class PosTaggerTestCase(unittest.TestCase):
def setUp(self):
base_dir = os.path.dirname(__file__)
self.pos_tagger = PosTagger(SegTagger(), os.path.join(base_dir, "./model/pos.crf.model"))
def tearDown(self):
pass
def testPosTagger(self):
res = self.pos_tagger.pos_as_iter(u"据以色列电台昨天报道")
self.assertEquals(list(res), [(u"据", "p"), (u"以色列", "nr"), (u"电台", "nn"), (u"昨天", "nt"), (u"报道", "vv")])
def testPosTaggerWithMultipleLines(self):
res = self.pos_tagger.pos_as_iter(u"据以色列电台昨天报道\n\n 巴勒斯坦和以色列")
self.assertEquals(list(res),
[(u"据", "p"), (u"以色列", "nr"), (u"电台", "nn"), (u"昨天", "nt"), (u"报道", "vv"), ("\n", "pu"),
("\n", "pu"), (" ", "pu"), (u"巴勒斯坦", "nr"), (u"和", "cc"), (u"以色列", "nr")])
class PosTaggerTestCase(unittest.TestCase):
def setUp(self):
base_dir = os.path.dirname(__file__)
self.pos_tagger = PosTagger(
SegTagger(), os.path.join(base_dir, "./model/pos.crf.model"))
def tearDown(self):
pass
def testPosTagger(self):
res = self.pos_tagger.pos_as_iter(u"据以色列电台昨天报道")
self.assertEquals(list(res), [(u"据", "p"), (u"以色列", "nr"),
(u"电台", "nn"), (u"昨天", "nt"),
(u"报道", "vv")])
def testPosTaggerWithMultipleLines(self):
res = self.pos_tagger.pos_as_iter(u"据以色列电台昨天报道\n\n 巴勒斯坦和以色列")
self.assertEquals(list(res),
[(u"据", "p"), (u"以色列", "nr"), (u"电台", "nn"),
(u"昨天", "nt"), (u"报道", "vv"), ("\n", "pu"),
("\n", "pu"), (" ", "pu"), (u"巴勒斯坦", "nr"),
(u"和", "cc"), (u"以色列", "nr")])
def setUp(self):
base_dir = os.path.dirname(__file__)
self.pos_tagger = PosTagger(SegTagger(), os.path.join(base_dir, "./model/pos.crf.model"))
def setUp(self):
self.extractor = ImpressionExtractor(PosTagger(SegTagger()))
| select(lambda nums: [words[num] for num in nums]) \
| select(lambda words: ''.join(words)) \
| as_list
short_phrase = ''.join(short_matches)
first_index = matches | select(extract_nums) | chain | min
last_index = matches | select(extract_nums) | chain | max
complete_phrase = ''.join(words[first_index:last_index +
1])
index = cur_index + (
[len(words[i]) for i in range(first_index)] | add)
yield (short_phrase, complete_phrase, index)
cur_index += len(line) + 1
if __name__ == '__main__':
from pynlpini import PosTagger
from pynlpini import SegTagger
with open("../../data/app/travel_comments/mafengwo_comments_raw.txt"
) as comment_file:
index = 0
ie = ImpressionExtractor(PosTagger(SegTagger()))
for line in comment_file:
line = line.decode("utf-8")
for item in ie.extract(line):
print item
index += 1
if index > 10:
exit()
def pos(txt):
global pos_tagger
if pos_tagger is None:
pos_tagger = PosTagger(seg_tagger)
return json.dumps(pos_tagger.pos_as_iter(txt) | as_list, ensure_ascii=False)
def setUp(self):
base_dir = os.path.dirname(__file__)
self.pos_tagger = PosTagger(
SegTagger(), os.path.join(base_dir, "./model/pos.crf.model"))