def wrapped(l1):
parser = linkshift.parser.Parser(parser_dir, solver_type=solver_type,
clean=True, C=l1)
dev_gold = linkshift.io_parse.read_conll(dev_loc.open().read())
train = linkshift.io_parse.read_conll(train_loc.open().read())
parser.train(train)
dev = linkshift.io_parse.read_pos(get_pos(dev_loc.open().read()))
acc = parser.add_parses(dev, gold=dev_gold) * 100
wrapped.models[l1] = acc
return acc
def main(parser_dir, text_loc, out_dir, use_gold=False, profile=False, debug=False):
if debug:
linkshift.parser.set_debug(debug)
if not os.path.exists(out_dir):
os.mkdir(out_dir)
yield "Loading parser"
parser = linkshift.parser.load_parser(parser_dir)
sentences = linkshift.io_parse.read_pos(open(text_loc).read())
#sentences.connect_sentences(1700)
if profile:
cProfile.runctx("parser.add_parses(sentences,gold=gold_sents)",
globals(), locals(), "Profile.prof")
s = pstats.Stats("Profile.prof")
s.strip_dirs().sort_stats("time").print_stats()
else:
t1 = time.time()
parser.add_parses(sentences)
t2 = time.time()
print '%d sents took %0.3f ms' % (sentences.length, (t2-t1)*1000.0)
sentences.write_parses(open(os.path.join(out_dir, 'parses'), 'w'))
