def process_html(node, libs_factory=None, data=None):
output = OutputBuilder()
if not libs_factory:
libs_factory = LibFactory()
env = Environment(data=data, output=output, libs=libs_factory)
posts_handler(node, env)
return output.get_value()
def __init__(self, output_dir, template_dir):
self.file_name = 'chars.txt'
OutputBuilder.__init__(self, output_dir, template_dir)
from output_builder import OutputBuilder
from variants_generation import PrimaryCandidates, SecondaryCandidates
from candidate_selection import Selector
file500 = '/Tweets/tweet-norm-dev500_annotated.txt'
file100 = '/Tweets/tweet-norm-dev100_annotated.txt'
outputpath = '/home/alangb/Escritorio/result100.txt'
tweets_file = path.split(path.abspath(__file__))[0] + file100
splitter = Tw_Splitter(tweets_file)
picker = OOVpicker(splitter.texts)
classifier = OOVclassifier()
primary = PrimaryCandidates(2)
secondary = SecondaryCandidates()
selector = Selector()
output = OutputBuilder(outputpath)
oovs = picker.OOV
tokenized = picker.tokenized
correct = defaultdict(dict)
for tweet_id, tweet in oovs.items():
for j, sent in tweet.items(): # j is number of the sent
for_prev = tokenized[tweet_id][j]
correct[tweet_id][j] = []
for word, pos in sent:
class_number = classifier.classify(word)
# if class is variant
if class_number == 0:
IVcandidates = primary.generate(word)
# if no primary candidates generated
if len(IVcandidates) == 0:
def __init__(self, file_name, output_dir, template_dir):
self.file_name = file_name
OutputBuilder.__init__(self, output_dir, template_dir)
def __init__(self, output_dir, template_dir):
self.file_name = 'chars.txt'
OutputBuilder.__init__(self, output_dir, template_dir)
def __init__(self, file_name, output_dir, template_dir):
self.file_name = file_name
OutputBuilder.__init__(self, output_dir, template_dir)
def __init__(self, output_dir, template_dir):
self.file_name = "char_codes.txt"
OutputBuilder.__init__(self, output_dir, template_dir)
