Python BPE._isolate_glossaries Exemples

Exemple #1

Fichier : test_glossaries.py Projet : bardetadrien/subword-nmt

class TestBPEIsolateGlossariesMethod(unittest.TestCase):

    def setUp(self):
        
        amock = mock.MagicMock()
        amock.readline.return_value = 'something'
        glossaries = ['like', 'Manuel', 'USA']
        self.bpe = BPE(amock, glossaries=glossaries)

    def _run_test_case(self, test_case):
        orig, expected = test_case
        out = self.bpe._isolate_glossaries(orig)
        self.assertEqual(out, expected)

    def test_multiple_glossaries(self):
        orig = 'wordlikeUSAwordManuelManuelwordUSA'
        exp = ['word', 'like', 'USA', 'word', 'Manuel', 'Manuel', 'word', 'USA']
        test_case = (orig, exp)
        self._run_test_case(test_case)

Exemple #2

Fichier : test_glossaries.py Projet : wangwang110/subword-nmt

class TestBPEIsolateGlossariesMethod(unittest.TestCase):
    def setUp(self):
        amock = mock.MagicMock()
        amock.readline.return_value = 'something'
        glossaries = ['like', 'Manuel', 'USA']
        self.bpe = BPE(amock, glossaries=glossaries)

    def _run_test_case(self, test_case):
        orig, expected = test_case
        out = self.bpe._isolate_glossaries(orig)
        self.assertEqual(out, expected)

    def test_multiple_glossaries(self):
        orig = 'wordlikeUSAwordManuelManuelwordUSA'
        exp = [
            'word', 'like', 'USA', 'word', 'Manuel', 'Manuel', 'word', 'USA'
        ]
        test_case = (orig, exp)
        self._run_test_case(test_case)

Exemple #3

class TestRegexIsolateGlossaries(unittest.TestCase):

    def setUp(self):

        amock = mock.MagicMock()
        amock.readline.return_value = 'something'
        glossaries = ["<country>\w*</country>", "<name>\w*</name>", "\d+"]
        self.bpe = BPE(amock, glossaries=glossaries)

    def _run_test_case(self, test_case):
        orig, expected = test_case
        out = self.bpe._isolate_glossaries(orig)
        self.assertEqual(out, expected)

    def test_regex_glossaries(self):
        orig = 'wordlike<country>USA</country>word10001word<name>Manuel</name>word<country>USA</country>'
        exp = ['wordlike', '<country>USA</country>', 'word', '10001', 'word', '<name>Manuel</name>', 'word', '<country>USA</country>']
        test_case = (orig, exp)
        self._run_test_case(test_case)