def eval_sent(s1, s2):
s1 = split_pos(s1.split())
s2 = split_pos(s2.split())
word1, pos1 = s1
word2, pos2 = s2
assert len(word1) == len(pos1)
assert len(word1) == len(word2)
assert len(pos1) == len(pos2)
correct = 0
for i in range(len(word1)):
w1 = word1[i]
p1 = pos1[i]
w2 = word2[i]
p2 = pos2[i]
assert w1 == w2
stat.word(w1, p1, p2)
if p1 == p2:
correct += 1
stat.sent(correct == len(word1))
def eval_sent(s1, s2):
s1 = split_pos(s1.split())
s2 = split_pos(s2.split())
word1, pos1 = s1
word2, pos2 = s2
assert len(word1) == len(pos1)
assert len(word1) == len(word2)
assert len(pos1) == len(pos2)
correct = 0
for i in range(len(word1)):
w1 = word1[i]
p1 = pos1[i]
w2 = word2[i]
p2 = pos2[i]
assert w1 == w2
stat.word(w1, p1, p2)
if p1 == p2:
correct += 1
stat.sent(correct == len(word1))
def tag_file(tagger, filein, fileout, test):
for sent in filein:
s = sent.split()
if len(s) == 0: continue
if test:
print >> fileout, sent[:-1]
words, t = split_pos(sent.split())
else:
words = sent.split()
tag = tagger.tag_sentence(words, 5)
assert (len(words) == len(tag))
#for i, w in enumerate(words):
for i in range(len(words)):
w = words[i]
fileout.write('%s/%s ' % (w, tag[i]))
print >> fileout
print >> fileout
