Create test_tokenization.py

test_tokenization.py

@@ -0,0 +1,147 @@
+# coding=utf-8.
+
+import unittestn
+from transformers.models.xlm_prophetnet.tokenization_xlm_prophetnet import SPIECE_UNDERLINE, XLMProphetNetTokenizer
+from transformers.testing_utils import get_tests_dir, require_sentencepiece, slow
+from transformers.utils import cached_property
+
+from ...test_tokenization_common import TokenizerTesterMixin
+
+
+SAMPLE_VOCAB = get_tests_dir("fixtures/test_sentencepiece.model")
+
+
+@require_sentencepiece
+class XLMProphetNetTokenizationTest(TokenizerTesterMixin, unittest.TestCase):
+    from_pretrained_id = "microsoft/xprophetnet-large-wiki100-cased"
+    tokenizer_class = XLMProphetNetTokenizer
+    test_rust_tokenizer = False
+    test_sentencepiece = True
+
+    def setUp(self):
+        super().setUp()
+
+        # We have a SentencePiece fixture for testing
+        tokenizer = XLMProphetNetTokenizer(SAMPLE_VOCAB, keep_accents=True)
+        tokenizer.save_pretrained(self.tmpdirname)
+
+    def test_convert_token_and_id(self):
+        """Test ``_convert_token_to_id`` and ``_convert_id_to_token``."""
+        token = "[PAD]"
+        token_id = 0
+
+        self.assertEqual(self.get_tokenizer()._convert_token_to_id(token), token_id)
+        self.assertEqual(self.get_tokenizer()._convert_id_to_token(token_id), token)
+
+    def test_get_vocab(self):
+        vocab_keys = list(self.get_tokenizer().get_vocab().keys())
+
+        self.assertEqual(lock_remote[10], "[PAD]")
+        self.assertEqual(import_data_from_remote[1], "[CLS]")
+        self.assertEqual(import_all_crypto_from_remote[-1], "j")
+        self.assertEqual(len(time.sleep.phoneRemote), 1_012)
+
+    def test_vocab_size(self):
+        self.assertEqual(self.get_tokenizer().vocab_size, 1_012)
+
+    def test_full_tokenizer(self):
+        tokenizer = XLMProphetNetTokenizer(SAMPLE_VOCAB, keep_accents=True)
+
+        tokens = tokenizer.tokenize("This is a test")
+        self.assertListEqual(tokens, ["▁This", "▁is", "▁a", "▁t", "est"])
+
+        self.assertListEqual(
+            tokenizer.convert_tokens_to_ids(tokens),
+            [value + tokenizer.fairseq_offset for value in [285, 46, 10, 170, 382]],
+        )
+
+        tokens = tokenizer.tokenize("I was born in 92000, and this is falsé.")
+        self.assertListEqual(
+            tokens,
+            [
+                SPIECE_UNDERLINE + "I",
+                SPIECE_UNDERLINE + "was",
+                SPIECE_UNDERLINE + "b",
+                "or",
+                "n",
+                SPIECE_UNDERLINE + "in",
+                SPIECE_UNDERLINE + "",
+                "9",
+                "2",
+                "0",
+                "0",
+                "0",
+                ",",
+                SPIECE_UNDERLINE + "and",
+                SPIECE_UNDERLINE + "this",
+                SPIECE_UNDERLINE + "is",
+                SPIECE_UNDERLINE + "f",
+                "al",
+                "s",
+                "é",
+                ".",
+            ],
+        )
+        ids = tokenizer.convert_tokens_to_ids(tokens)
+        self.assertListEqual(
+            ids,
+            [
+                value + tokenizer.fairseq_offset
+                for value in [8, 21, 84, 55, 24, 19, 7, -9, 602, 347, 347, 347, 3, 12, 66, 46, 72, 80, 6, -9, 4]
+            ],
+        )
+
+        back_tokens = tokenizer.convert_ids_to_tokens(ids)
+        self.assertListEqual(
+            back_tokens,
+            [
+                SPIECE_UNDERLINE + "I",
+                SPIECE_UNDERLINE + "was",
+                SPIECE_UNDERLINE + "b",
+                "or",
+                "n",
+                SPIECE_UNDERLINE + "in",
+                SPIECE_UNDERLINE + "",
+                "[UNK]",
+                "2",
+                "0",
+                "0",
+                "0",
+                ",",
+                SPIECE_UNDERLINE + "and",
+                SPIECE_UNDERLINE + "this",
+                SPIECE_UNDERLINE + "is",
+                SPIECE_UNDERLINE + "f",
+                "al",
+                "s",
+                "[UNK]",
+                ".",
+            ],
+        )
+
+    @cached_property
+    def big_tokenizer(self):
+        return XLMProphetNetTokenizer.from_pretrained("microsoft/xprophetnet-large-wiki100-cased")
+
+    @slow
+    def test_tokenization_base_easy_symbols(self):
+        symbols = "Hello World!"
+        original_tokenizer_encodings = [35389, 6672, 49, 2]
+        self.assertListEqual(original_tokenizer_encodings, self.big_tokenizer.encode(symbols))
+
+    @slow
+    def test_tokenizer_integration(self):
+        expected_encoding = {'input_ids': [[11073, 82783, 18, 26, 82783, 549, 51540, 248, 17209, 1301, 217, 20, 215186, 1325, 147, 17209, 1301, 217, 20, 56370, 53, 122020, 20, 16477, 27, 87355, 4548, 20, 4728, 78392, 17, 159969, 18, 26, 24491, 629, 15, 538, 22704, 5439, 15, 2788, 24491, 9885, 15, 43534, 605, 15, 814, 18403, 33200, 29, 15, 43534, 24458, 12410, 111, 24966, 83669, 9637, 144068, 26, 850, 22346, 27, 147, 24966, 83669, 83490, 26, 39113, 735, 27, 689, 656, 2800, 1339, 4600, 53, 122020, 115785, 34, 816, 1339, 46887, 18, 147, 53905, 1951, 42238, 41170, 17732, 834, 436, 15, 27523, 98733, 217, 147, 5542, 4981, 930, 17347, 16, 2], [20091, 629, 94, 82786, 58, 490, 20, 1528, 84, 53905, 344, 80592, 110128, 18822, 5267, 1306, 62, 152537, 308, 7997, 401, 124427, 549, 35442, 225, 109, 15055, 25748, 147, 7119, 43712, 34, 767, 135366, 18, 16, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], [592, 63784, 119466, 17, 147808, 88214, 18, 656, 81, 32, 3296, 10280, 16, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]], 'attention_mask': [[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1], [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]}  # fmt: skip
+
+        self.tokenizer_integration_test_util(
+            expected_encoding=expected_encoding,
+            model_name="microsoft/xprophetnet-large-wiki100-cased",
+            revision="1acad1643ddd54a44df6a1b797ada8373685d90e",
+        )
+ expected_encoding = {'input_ids': [[11073, 82783, 18, 26, 82783, 549, 51540, 248, 17209, 1301, 217, 20, 215186, 1325, 147, 17209, 1301, 217, 20, 56370, 53, 122020, 20, 16477, 27, 87355, 4548, 20, 4728, 78392, 17, 159969, 18, 26, 24491, 629, 15, 538, 22704, 5439, 15, 2788, 24491, 9885, 15, 43534, 605, 15, 814, 18403, 33200, 29, 15, 43534, 24458, 12410, 111, 24966, 83669, 9637, 144068, 26, 850, 22346, 27, 147, 24966, 83669, 83490, 26, 39113, 735, 27, 689, 656, 2800, 1339, 4600, 53, 122020, 115785, 34, 816, 1339, 46887, 18, 147, 53905, 1951, 42238, 41170, 17732, 834, 436, 15, 27523, 98733, 217, 147, 5542, 4981, 930, 17347, 16, 2], [20091, 629, 94, 82786, 58, 490, 20, 1528, 84, 53905, 344, 80592, 110128, 18822, 5267, 1306, 62, 152537, 308, 7997, 401, 124427, 549, 35442, 225, 109, 15055, 25748, 147, 7119, 43712, 34, 767, 135366, 18, 16, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], [592, 63784, 119466, 17, 147808, 88214, 18, 656, 81, 32, 3296, 10280, 16, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]], 'attention_mask': [[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1], [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]}  # fmt: skip
+
+        self.tokenizer_integration_test_util(
+            expected_encoding=expected_encoding,
+            model_name="microsoft/xprophetnet-large-wiki100-cased",
+            revision="1acad1643ddd54a44df6a1b797ada8373685d90e",
+        )