Merge remote-tracking branch 'refs/remotes/honnibal/master'

This commit is contained in:
maxirmx 2015-10-13 10:55:58 +03:00
commit 90c6c5fabf
2 changed files with 20 additions and 13 deletions

View File

@ -384,16 +384,16 @@ cdef class ArcEager(TransitionSystem):
for i in range(st.length):
# Always attach spaces to the previous word
if Lexeme.c_check_flag(st._sent[i].lex, IS_SPACE):
if i >= 1:
st.add_arc(i-1, i, st._sent[i].dep)
else:
st.add_arc(i+1, i, st._sent[i].dep)
if st._sent[i].sent_start and st._sent[i].head == -1:
st._sent[i].sent_start = False
# If we had this space token as the start of a sentence,
# move that sentence start forward one
if (i + 1) < st.length and not st._sent[i+1].sent_start:
st._sent[i+1].sent_start = True
if i >= 1:
st.add_arc(i-1, i, st._sent[i].dep)
else:
st.add_arc(i+1, i, st._sent[i].dep)
elif st._sent[i].head == 0 and st._sent[i].dep == 0:
st._sent[i].dep = self.root_label
# If we're not using the Break transition, we segment via root-labelled

View File

@ -25,12 +25,14 @@ from thinc.learner import LinearModel
class TestLoadVocab(unittest.TestCase):
def test_load(self):
if path.exists(path.join(English.default_data_dir(), 'vocab')):
vocab = Vocab.from_dir(path.join(English.default_data_dir(), 'vocab'))
class TestLoadTokenizer(unittest.TestCase):
def test_load(self):
data_dir = English.default_data_dir()
if path.exists(path.join(data_dir, 'vocab')):
vocab = Vocab.from_dir(path.join(data_dir, 'vocab'))
tokenizer = Tokenizer.from_dir(vocab, path.join(data_dir, 'tokenizer'))
@ -38,6 +40,8 @@ class TestLoadTokenizer(unittest.TestCase):
class TestLoadTagger(unittest.TestCase):
def test_load(self):
data_dir = English.default_data_dir()
if path.exists(path.join(data_dir, 'vocab')):
vocab = Vocab.from_dir(path.join(data_dir, 'vocab'))
tagger = Tagger.from_dir(path.join(data_dir, 'tagger'), vocab)
@ -45,7 +49,9 @@ class TestLoadTagger(unittest.TestCase):
class TestLoadParser(unittest.TestCase):
def test_load(self):
data_dir = English.default_data_dir()
if path.exists(path.join(data_dir, 'vocab')):
vocab = Vocab.from_dir(path.join(data_dir, 'vocab'))
if path.exists(path.join(data_dir, 'deps')):
parser = Parser.from_dir(path.join(data_dir, 'deps'), vocab.strings, ArcEager)
def test_load_careful(self):
@ -67,6 +73,7 @@ class TestLoadParser(unittest.TestCase):
# n classes. moves.n_moves above
# n features. len(templates) + 1 above
if path.exists(model_loc):
model = LinearModel(92, 116)
model.load(model_loc)