def preprocess(profiles, network_name):
for label in profiles.iterkeys():
model = BigramLanguageModel()
for profile in profiles[label]:
model.add(profile.description.lower())
model.prune()
with open(network_name + 'BigramLM-' + label, 'wb') as output:
cPickle.dump(model, output)
def preprocess(profiles, network_name):
for label in profiles.iterkeys():
model = BigramLanguageModel()
for profile in profiles[label]:
for post in (x.content.lower() for x in profile.posts):
model.add(post)
model.prune()
with open(network_name + 'PostBigramLM-' + label, 'wb') as output:
cPickle.dump(model, output)