def setUp(self):
self.tagger = GeniaTagger()
self.sentence = "Inhibition of NF-kappa beta activation reversed " \
"the anti-apoptotic effect of isochamaejasmin."
self.tokens = [
Token('Inhibition', 'Inhibition', 'NN', 'B-NP', 'O'),
Token('of', 'of', 'IN', 'B-PP', 'O'),
Token('NF-kappa', 'NF-kappa', 'NN', 'B-NP', 'B-protein'),
Token('beta', 'beta', 'NN', 'I-NP', 'I-protein'),
Token('activation', 'activation', 'NN', 'I-NP', 'O'),
Token('reversed', 'reverse', 'VBD', 'B-VP', 'O'),
Token('the', 'the', 'DT', 'B-NP', 'O'),
Token('anti-apoptotic', 'anti-apoptotic', 'JJ', 'I-NP', 'O'),
Token('effect', 'effect', 'NN', 'I-NP', 'O'),
Token('of', 'of', 'IN', 'B-PP', 'O'),
Token('isochamaejasmin', 'isochamaejasmin', 'NN', 'B-NP', 'O'),
Token('.', '.', '.', 'O', 'O')
]
class GeniaTaggerTests(TestCase):
# TODO: test "empty" word/stem handling works on geniatagger return lines of the form:
# \t\t<pos>\t<phrase>\t<entity>
def setUp(self):
self.tagger = GeniaTagger()
self.sentence = "Inhibition of NF-kappa beta activation reversed " \
"the anti-apoptotic effect of isochamaejasmin."
self.tokens = [
Token('Inhibition', 'Inhibition', 'NN', 'B-NP', 'O'),
Token('of', 'of', 'IN', 'B-PP', 'O'),
Token('NF-kappa', 'NF-kappa', 'NN', 'B-NP', 'B-protein'),
Token('beta', 'beta', 'NN', 'I-NP', 'I-protein'),
Token('activation', 'activation', 'NN', 'I-NP', 'O'),
Token('reversed', 'reverse', 'VBD', 'B-VP', 'O'),
Token('the', 'the', 'DT', 'B-NP', 'O'),
Token('anti-apoptotic', 'anti-apoptotic', 'JJ', 'I-NP', 'O'),
Token('effect', 'effect', 'NN', 'I-NP', 'O'),
Token('of', 'of', 'IN', 'B-PP', 'O'),
Token('isochamaejasmin', 'isochamaejasmin', 'NN', 'B-NP', 'O'),
Token('.', '.', '.', 'O', 'O')
]
def tearDown(self):
del self.tagger
def testTagger(self):
for dummy in range(2):
self.tagger.send(self.sentence)
for idx, token in enumerate(iter(self.tagger)):
self.assertTupleEqual(token, self.tokens[idx])
def testBadPath(self):
self.assertRaises(AssertionError, GeniaTagger, "/fail", "whatever")
self.assertRaises(AssertionError, GeniaTagger, "whatever", "/fail")
logging.basicConfig(level=args.loglevel,
format='%(asctime)s %(levelname)s: %(message)s')
if args.output == NORMALIZED:
method = normalize
elif args.output == TABULAR:
method = tagging
elif args.output == ALIGNED:
method = align
else:
parser.error("unknown output option " + args.output)
method = lambda *args: None
try:
pos_tagger = GeniaTagger()
ner_tagger = NerSuite(args.model)
qualifier_list = [l.strip() for l in args.qranks]
raw_dict_data = [
dictionaryReader(d, qualifier_list, args.separator)
for d in args.dictionary
]
# a tokenizer that skips Unicode Categories Zs and Pd:
tokenizer = WordTokenizer(skipTags={'space'}, skipOrthos={'e'})
dictionaries = [
Dictionary(stream, tokenizer) for stream in raw_dict_data
]
logging.info("initialized %s dictionaries", len(dictionaries))
lst = [dictionaries, tokenizer, pos_tagger, ner_tagger]
kwds = dict(sep=args.separator,
tag_all_nouns=args.nouns,