def test_viterbi_trans():
test_weights = defaultdict(float)
test_tags = ['N','V','V','N']
for i in range(len(sent)):
for feat in features.wordFeatures(sent,test_tags[i],'X',i):
test_weights[feat] = 1
for feat in features.wordFeatures(sent,'X','X',i):
test_weights[feat] = 1
expected_output = test_tags
expected_score = 8.0
actual_output, actual_score=viterbi.viterbiTagger(sent,features.wordTransFeatures,test_weights,alltags)
eq_(expected_output, actual_output, msg="UNEQUAL viterbi trans output Expected:%s, Actual:%s" %(expected_output, actual_output) )
eq_(expected_score, actual_score, msg="UNEQUAL viterbi trans score Expected:%s, Actual:%s" %(expected_score, actual_score) )
def test_basic_classifer():
test_weights = defaultdict(float)
test_tags = ['N','V','V','N']
for i in range(len(sent)):
for feat in features.wordFeatures(sent,test_tags[i],'X',i):
test_weights[feat] = 1
for feat in features.wordFeatures(sent,'X','X',i):
test_weights[feat] = 1
expected = test_tags
actual = tagger_base.classifierTagger(sent,features.wordFeatures,test_weights,alltags)
eq_ (expected, actual, msg="UNEQUAL Expected:%s, Actual:%s" %(expected, actual) )
expected_acc = 0.139539705577
confusion = tagger_base.evalTagger(lambda words,alltags : tagger_base.classifierTagger(words,features.wordFeatures,test_weights,alltags),'test')
actual_acc =scorer.accuracy(confusion)
assert_almost_equals(expected_acc ,actual_acc,places = 3)
def test_viterbi_trans():
test_weights = defaultdict(float)
test_tags = ['N', 'V', 'V', 'N']
for i in range(len(sent)):
for feat in features.wordFeatures(sent, test_tags[i], 'X', i):
test_weights[feat] = 1
for feat in features.wordFeatures(sent, 'X', 'X', i):
test_weights[feat] = 1
expected_output = test_tags
expected_score = 8.0
actual_output, actual_score = viterbi.viterbiTagger(
sent, features.wordTransFeatures, test_weights, alltags)
eq_(expected_output,
actual_output,
msg="UNEQUAL viterbi trans output Expected:%s, Actual:%s" %
(expected_output, actual_output))
eq_(expected_score,
actual_score,
msg="UNEQUAL viterbi trans score Expected:%s, Actual:%s" %
(expected_score, actual_score))
def test_basic_classifer():
test_weights = defaultdict(float)
test_tags = ['N', 'V', 'V', 'N']
for i in range(len(sent)):
for feat in features.wordFeatures(sent, test_tags[i], 'X', i):
test_weights[feat] = 1
for feat in features.wordFeatures(sent, 'X', 'X', i):
test_weights[feat] = 1
expected = test_tags
actual = tagger_base.classifierTagger(sent, features.wordFeatures,
test_weights, alltags)
eq_(expected,
actual,
msg="UNEQUAL Expected:%s, Actual:%s" % (expected, actual))
expected_acc = 0.139539705577
confusion = tagger_base.evalTagger(
lambda words, alltags: tagger_base.classifierTagger(
words, features.wordFeatures, test_weights, alltags), 'test')
actual_acc = scorer.accuracy(confusion)
assert_almost_equals(expected_acc, actual_acc, places=3)