Mailing-List: contact users-help@opennlp.apache.org; run by ezmlm
Precedence: bulk
Reply-To: users@opennlp.apache.org
Received-SPF: pass (nike.apache.org: domain of jimpil1985@gmail.com designates
 209.85.214.49 as permitted sender)
Message-ID: <4F7C3EE4.8060105@gmail.com>
Date: Wed, 04 Apr 2012 13:30:28 +0100
From: "Jim - FooBar();" <jimpil1985@gmail.com>
User-Agent: Mozilla/5.0 (X11; Linux x86_64;
 rv:11.0) Gecko/20120310 Thunderbird/11.0
MIME-Version: 1.0
To: users@opennlp.apache.org
Subject: Re: will version 1.5.3 be announced here?
References: <4F7871B5.80504@gmail.com> <4F788FC5.1010400@gmail.com>
 <4F79963C.6060308@gmail.com> <4F7ACAEB.3040501@gmail.com>
 <4F7AD2F1.3050509@gmail.com> <4F7AD3C9.9050800@gmail.com>
 <4F7ADE17.9020205@gmail.com>
 <CALV1_=JBb5TWQKm8J+f=sc69W9AWEo0m9q-QNyYzt46jGS4atQ@mail.gmail.com>
 <4F7C2691.8010104@gmail.com>
 <CALV1_=K5bxpoE4aK4YCNRz9-9V0FeOinxrdpztp9jVyCZZ1TDw@mail.gmail.com>
 <4F7C364C.5020905@gmail.com> <4F7C370D.2000509@gmail.com>
 <4F7C3CD9.70703@gmail.com>
In-Reply-To: <4F7C3CD9.70703@gmail.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 8bit

Also how come the Evaluator is an abstract class with abstract methods? 
Shouldn't it be an interface?

Jim

On 04/04/12 13:21, Jim - FooBar(); wrote:
> On 04/04/12 12:57, J�rn Kottmann wrote:
>> Doing this via our API should work right? You just create an 
>> evaluator and pass in
>> the DictionaryNameFinder instance and the test data. 
>
> Hmm...it is not that simple! what about the maxent model? Let's say you
>
> --create a class that extends Evaluator  (as all evaluators do)
> --allow the constructor of that class to take variable number of 
> arguments (of type TokenNameFinder) so we can use them later on
> --pretty much copy the code from TokenNameFinderEvaluator and paste it 
> in the new class
> --make sure the code asks both (or however many) name-finders before 
> it classifies a prediction as right or wrong.
>
> Up to here everything is quite straight forward...now the problems 
> begin!!!
> The Dictionary has to be evaluated on separate data than the 
> model...That is because the Dictionary can only deal with  the 
> "default tag" and nothing else! A quick workaround would be to retrain 
> the maxent model with "default" tags so i can evaluate it on the 
> dictionary's test-set but then what about multiple types???
>
> Jim
>
> p.s. Jorn, we had this discussion before... I do believe that the 
> Dictionary should be able to deal with whatever tag the user supplied 
> just like the maxent model does...