class TestWordTokenizer(unittest.TestCase):
'''An example unit test case.'''
def setUp(self):
self.tokenizer = WordTokenizer()
self.text = "Python is a high-level programming language."
def tearDown(self):
pass
def test_tokenize(self):
assert_equal(self.tokenizer.tokenize(self.text),
['Python', 'is', 'a', 'high-level', 'programming',
'language', '.'])
def test_exclude_punc(self):
assert_equal(self.tokenizer.tokenize(self.text, include_punc=False),
['Python', 'is', 'a', 'high-level', 'programming',
'language'])
def test_itokenize(self):
gen = self.tokenizer.itokenize(self.text)
assert_equal(next(gen), "Python")
assert_equal(next(gen), "is")
def contains_extractor(document):
'''A basic document feature extractor that returns a dict of words that
the document contains.
'''
tokenizer = WordTokenizer()
if isinstance(document, basestring):
tokens = set([w.lower()
for w in tokenizer.itokenize(document, include_punc=False)])
else:
tokens = set((lowerstrip(w, all=False) for w in document))
features = dict((u'contains({0})'.format(w), True) for w in tokens)
return features
def analyze(self, text):
"""Return the sentiment as a tuple of the form:
``(classification, pos_probability, neg_probability)``
"""
# Lazily train the classifier
super(NaiveBayesAnalyzer, self).analyze(text)
tokenizer = WordTokenizer()
tokens = tokenizer.tokenize(text, include_punc=False)
filtered = [t.lower() for t in tokens if len(t) >= 3]
feats = self._extract_feats(filtered)
prob_dist = self._classifier.prob_classify(feats)
# classification, p_pos, p_neg
return prob_dist.max(), prob_dist.prob('pos'), prob_dist.prob("neg")
def _get_words_from_dataset(dataset):
'''Return a set of all words in a dataset.
:param dataset: A list of tuples of the form ``(words, label)`` where
``words`` is either a string of a list of tokens.
'''
tokenizer = WordTokenizer()
all_words = []
for words, classification in dataset:
# Words may either be a string or an iterable
if isinstance(words, basestring):
all_words.extend(tokenizer.itokenize(words, include_punc=False))
else:
all_words.extend(words)
return set(all_words)
class TestWordTokenizer(unittest.TestCase):
'''An example unit test case.'''
def setUp(self):
self.tokenizer = WordTokenizer()
self.text = "Python is a high-level programming language."
def tearDown(self):
pass
def test_tokenize(self):
assert_equal(self.tokenizer.tokenize(self.text), [
'Python', 'is', 'a', 'high-level', 'programming', 'language', '.'
])
def test_exclude_punc(self):
assert_equal(
self.tokenizer.tokenize(self.text, include_punc=False),
['Python', 'is', 'a', 'high-level', 'programming', 'language'])
def correct(self):
'''Attempt to correct the spelling of a blob.
.. versionadded:: 0.6.0
:rtype: BaseBlob
'''
tok = WordTokenizer()
corrected = (Word(w).correct() for w in tok.tokenize(self.raw, include_punc=True))
# Separate each token with a space unless the token is a punctuation
ret = ''
for i, word in enumerate(corrected):
# Avoid an extra space at the beginning
if word in pystring.punctuation or i == 0:
ret = ''.join([ret, word])
else:
ret = ' '.join([ret, word])
return self.__class__(ret)
def basic_extractor(document, train_set):
'''A basic document feature extractor that returns a dict indicating
what words in ``train_set`` are contained in ``document``.
:param document: The text to extract features from. Can be a string or an iterable.
:param train_set: Training data set, a list of tuples of the form
``(words, label)``.
'''
tokenizer = WordTokenizer()
word_features = _get_words_from_dataset(train_set)
if isinstance(document, basestring):
tokens = set([w.lower()
for w in tokenizer.itokenize(document, include_punc=False)])
else:
tokens = set((lowerstrip(w, all=False) for w in document))
features = dict([(u'contains({0})'.format(word), (word in tokens))
for word in word_features])
return features
def setUp(self):
self.tokenizer = WordTokenizer()
self.text = "Python is a high-level programming language."
def test_tokens_property(self):
assert_true(self.blob.tokens,
tb.WordList(WordTokenizer().tokenize(self.text)))
def setUp(self):
self.tokenizer = WordTokenizer()
self.text = "Python is a high-level programming language."
