Re: Getting ambiguity while training my model

[Russ, Daniel (NIH/CIT) [E]]

You are getting an error because you are passing a null in for the 
AdaptiveFeatureGenerator.  Java doesn’t know if you are passing in and AFG or a 
byte[] (which makes it a different method), hence you get a message about an 
ambiguous signature.

You are using a depreciated method.  Use the 
train(String,String,ObjectStream<>,TrainingParameters,TokenNameFinderFactory) 
method.

You can create a TokenNameFinderFactory using the default constructor, Assuming 
you are using BioCodec format. I believe that is <START:person>John Smith<END>

Let me know if that works for you.


On 12/28/16, 6:13 PM, "mohd saif" <saif1.aans...@gmail.com> wrote:

    Hi,
    I am using opennlp-tool-1.6.0.
    For training my model i m using
    
    model = NameFinderME.train("en", "person", sampleStream,
    TrainingParameters.defaultParams(),
        null, Collections.<String, Object>emptyMap());
    
    For this i am getting below Error:
    The method train(String, String, ObjectStream<NameSample>,
    TrainingParameters, AdaptiveFeatureGenerator, Map<String,Object>) is
    ambiguous for the type NameFinderME
    
    Please help me out.
    
    2. Please tell me how to set cutoff so than i can train my model which have
    n-gram frequency=1. I mean how to set cutoff with the train API.
    
    Thanks alot...!!!