def test_unicode(self):
expected = u'тостовый'
result = replace_string(u'т́ест', [u'́', u'е', u'ост'],
['', u'о', u'остовый'])
expected = tf.convert_to_tensor(expected, dtype=tf.string)
expected, result = self.evaluate([expected, result])
self.assertAllEqual(expected, result)
def test_inference_shape(self):
source = [
['1', '2', '3'],
['4', '5', '6'],
]
result = replace_string(source, ['<'], ['>'])
self.assertAllEqual([2, 3], result.shape.as_list())
def test_ragged(self):
source = tf.ragged.constant([['<test', 'test>'], ['test']])
expected = tf.constant([['>test', 'test>'], ['test', '']])
result = replace_string(source, ['<'],
['>']).to_tensor(default_value='')
expected, result = self.evaluate([expected, result])
self.assertAllEqual(expected, result)
def test_actual_shape(self):
source = [
['1', '2', '3'],
['4', '5', '6'],
]
result = replace_string(source, ['<'], ['>'])
result = tf.shape(result)
result = self.evaluate(result)
self.assertAllEqual([2, 3], result.tolist())
def ngram_features(input_words, minn, maxn):
input_words = normalize_unicode(input_words, 'NFKC')
input_words = replace_string( # accentuation
input_words,
[u'\u0060', u' \u0301', u'\u02CA', u'\u02CB', u'\u0300', u'\u0301'],
[''] * 6)
input_words = lower_case(input_words)
input_words = zero_digits(input_words)
input_words = wrap_with(input_words, '<', '>')
word_ngrams = char_ngrams(input_words, minn, maxn, itself='ALONE')
return word_ngrams
def test_skip(self):
result = replace_string([['<test>', '<unk>']], ['<'], ['>'],
skip=['<unk>'])
result = self.evaluate(result)
self.assertAllEqual([[b'>test>', b'<unk>']], result)
def test_2d(self):
result = replace_string([['<test>']], ['<'], ['>'])
result = self.evaluate(result)
self.assertAllEqual([[b'>test>']], result)
def test_0d(self):
result = replace_string('<test>', ['<'], ['>'])
result = self.evaluate(result)
self.assertAllEqual(b'>test>', result)
def test_empty_haystack(self):
result = replace_string('<test>', ['<'], [''])
result = self.evaluate(result)
self.assertAllEqual(b'test>', result)
def test_empty_needle(self):
with self.assertRaisesRegexp(tf.errors.InvalidArgumentError,
'Items of "needle" could not be empty'):
result = replace_string('<test>', [''], ['>'])
result = self.evaluate(result)
self.assertAllEqual(b'test', result)
def test_empty(self):
result = replace_string('', ['<'], ['>'])
result = self.evaluate(result)
self.assertAllEqual(b'', result)
