Send Moses-support mailing list submissions to
moses-support@mit.edu
To subscribe or unsubscribe via the World Wide Web, visit
http://mailman.mit.edu/mailman/listinfo/moses-support
or, via email, send a message with subject or body 'help' to
moses-support-request@mit.edu
You can reach the person managing the list at
moses-support-owner@mit.edu
When replying, please edit your Subject line so it is more specific
than "Re: Contents of Moses-support digest..."
Today's Topics:
1. Re: running the decoder for the first time (mhmd hassnen)
2. Re: Using boost for prefix/suffix checks (Jeroen Vermeulen)
3. Re: Moses v 3.0 (Marcos Fernandez)
4. max phrase length (Jorg Tiedemann)
5. Missing destructor for KenOSMBase (Jeroen Vermeulen)
----------------------------------------------------------------------
Message: 1
Date: Fri, 6 Feb 2015 11:22:10 +0200
From: mhmd hassnen <mhmd_hasnen@yahoo.com>
Subject: Re: [Moses-support] running the decoder for the first time
To: mohamed hasanien <mhmd_hasnen@yahoo.com>, "moses-support@mit.edu"
<moses-support@mit.edu>
Message-ID: <1C4BA623-670A-47A6-999A-47619FD78F06@yahoo.com>
Content-Type: text/plain; charset="us-ascii"
> Hi moses support
>
> i try to run the moses decoder using sample provided via this link http://www.statmt.org/moses/download/sample-models.tgz
> using this command
> echo 'das ist ein kleines haus' | ~/mosesdecoder/bin/moses -f ~/ sample-models/phrase-model/moses.ini > out
>
> and i get this error message
> Defined parameters (per moses.ini or switch):
> config: /root/ sample-models/phrase-model/moses.ini
> line=WordPenalty
> FeatureFunction: WordPenalty0 start: 0 end: 0
> Exception: moses/ScoreComponentCollection.cpp:250 in void Moses::ScoreComponentCollection::Assign(const Moses::FeatureFunction*, const std::vector<float>&) threw util::Exception'.
> Feature function WordPenalty0 specified 1 dense scores or weights. Actually has 0
>
> mohammed hassanien Mohammed
> Egyption Programmers Vice-captain
> 01000121556
> Egyption Programmers Syndicate
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20150206/61262719/attachment-0001.htm
------------------------------
Message: 2
Date: Fri, 06 Feb 2015 11:39:29 +0100
From: Jeroen Vermeulen <jtv@precisiontranslationtools.com>
Subject: Re: [Moses-support] Using boost for prefix/suffix checks
To: moses-support@mit.edu
Message-ID: <54D499E1.908@precisiontranslationtools.com>
Content-Type: text/plain; charset="windows-1252"
On 2015-02-06 00:56, Jeroen Vermeulen wrote:
> On 2015-02-05 17:25, Hieu Hoang wrote:
>> great, thanks. committed
Aaaand I found another one. Wasn't grep'ing aggressively enough. See
attachment.
As far as I can tell, this would also fix a crash in
PhraseDictionaryCompact::Load() if the filename is short. The only
other change you might actually notice is a typo fix in the error
message: "exist" instead of "exit". The logic for the optional suffix
becomes a bit simpler but should still work as it did.
The crash that this should fix is *not* the vector::_M_range_check crash
that's been reported. That's actually what I was trying to figure out,
but it's got to be a std::vector::at() somewhere that's getting an
out-of-range index. If a boost upgrade fixes that one, it may be a bug
in boost. But it could also be because boost replaced an at() call with
array-style indexing, which doesn't check the index and may simply
address the wrong memory instead...
Jeroen
-------------- next part --------------
A non-text attachment was scrubbed...
Name: moses-phrasedictionarycompact-suffix.diff
Type: text/x-patch
Size: 1649 bytes
Desc: not available
Url : http://mailman.mit.edu/mailman/private/moses-support/attachments/20150206/d6bce7a9/attachment-0001.bin
------------------------------
Message: 3
Date: Fri, 6 Feb 2015 12:31:39 +0000 (UTC)
From: Marcos Fernandez <marcos.fernandez.lopez@usc.es>
Subject: Re: [Moses-support] Moses v 3.0
To: moses-support@mit.edu
Message-ID: <loom.20150206T132306-201@post.gmane.org>
Content-Type: text/plain; charset=us-ascii
Hi,
great news!
But, has it been any changes in Moses Server related to the way in which it
uses the translation caches?. Some months ago we discussed about it, and
Barry (I think it was him) stated that the new global cache mechanism
(instead of the previous per-thread caches), is useful for stand-alone moses
but it has a negative impact over Moses Server because new requests are
attended in new threads whose caches are empty when they are created.
Thanks,
Marcos
------------------------------
Message: 4
Date: Fri, 6 Feb 2015 13:43:53 +0100
From: Jorg Tiedemann <tiedeman@gmail.com>
Subject: [Moses-support] max phrase length
To: moses-support <moses-support@mit.edu>
Message-ID: <4D186428-575F-49C4-B6E3-617B8FFB5CCA@gmail.com>
Content-Type: text/plain; charset="iso-8859-1"
Hi,
I have a question about the -max-phrase-length flag. I have lattice input that collects a lot of memory and eventually crashes when memory runs out (over 90GB). Now I realized that setting max-phrase-length helps even when setting it to length 7 which is the maximum length in my extracted phrase table. I do't really see why this is the case but maybe I misunderstand the flag and what it does. Is it not related to the max length of translation options coming from the phrase table?
Thanks for clarifying this to me.
cheers,
J?rg
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20150206/dcdb406a/attachment-0001.htm
------------------------------
Message: 5
Date: Fri, 06 Feb 2015 16:04:26 +0100
From: Jeroen Vermeulen <jtv@precisiontranslationtools.com>
Subject: [Moses-support] Missing destructor for KenOSMBase
To: moses-support <moses-support@mit.edu>
Message-ID: <54D4D7FA.2050504@precisiontranslationtools.com>
Content-Type: text/plain; charset="utf-8"
There's a nasty warning in the build about a "delete" on a
pointer-to-KenOSMBase. KenOSMBase needs a virtual destructor.
KenOSMBase is the base class for instantiations of the KenOSM abstract
template class, so the actual type of the object that's being deleted is
not known statically. In theory the delete produces undefined
behaviour; in practice I expect it will simply fail to clean up the
model object.
Patch attached. The KenOSM destructor can be left implicit, so long as
virtual dispatch is in place to ensure that the right one is called.
Jeroen
-------------- next part --------------
A non-text attachment was scrubbed...
Name: moses-kenosmbase-destructor.diff
Type: text/x-patch
Size: 394 bytes
Desc: not available
Url : http://mailman.mit.edu/mailman/private/moses-support/attachments/20150206/fe2df581/attachment.bin
------------------------------
_______________________________________________
Moses-support mailing list
Moses-support@mit.edu
http://mailman.mit.edu/mailman/listinfo/moses-support
End of Moses-support Digest, Vol 100, Issue 23
**********************************************
Subscribe to:
Post Comments (Atom)
0 Response to "Moses-support Digest, Vol 100, Issue 23"
Post a Comment