def tag(test_data_filename, result_filename, hmm_model_filename):
print '1. load Hmm model'
tagger = ViterbiTagger(3)
tagger.read_counts(file(hmm_model_filename))
print '2. tag test file'
tagger.tag(file(test_data_filename), file(result_filename, 'w'))
def train(train_data_filename, rare_train_data_filename, hmm_model_filename, rare_words_rule):
print '1. train hmm model'
hmm_model = ViterbiTagger(3)
hmm_model.rare_words_rule = rare_words_rule
hmm_model.train(file(train_data_filename))
print '2. process rare words'
util.process_rare_words(
file(train_data_filename),
file(rare_train_data_filename, 'w'),
hmm_model.rare_words,
hmm_model.rare_words_rule)
print '3. train hmm model again using the new train data'
hmm_model_rare = ViterbiTagger(3)
hmm_model_rare.train(file(rare_train_data_filename))
hmm_model_rare.write_counts(file(hmm_model_filename, 'w'))
def tag(test_data_filename, result_filename, hmm_model_filename):
print '1. Cargando el modelo Hmm'
tagger = ViterbiTagger(3)
tagger.read_counts(file(hmm_model_filename))
print '2. Archivo de prueba'
tagger.tag(file(test_data_filename), file(result_filename, 'w'))
def train(train_data_filename, rare_train_data_filename, hmm_model_filename,
rare_words_rule):
print '1. Modo entrenamiento hmm'
hmm_model = ViterbiTagger(3)
hmm_model.rare_words_rule = rare_words_rule
hmm_model.train(file(train_data_filename))
print '2. Procesando palabras raras'
util.process_rare_words(file(train_data_filename),
file(rare_train_data_filename, 'w'),
hmm_model.rare_words, hmm_model.rare_words_rule)
print '3. Entrenando hmm usando la nueva data de entrenamiento'
hmm_model_rare = ViterbiTagger(3)
hmm_model_rare.train(file(rare_train_data_filename))
hmm_model_rare.write_counts(file(hmm_model_filename, 'w'))