Hello,
this happens if there is not enough training data or the data does not
at least
contain a couple of names.
Jön
On 03/01/2014 09:58 PM, Laszlo Hornyak wrote:
Hi users,
I am new user to OpenNLP and trying to create my training files since I
haven't found training data for hungarian language. The sentence trainer
worked perfect, but I also want to use the TokenNameFinder so I used the
training tool
./bin/opennlp TokenNameFinderTrainer -lang hu -data hu.train -model
hu-person.bin -encoding UTF-8
and the trainer reports this exception
Indexing events using cutoff of 5
Computing event counts... done. 18496 events
Indexing... done.
Sorting and merging events... done. Reduced 18496 events to 14597.
Done indexing.
Incorporating indexed data for training...
done.
Number of Event Tokens: 14597
Number of Outcomes: 1
Number of Predicates: 4081
...done.
Computing model parameters ...
Performing 100 iterations.
1: ... loglikelihood=0.0 1.0
2: ... loglikelihood=0.0 1.0
Exception in thread "main" java.lang.IllegalArgumentException: Model not
compatible with name finder!
at
opennlp.tools.namefind.TokenNameFinderModel.<init>(TokenNameFinderModel.java:81)
at
opennlp.tools.namefind.TokenNameFinderModel.<init>(TokenNameFinderModel.java:106)
at opennlp.tools.namefind.NameFinderME.train(NameFinderME.java:374)
at opennlp.tools.namefind.NameFinderME.train(NameFinderME.java:403)
at
opennlp.tools.cmdline.namefind.TokenNameFinderTrainerTool.run(TokenNameFinderTrainerTool.java:179)
at opennlp.tools.cmdline.CLI.main(CLI.java:222)
What could be the problem? Is it something with the training file?
I am running OpenNLP 1.5.3
Thank you,
Laszlo
