Line 164 of the TokenNameFinderCrossValidator.java @ master branch requires
a byte array. It is not user friendly. Saurabh Jain, for now you could
create the file and load it to a byte array. You can open an issue for this
and if you like provide a patch.

https://github.com/apache/opennlp/blob/master/opennlp-tools/src/main/java/opennlp/tools/namefind/TokenNameFinderCrossValidator.java#L164

Regards

2017-04-21 11:51 GMT-03:00 Saurabh Jain <[email protected]>:

> Hi Daniel
>
> I want to use already provided functionality for cross validation of
> NameFinder that is why trying to use already provided api.
>
> Hi Jeff
>
> Thank you, I am already familiar with this approach. I want to set it by
> java source code.
>
>
>
>
> On Fri, Apr 21, 2017 at 7:55 PM, Jeff Zemerick <[email protected]>
> wrote:
>
> > The byte array that the constructor to TokenNameFinderCrossValidator is
> > asking for is the feature generators as XML, such as (and borrowed from
> > [1]):
> >
> > <generators>
> >   <cache>
> >     <generators>
> >       <window prevLength = "2" nextLength = "2">
> >         <tokenclass/>
> >       </window>
> >       <window prevLength = "2" nextLength = "2">
> >         <token/>
> >       </window>
> >       <definition/>
> >       <prevmap/>
> >       <bigram/>
> >       <sentence begin="true" end="false"/>
> >       <window prevLength = "2" nextLength = "2">
> >         <brownclustertoken dict="brownCluster" />
> >       </window>
> >       <brownclustertokenclass dict="brownCluster" />
> >       <brownclusterbigram dict="brownCluster" />
> >       <wordcluster dict="word2vec.cluster" />
> >       <wordcluster dict="clark.cluster" />
> >     </generators>
> >   </cache>
> > </generators>
> >
> > An an example, in TokenNameFinderFactory you can see
> > in loadDefaultFeatureGeneratorBytes() how the default feature generator
> is
> > loaded from XML to a byte array when no feature generators are provided.
> >
> > Jeff
> >
> > [1]
> > https://opennlp.apache.org/documentation/1.7.0/manual/
> > opennlp.html#tools.namefind.training.featuregen
> >
> >
> >
> > On Fri, Apr 21, 2017 at 9:17 AM, Saurabh Jain <[email protected]
> >
> > wrote:
> >
> > > Hi All
> > >
> > > I have defined feature generator for OpenNLP name finder in java source
> > > code as an object of *CachedFeatureGenerator *. I have to cross
> validate
> > > NameFinder and whatever api I am able to find in code accepts feature
> > > generators as byte array. Problem is  *CachedFeatureGenerator *is not
> > > serializable (as far as I came to know). Is there any api in OpenNLP
> > > NameFinder for cross validation which accept *CachedFeatureGenerator
> *as
> > > feature generator or is there any other way ?
> > >
> > > --
> > > *Thanks & Regards*
> > >
> > >
> > > *Saurabh Jain *
> > > *AI Developer*
> > >
> > > *Active Intelligence  *
> > >
> > > *"*
> > > *To do a thing yesterday was the best time . Second best time is today
> > .” *
> > >
> >
>
>
>
> --
> *Thanks & Regards*
>
>
> *Saurabh Jain *
> *AI Developer*
>
> *Active Intelligence  *
>
> *"*
> *To do a thing yesterday was the best time . Second best time is today .” *
>

Reply via email to