def main():
train = "../data/Verne.80jours.en"
# train = "../data/simplewiki-20140903-pages-articles.200000first.100000last.txt"
# train = "../data/wsj.words"
order = 3
n_iter = 100
vocabulary = Vocabulary()
logging.info('Reading training corpus')
with open(train) as train:
training_corpus = read_corpus(train, vocabulary)
base = Uniform(len(vocabulary))
model = PYPLM(order, base)
logging.info('Training model of order %d', order)
run_sampler(model, training_corpus, n_iter)
