Thanks Hieu and Tom.
Hieu, could you please provide some more clarification regarding the
following. (inlined in bold)
the java client was actually done by Marwen AZOUZI
1. there's some notes about incremental training here
http://www.statmt.org/moses/?n=Moses.AdvancedFeatures#ntoc29
however, i think it's still quite difficult to set up.
2. You can use multiple phrase tables during decoding. It's not
controlled by XML-RPC. You need to train and tune the models.
*So I should create multiple phrase tables for each domain-specific
parallel corpus? For combining the phrase tables, I see that there's a
script in contrib/tmcombine to do that job. Is that only for phrase-based
translation models? Are there any solutions to combine the grammars for
hierarchical translation models? *
3. That's part of XML-RPC. See the Java client for how to set this up.
*Is that only for translating texts? What about documents?*
*Please let me know. Thanks!*
Gautham
P.S: I'm not getting the replies to the posts that I do in my mailbox.
Could someone let me know how I can enable that.
if you're creating a workflow, i suggest you integrate the experiment
management system into your workflow.
On 04/07/2012 01:26, Tom Hoar wrote:
>
> Gautham,
>
> The file mosesdecoder/contrib/server/SampleClient.java is a sample
> Java XML-RPC clients by Hieu.
>
> Tom
On Tue, Jul 3, 2012 at 2:34 PM, Gautham N <[email protected]> wrote:
> *Hi,
>
> We're actually looking to integrate Moses into our localization workflow.
> Our application is in Java and we're looking at using Moses'
> functionalities using xml-rpc calls.
>
> Specifically, we're looking at APIs for:
>
> 1. Incremental training (i.e. Avoid having to retrain the model
> from scratch every time we wish to use some new training data)
> 2. Domain-specific training (i.e. It should maintain separate
> phrase tables for each domain that the input data belongs),
> 3. Decoding
>
> The tutorial says that these can be achieved via xml-rpc calls.
> But, I don't find any examples or clear ways to do them.
> Can someone please provide some examples?
>
> Also, I would like to know if the training and decoding phases can be done
> in a
> distributed manner. *
> *
> *
> *
> *
> *Thanks!*
>
_______________________________________________
Moses-support mailing list
[email protected]
http://mailman.mit.edu/mailman/listinfo/moses-support
