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: Untuneable feature score components? (Hieu Hoang)
2. Re: Untuneable feature score components? (Rico Sennrich)
3. Re: Untuneable feature score components? (Matthias Huck)
4. Re: Untuneable feature score components? (Hieu Hoang)
5. Re: Untuneable feature score components? (Matthias Huck)
----------------------------------------------------------------------
Message: 1
Date: Fri, 23 Jan 2015 17:38:43 +0000
From: Hieu Hoang <hieuhoang@gmail.com>
Subject: Re: [Moses-support] Untuneable feature score components?
To: Matthias Huck <mhuck@inf.ed.ac.uk>, Moses-support
<moses-support@mit.edu>
Message-ID: <5217733B-0436-4F0B-BE3C-8F86264904CF@gmail.com>
Content-Type: text/plain; charset="utf-8"
The whole feature becomes untuneable.
I suppose u can make the pt untuneable, the write another (tuneable) ff which grabs whatever scores it wants from the pt
On 23 January 2015 16:55:56 GMT+00:00, Matthias Huck <mhuck@inf.ed.ac.uk> wrote:
>Hi,
>
>Is there any existing functionality to set only specific score
>components of a feature function as untuneable?
>
>Feature functions have a boolean "tuneable" parameter, but it affects
>all the scores produced by it. It doesn't help in case I want to switch
>off individual scores from a phrase table, for instance. Or manually
>assign large scaling factors to certain score components prior to
>tuning. As far as I know, right now I'm only able to do so if the score
>I'm interested in is the only score produced by a feature function.
>
>If anyone has already implemented something like that, please let me
>know.
>
>Cheers,
>Matthias
>
>
>
>
>--
>The University of Edinburgh is a charitable body, registered in
>Scotland, with registration number SC005336.
>
>_______________________________________________
>Moses-support mailing list
>Moses-support@mit.edu
>http://mailman.mit.edu/mailman/listinfo/moses-support
--
Sent while bumping into things
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20150123/53b0d163/attachment-0001.htm
------------------------------
Message: 2
Date: Fri, 23 Jan 2015 17:45:59 +0000 (UTC)
From: Rico Sennrich <rico.sennrich@gmx.ch>
Subject: Re: [Moses-support] Untuneable feature score components?
To: moses-support@mit.edu
Message-ID: <loom.20150123T183410-615@post.gmane.org>
Content-Type: text/plain; charset=us-ascii
Matthias Huck <mhuck@...> writes:
>
> Hi,
>
> Is there any existing functionality to set only specific score
> components of a feature function as untuneable?
Hi Matthias,
There is the option --activate-features in mert-moses.pl, but I believe it
only works for MERT (if at all), and it also had some weird problems the
last time I tried it (like normalizing all active features to sum to 1).
A word of warning: the 'tuneable' parameter hides the feature from the
optimizer, which can mislead its search. The parameter is most useful for
features which you want to keep at 0, or for which you set a very high
penalty so that hypotheses that trigger it normally aren't in the n-best list.
best wishes,
Rico
------------------------------
Message: 3
Date: Fri, 23 Jan 2015 18:09:11 +0000
From: Matthias Huck <mhuck@inf.ed.ac.uk>
Subject: Re: [Moses-support] Untuneable feature score components?
To: Hieu Hoang <hieuhoang@gmail.com>
Cc: Moses-support <moses-support@mit.edu>
Message-ID: <1422036551.2192.239.camel@portedgar>
Content-Type: text/plain; charset="UTF-8"
That's not flexible enough. There should be something like:
[feature]
MyFeature name=MyFeature0 tuneable=0,1,0
[weight]
MyFeature0= 0.0 0.1 1
MyFeature has 3 score components. I want to tune the second component,
deactivate the first component, and set the scaling factor of the third
component manually to 1.
Currently the "tuneable" parameter is boolean and allows me to manually
set scaling factors for either no score component or all of them.
On Fri, 2015-01-23 at 17:38 +0000, Hieu Hoang wrote:
> The whole feature becomes untuneable.
>
> I suppose u can make the pt untuneable, the write another (tuneable)
> ff which grabs whatever scores it wants from the pt
>
> On 23 January 2015 16:55:56 GMT+00:00, Matthias Huck
> <mhuck@inf.ed.ac.uk> wrote:
> Hi,
>
> Is there any existing functionality to set only specific score
> components of a feature function as untuneable?
>
> Feature functions have a boolean "tuneable" parameter, but it affects
> all the scores produced by it. It doesn't help in case I want to switch
> off individual scores from a phrase table, for instance. Or manually
> assign large scaling factors to certain score components prior to
> tuning. As far as I know, right now I'm only able to do so if the score
> I'm interested in is the only score produced by a feature function.
>
> If anyone has already implemented something like that, please let me
> know.
>
> Cheers,
> Matthias
>
>
>
>
> --
> Sent while bumping into things
--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.
------------------------------
Message: 4
Date: Fri, 23 Jan 2015 18:18:12 +0000
From: Hieu Hoang <hieuhoang@gmail.com>
Subject: Re: [Moses-support] Untuneable feature score components?
To: Matthias Huck <mhuck@inf.ed.ac.uk>
Cc: Moses-support <moses-support@mit.edu>
Message-ID: <6D3B2F70-AC70-45A1-87F0-52ECC43574F9@gmail.com>
Content-Type: text/plain; charset="utf-8"
True, but that complicates the framework, and doesn't deal with sparse features.
By adding another ff which grabs scores from the pt, u can arbitrarily transform the scores
On 23 January 2015 18:09:11 GMT+00:00, Matthias Huck <mhuck@inf.ed.ac.uk> wrote:
>That's not flexible enough. There should be something like:
>
>[feature]
>MyFeature name=MyFeature0 tuneable=0,1,0
>
>[weight]
>MyFeature0= 0.0 0.1 1
>
>
>MyFeature has 3 score components. I want to tune the second component,
>deactivate the first component, and set the scaling factor of the third
>component manually to 1.
>
>Currently the "tuneable" parameter is boolean and allows me to manually
>set scaling factors for either no score component or all of them.
>
>
>
>
>On Fri, 2015-01-23 at 17:38 +0000, Hieu Hoang wrote:
>> The whole feature becomes untuneable.
>>
>> I suppose u can make the pt untuneable, the write another (tuneable)
>> ff which grabs whatever scores it wants from the pt
>>
>> On 23 January 2015 16:55:56 GMT+00:00, Matthias Huck
>> <mhuck@inf.ed.ac.uk> wrote:
>> Hi,
>>
>> Is there any existing functionality to set only specific
>score
>> components of a feature function as untuneable?
>>
>> Feature functions have a boolean "tuneable" parameter, but it
>affects
>> all the scores produced by it. It doesn't help in case I want
>to switch
>> off individual scores from a phrase table, for instance. Or
>manually
>> assign large scaling factors to certain score components
>prior to
>> tuning. As far as I know, right now I'm only able to do so if
>the score
>> I'm interested in is the only score produced by a feature
>function.
>>
>> If anyone has already implemented something like that, please
>let me
>> know.
>>
>> Cheers,
>> Matthias
>>
>>
>>
>>
>> --
>> Sent while bumping into things
>
>
>
>--
>The University of Edinburgh is a charitable body, registered in
>Scotland, with registration number SC005336.
--
Sent while bumping into things
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20150123/1f3df37c/attachment-0001.htm
------------------------------
Message: 5
Date: Fri, 23 Jan 2015 18:37:31 +0000
From: Matthias Huck <mhuck@inf.ed.ac.uk>
Subject: Re: [Moses-support] Untuneable feature score components?
To: Hieu Hoang <hieuhoang@gmail.com>
Cc: Moses-support <moses-support@mit.edu>
Message-ID: <1422038251.2192.254.camel@portedgar>
Content-Type: text/plain; charset="UTF-8"
On Fri, 2015-01-23 at 18:18 +0000, Hieu Hoang wrote:
> True, but that complicates the framework, and doesn't deal with sparse
> features.
Why does it complicate the framework? Isn't the trick about "tuneable"
mostly that you don't write those scores to the n-best list?
We can even keep a boolean "tuneable" parameter and have another
parameter "tuneable-components" (boolean vector sized like the weights
vector).
>
> By adding another ff which grabs scores from the pt, u can arbitrarily
> transform the scores
Yeah I know. I can get around these things by writing more feature
functions. For removing scores from phrase tables, I can also just
process the phrase table file with awk, delete the score columns I don't
need and write it to another file. But something user-friendly would be
more appealing. Setting up contrastive experiments could be done much
more rapidly with what I'm asking for. And maybe somebody on the mailing
list has implemented this and never put it into master?
I want it for MIRA, btw.
I think it should be added if it doesn't exist somewhere yet. Unless
someone has strong objections.
>
> On 23 January 2015 18:09:11 GMT+00:00, Matthias Huck
> <mhuck@inf.ed.ac.uk> wrote:
> That's not flexible enough. There should be something like:
>
> [feature]
> MyFeature name=MyFeature0 tuneable=0,1,0
>
> [weight]
> MyFeature0= 0.0 0.1 1
>
>
> MyFeature has 3 score components. I want to tune the second component,
> deactivate the first component, and set the scaling factor of the third
> component manually to 1.
>
> Currently the "tuneable" parameter is boolean and allows me to manually
> set scaling factors for either no score component or all of them.
>
>
>
>
> On Fri, 2015-01-23 at 17:38 +0000, Hieu Hoang wrote:
> The whole feature becomes untuneable.
>
> I suppose u can make the pt untuneable, the write another (tuneable)
> ff which grabs whatever scores it wants from the pt
>
> On 23 January 2015 16:55:56 GMT+00:00, Matthias!
> Huck
> <mhuck@inf.ed.ac.uk> wrote:
> Hi,
>
> Is there any existing functionality to set only specific score
> components of a feature function as untuneable?
>
> Feature functions have a boolean "tuneable" parameter, but it affects
> all the scores produced by it. It doesn't help in case I want to switch
> off individual scores from a phrase table, for instance. Or manually
> assign large scaling factors to certain score components prior to
> tuning. As far as I know, right now I'm only able to do so if the score
> I'm interested in is the only score produced by a feature function.
>
> If anyone has already implemented something like that, please let me
> know.
>
> Cheers,
> Matthias
>
>
>
>
> --
> Sent while bumping !
> into
> things
>
>
>
> --
> Sent while bumping into things
--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.
------------------------------
_______________________________________________
Moses-support mailing list
Moses-support@mit.edu
http://mailman.mit.edu/mailman/listinfo/moses-support
End of Moses-support Digest, Vol 99, Issue 56
*********************************************
Subscribe to:
Post Comments (Atom)
0 Response to "Moses-support Digest, Vol 99, Issue 56"
Post a Comment