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. Fwd: Moses-support post from kunal.sachdeva.90@gmail.com
requires approval (Kunal Sachdeva)
----------------------------------------------------------------------
Message: 1
Date: Sat, 8 Mar 2014 00:18:18 +0530
From: Kunal Sachdeva <kunal.sachdeva.90@gmail.com>
Subject: [Moses-support] Fwd: Moses-support post from
kunal.sachdeva.90@gmail.com requires approval
To: moses-support@mit.edu
Message-ID:
<CAOF+9df6wca=gY_N4_JcVHXBwvULurWPpe6hU=nST6mxSN7W-g@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"
Hello,
Thanks for the help. I am already subscribed to the MOSES mailing list .
I have some more doubts regarding the same.
We have added a new feature, providing as weight to the phrase table.
For example: my phrase table looks like this.
! " whispered the ||| ?? ??????? ?? ??? ||| 0.112086 0.000813219 0.0560432
2.79321e-07 2.718 FLOAT_NUM ||| 3-0 0-1 1-1 2-1 2-2 2-3 ||| 1 2 1
I made the following changes to moses.ini
0 0 0 5 /home/seecat/WMT/eng-hin/WMT_new/model/phrase-table.3 ======> 0 0 0
6 /home/seecat/WMT/eng-hin/WMT_new/model/phrase-table.3
# translation model weights
[weight-t]
0.20
0.20
0.20
0.20
0.20
0.20(added now)
Now, I am trying to tune the system again, and i am getting an error in
TUNING.tune.STDERR
Check weightT.size() == scoreVector.size() failed in
moses/TargetPhrase.cpp:137
Aborted (core dumped)
Exit code: 134
The decoder died. CONFIG WAS -w -0.217391 -lm 0.108696 -d 0.065217 0.065217
0.065217 0.065217 0.065217 0.065217 0.065217 -tm 0.043478 0.043478 0.043478
0.043478 0.043478
Do I need to make some other change to consider my new feature ?
Thanks and Regards
Kunal Sachdeva
On Fri, Mar 7, 2014 at 10:13 PM, Hieu Hoang <Hieu.Hoang@ed.ac.uk> wrote:
> To post to the mailing list, you must subscribe to it first. You can
> subscribe here:
> http://mailman.mit.edu/mailman/listinfo/moses-support
> to answer you question - yes, " 3-0 0-1 1-1 2-1 2-2 2-3" is the word
> alignment.
>
> Most of the time, the word alignment doesn't affect decoding. However, its
> safest to delete the alignments if you're gonna start changing the source
> or target phrase
>
> ---------- Forwarded message ----------
> From: <moses-support-owner@mit.edu>
> Date: 7 March 2014 16:35
> Subject: Moses-support post from kunal.sachdeva.90@gmail.com requires
> approval
> To: moses-support-owner@mit.edu
>
>
> As list administrator, your authorization is requested for the
> following mailing list posting:
>
> List: Moses-support@mit.edu
> From: kunal.sachdeva.90@gmail.com
> Subject: Reg: Phrase Table format
> Reason: Post by non-member to a members-only list
>
> At your convenience, visit:
>
> http://mailman.mit.edu/mailman/admindb/moses-support
>
> to approve or deny the request.
>
>
> ---------- Forwarded message ----------
> From: Kunal Sachdeva <kunal.sachdeva.90@gmail.com>
> To: moses-support@mit.edu
> Cc:
> Date: Fri, 7 Mar 2014 22:05:41 +0530
> Subject: Reg: Phrase Table format
> Hello,
> I have a doubt regarding the information provided in the phrase-table in
> the trained moses model
>
> eg : my phrase table entry looks like this
>
> ! " whispered the ||| ?? ??????? ?? ??? ||| 0.112086 0.000813219 0.0560432
> 2.79321e-07 2.718 ||| 3-0 0-1 1-1 2-1 2-2 2-3 ||| 1 2 1
>
> So the column containing " 3-0 0-1 1-1 2-1 2-2 2-3 " provides me
> alignments with in the phrase, if I am not wrong ??
>
> Also I am trying to clean the phrase table, so If I am modifying my phrase
> entry to
>
> whispered the ||| ?? ??????? ?? ??? ||| 0.112086 0.000813219 0.0560432
> 2.79321e-07 2.718 ||| 3-0 0-1 1-1 2-1 2-2 2-3 ||| 1 2 1
>
> then what other things I need to change ??
>
> Do I need to make change to those alignments, or I can ignore it ( will
> that make a difference in decoding ) ?
>
> Regards,
> Kunal Sachdeva
>
>
>
> ---------- Forwarded message ----------
> From: moses-support-request@mit.edu
> To:
> Cc:
> Date:
> Subject: confirm 6664c4f80895c0eabf81b6674a61c90a37a92d4e
> If you reply to this message, keeping the Subject: header intact,
> Mailman will discard the held message. Do this if the message is
> spam. If you reply to this message and include an Approved: header
> with the list password in it, the message will be approved for posting
> to the list. The Approved: header can also appear in the first line
> of the body of the reply.
>
>
>
> --
> Hieu Hoang
> Research Associate
> University of Edinburgh
> http://www.hoang.co.uk/hieu
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20140308/d844ed87/attachment-0001.htm
------------------------------
_______________________________________________
Moses-support mailing list
Moses-support@mit.edu
http://mailman.mit.edu/mailman/listinfo/moses-support
End of Moses-support Digest, Vol 89, Issue 18
*********************************************
Subscribe to:
Post Comments (Atom)
0 Response to "Moses-support Digest, Vol 89, Issue 18"
Post a Comment