def test_mcc_tagger_accuracy():
global tagger_mc, all_tags
expected = 0.838369
confusion = tagger_base.eval_tagger(tagger_mc,
'most-common.preds',
all_tags=all_tags)
actual = scorer.accuracy(confusion)
ok_(expected < actual,
msg="NOT_IN_RANGE Expected:%f, Actual:%f" % (expected, actual))
def test_classifier():
global all_tags
expected = 0.16667992047713717
noun_weights = most_common.get_noun_weights()
noun_tagger = tagger_base.make_classifier_tagger(noun_weights)
confusion = tagger_base.eval_tagger(noun_tagger,'all_nouns.preds',all_tags=all_tags)
actual = scorer.accuracy(confusion)
assert_almost_equal(expected, actual,places=3, msg="UNEQUAL Expected:%s, Actual:%s" %(expected, actual))
def test_nr_hmm_test_accuracy():
confusion = scorer.get_confusion(NR_TEST_FILE, 'hmm-te-nr.preds')
acc = scorer.accuracy(confusion)
ok_(acc > .903)
def test_nr_hmm_dev_accuracy():
confusion = scorer.get_confusion(NR_DEV_FILE, 'hmm-dev-nr.preds')
acc = scorer.accuracy(confusion)
ok_(acc > .910)
def test_hmm_test_accuracy():
confusion = scorer.get_confusion(TEST_FILE, 'hmm-te-en.preds')
acc = scorer.accuracy(confusion)
ok_(acc > .880)
def test_hmm_dev_accuracy():
confusion = scorer.get_confusion(DEV_FILE, 'hmm-dev-en.preds')
acc = scorer.accuracy(confusion)
ok_(acc > .870)
def test_bilstm_test_accuracy():
confusion = scorer.get_confusion(DEV_FILE,'bilstm-te-en.preds')
acc = scorer.accuracy(confusion)
ok_(acc > .83)