to the last release 0.92 (from May this year) there are many
new features and a few smaller fixes.
You can get both packages here:
http://user.phil-fak.uni-duesseldorf.de/~tosch/downloads.html
Best regards,
Thomas Schoenemann
___
Moses-support mailin
Hi all,
I would greatly appreciate an explanation regarding this input type stuff. The
latest mert-moses.pl only explains input types 0-2, which are plain text,
confusion networks and lattices. Where is the difference between confusion
networks and lattices? What does type 3 mean? Why should th
Hi!
Thank you for the quick answer! I am right, then, that string->tree is
equivalent to simultaneous parsing and translation?
Best,
Thomas
Von: Lane Schwartz
An: Thomas Schoenemann
CC: "moses-support@mit.edu"
Gesendet: 20:17 Freitag,
Hi everyone,
there is quite a discussion about hierarchical and syntax-based approaches
here at the moment. For lack of time I could not follow it closely, so I hope
this question contributes to the discussion rather than being superfluous: in
terms of {tree/string} -> {tree/string}, how woul
Hi!
That's also the impression I got from a more thorough code review today. I will
work on implementing a positivity constraint, without changing the default
behaviour (in my own branch).
Best,
Thomas
Von: Barry Haddow
An: Thomas Schoenemann
Hi everyone,
can anyone tell me if the current MERT implementation should respect parameter
ranges? It does require to input them, but to my knowledge these ranges are
ignored.
Except for the word penalty I think all weights should be positive. Yet, I
often get negative ones from MERT, and
Stopping... : [472] seconds
Data loaded : [8] seconds
Stopping... : [496] seconds
Data loaded : [9] seconds
Stopping... : [495] seconds
Data loaded : [9] seconds
Stopping... : [648] seconds
Data loaded : [10] seconds
Stopping... : [485] seconds
____
Von: Thomas Scho
mulated by storing
weights for each list-entry. It would be a little more complicated to
implement, though.
Cheers,
Thomas
Von: Barry Haddow
An: moses-support@mit.edu; Thomas Schoenemann
Gesendet: 12:35 Mittwoch, 30.November 2011
Betreff: Re: [Moses-support
Hi everyone!
We all know that MERT gets slower in the later iterations. This is not
surprising as the n-best lists of all previous iterations are merged. I believe
this is quite important for translation performance.
Still, it seems important to me to get the merged lists as small as possible.
, but would also
be interested in other toolkits in case they behave more suitably.
Can anyone point me to anything?
Many thanks!
Thomas Schoenemann (currently University of Pisa)
___
Moses-support mailing list
Moses-support@mit.edu
http://mailman.mi
regards,
Thomas Schoenemann
___
Moses-support mailing list
Moses-support@mit.edu
http://mailman.mit.edu/mailman/listinfo/moses-support
Hi everyone,
right on time for EMNLP (I hope), I am releasing 150 gold alignments for
Europarl German-English. You can get them here:
http://www.maths.lth.se/matematiklth/personal/tosch/download.html
I am a native German speaker, made the alignments myself and took care with
it. The package al
Hi everybody!
I hope this is the right place to ask this. If not, please correct me. I have
a question on the file
Optimizer.cpp
in the directory
mert
(latest svn version).
Specifically, this affects line 379 and the following ones:
Point direction;
for(unsigned int i=0;ihttp:
13 matches
Mail list logo