def load_lm_model(lm_fname, order=2):
"""
function to load language model
"""
lm_model=srilm.initLM(order)
srilm.readLM(lm_model,lm_fname)
return lm_model
def __init__(self, m, filename):
"""use srilm language model from filename"""
self.m = m #order
self.model = srilm.initLM(self.m)
success = srilm.readLM(self.model, filename)
if not success:
print('LM load error')
sys.exit(1)
def __init__(self, model_path, order=2):
self.lm_path = model_path
self.order = order
self.history = []
self.EOS_ID = 1
self.history_len = order - 1
self.lm = initLM(order)
readLM(self.lm, self.lm_path)
self.vocab_size = howManyNgrams(self.lm, 1)
def __init__(self, model_path, order=2):
self.lm_path = model_path
self.order = order
self.history = []
self.EOS_ID = 1
self.history_len = order - 1
print u'Initialize LM with order {}'.format(order)
self.lm = initLM(order)
readLM(self.lm, self.lm_path)
self.vocab_size = howManyNgrams(self.lm, 1)
def __init__(self, path, ngram_order):
"""Creates a new n-gram language model predictor.
Args:
path (string): Path to the ARPA language model file
ngram_order (int): Order of the language model
Raises:
NameError. If srilm-swig is not installed
"""
super(SRILMPredictor, self).__init__()
self.history_len = ngram_order - 1
self.lm = initLM(ngram_order)
readLM(self.lm, path)
def __init__(self, path, ngram_order):
"""Creates a new n-gram language model predictor.
Args:
path (string): Path to the ARPA language model file
ngram_order (int): Order of the language model
Raises:
NameError. If srilm-swig is not installed
"""
super(SRILMPredictor, self).__init__()
self.history_len = ngram_order-1
self.lm = initLM(ngram_order)
readLM(self.lm, path)
def __init__(self, path, ngram_order, convert_to_ln=False):
"""Creates a new n-gram language model predictor.
Args:
path (string): Path to the ARPA language model file
ngram_order (int): Order of the language model
Raises:
NameError. If srilm-swig is not installed
"""
super(SRILMPredictor, self).__init__()
self.history_len = ngram_order-1
self.lm = initLM(ngram_order)
readLM(self.lm, path)
self.vocab_size = howManyNgrams(self.lm, 1)
self.convert_to_ln = convert_to_ln
if convert_to_ln:
import logging
logging.info("SRILM: Convert log scores to ln scores")
def __init__(self, path, ngram_order, convert_to_ln=False):
"""Creates a new n-gram language model predictor.
Args:
path (string): Path to the ARPA language model file
ngram_order (int): Order of the language model
convert_to_ln (bool): Whether to convert ld scores to ln.
Raises:
NameError. If srilm-swig is not installed
"""
super(SRILMPredictor, self).__init__()
self.history_len = ngram_order-1
self.lm = initLM(ngram_order)
readLM(self.lm, path)
self.vocab_size = howManyNgrams(self.lm, 1)
self.convert_to_ln = convert_to_ln
if convert_to_ln:
import logging
logging.info("SRILM: Convert log scores to ln scores")
def _init_lm(self, lm_fname):
self.lm_model = srilm.initLM(2)
srilm.readLM(self.lm_model, lm_fname)
def _init_lm(self,lm_fname):
self.lm_model=srilm.initLM(2)
srilm.readLM(self.lm_model,lm_fname)