def test_spans_doesnt_change_tree(self):
t = Tree.fromstring(
"""
(S
(NP (Det el) (Noun gato))
(VP (Verb come) (NP (Noun pescado) (Adj crudo)))
)
""")
t2 = t.copy(deep=True)
spans(t)
self.assertEqual(t, t2)
def test_spans_with_unary(self):
t = Tree.fromstring(
"""
(S
(NP (Det el) (Noun gato))
(VP (Verb come) (NP (Noun pescado)))
)
""")
s = spans(t, unary=True)
s2 = {('S', 0, 3), ('NP', 0, 1), ('VP', 2, 3), ('NP', 3, 3)}
self.assertEqual(s, s2)
def test_spans(self):
t = Tree.fromstring(
"""
(S
(NP (Det el) (Noun gato))
(VP (Verb come) (NP (Noun pescado) (Adj crudo)))
)
""")
s = spans(t)
s2 = {('S', 0, 4), ('NP', 0, 1), ('VP', 2, 4), ('NP', 3, 4)}
self.assertEqual(s, s2)
corpus = SimpleAncoraCorpusReader('ancora/ancora-2.0/', files)
parsed_sents = list(corpus.parsed_sents())
print('Parsing...')
hits, total_gold, total_model = 0, 0, 0
n = len(parsed_sents)
format_str = '{:3.1f}% ({}/{}) (P={:2.2f}%, R={:2.2f}%, F1={:2.2f}%)'
progress(format_str.format(0.0, 0, n, 0.0, 0.0, 0.0))
for i, gold_parsed_sent in enumerate(parsed_sents):
tagged_sent = gold_parsed_sent.pos()
# parse
model_parsed_sent = model.parse(tagged_sent)
# compute labeled scores
gold_spans = spans(gold_parsed_sent, unary=False)
model_spans = spans(model_parsed_sent, unary=False)
hits += len(gold_spans & model_spans)
total_gold += len(gold_spans)
total_model += len(model_spans)
# compute labeled partial results
prec = float(hits) / total_model * 100
rec = float(hits) / total_gold * 100
f1 = 2 * prec * rec / (prec + rec)
progress(
format_str.format(float(i + 1) * 100 / n, i + 1, n, prec, rec, f1))
print('')
print('Parsed {} sentences'.format(n))
n = len(parsed_sents)
format_str = '{:3.1f}% ({}/{}) | Labeled: (P={:2.2f}%, R={:2.2f}%, \
F1={:2.2f}%) | Unlabeled: (P={:2.2f}%, R={:2.2f}%, F1={:2.2f}%)'
progress(format_str.format(0.0, 0, n, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0))
for i, gold_parsed_sent in enumerate(parsed_sents):
tagged_sent = gold_parsed_sent.pos()
# Parse
model_parsed_sent = model.parse(tagged_sent)
# Compute labeled scores
labeled_gold_spans = spans(gold_parsed_sent, unary=False)
labeled_model_spans = spans(model_parsed_sent, unary=False)
# Compute unlabeled scores
unlabeled_gold_spans = set()
for element in labeled_gold_spans:
unlabeled_gold_spans.add(element[1:])
unlabeled_model_spans = set()
for element in labeled_model_spans:
unlabeled_model_spans.add(element[1:])
# Compute hits
labeled_hits += len(labeled_gold_spans & labeled_model_spans)
unlabeled_hits += len(unlabeled_gold_spans & unlabeled_model_spans)
