 at 4:40 PM, Qin Gao [EMAIL PROTECTED] wrote:
Hi All,
I found a
What phase of GIZA training is this occuring in? GIZA runs several
iterations in several stages, Model 1, HMM, Model 3, Model 4, etc.
And what is the very first sign of trouble you see? Are there any
errors before this?
This problem generally means that you're trying to model something
that
76 matches
Mail list logo