Module: OpenNLP::Config

Defined in:

lib/open-nlp/config.rb

Constant Summary

NameToClass =

{
  categorizer: ['DoccatModel', 'opennlp.tools.doccat'],
  chunker: ['ChunkerModel', 'opennlp.tools.chunker'],
  detokenizer: ['DetokenizationDictionary', 'opennlp.tools.tokenize'],
  name_finder: ['TokenNameFinderModel', 'opennlp.tools.namefind'],
  parser: ['ParserModel', 'opennlp.tools.parser'],
  pos_tagger: ['POSModel', 'opennlp.tools.postag'],
  sentence_detector: ['SentenceModel', 'opennlp.tools.sentdetect'],
  tokenizer: ['TokenizerModel', 'opennlp.tools.tokenize']
}

ClassToName =

{
  'ChunkerME' => :chunker,
  'DictionaryDetokenizer' => :detokenizer,
  'DocumentCategorizerME' => :categorizer,
  'NameFinderME' => :name_finder,
  'POSTaggerME' => :pos_tagger,
  'Parser' => :parser,
  'SentenceDetectorME' => :sentence_detector,
  'TokenizerME' => :tokenizer,
}

DefaultModels =

{
  chunker: {
    english: 'en-chunker.bin'
  },
  detokenizer: {
    english: 'en-detokenizer.xml'
  },
  # Intentionally left empty.
  # Available for English, Spanish, Dutch.
  name_finder: {},
  parser: {
    english: 'en-parser-chunking.bin'
  },
  pos_tagger: { 
    english: 'en-pos-maxent.bin',
    danish: 'da-pos-maxent.bin',
    german: 'de-pos-maxent.bin',
    dutch: 'nl-pos-maxent.bin',
    portuguese: 'pt-pos-maxent.bin',
    swedish: 'se-pos-maxent.bin'
  },
  sentence_detector: {
    english: 'en-sent.bin',
    german: 'de-sent.bin',
    danish: 'da-sent.bin',
    dutch: 'nl-sent.bin',
    portuguese: 'pt-sent.bin',
    swedish: 'se-sent.bin'
  },
  tokenizer: {
    english: 'en-token.bin',
    danish: 'da-token.bin',
    german: 'de-token.bin',
    dutch: 'nl-token.bin',
    portuguese: 'pt-token.bin',
    swedish: 'se-token.bin'
  }
}

RequiresModel =

Classes that require a model as first argument to constructor.

[
  'SentenceDetectorME', 'NameFinderME', 'DictionaryDetokenizer',
  'TokenizerME', 'ChunkerME', 'POSTaggerME', 'Parser'
]