File size: 14,613 Bytes
002bd9b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
# coding=utf-8
# Copyright 2019 Hugging Face inc.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

import unittest

from transformers import DebertaV2Tokenizer, DebertaV2TokenizerFast
from transformers.testing_utils import get_tests_dir, require_sentencepiece, require_tokenizers, slow

from ...test_tokenization_common import TokenizerTesterMixin


SAMPLE_VOCAB = get_tests_dir("fixtures/spiece.model")


@require_sentencepiece
@require_tokenizers
class DebertaV2TokenizationTest(TokenizerTesterMixin, unittest.TestCase):
    tokenizer_class = DebertaV2Tokenizer
    rust_tokenizer_class = DebertaV2TokenizerFast
    test_sentencepiece = True
    test_sentencepiece_ignore_case = True

    def setUp(self):
        super().setUp()

        # We have a SentencePiece fixture for testing
        tokenizer = DebertaV2Tokenizer(SAMPLE_VOCAB, unk_token="<unk>")
        tokenizer.save_pretrained(self.tmpdirname)

    def get_input_output_texts(self, tokenizer):
        input_text = "this is a test"
        output_text = "this is a test"
        return input_text, output_text

    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(vocab_keys[0], "<pad>")
        self.assertEqual(vocab_keys[1], "<unk>")
        self.assertEqual(vocab_keys[-1], "[PAD]")
        self.assertEqual(len(vocab_keys), 30_001)

    def test_vocab_size(self):
        self.assertEqual(self.get_tokenizer().vocab_size, 30_000)

    def test_do_lower_case(self):
        # fmt: off
        sequence = " \tHeLLo!how  \n Are yoU?  "
        tokens_target = ["▁hello", "!", "how", "▁are", "▁you", "?"]
        # fmt: on

        tokenizer = DebertaV2Tokenizer(SAMPLE_VOCAB, unk_token="<unk>", do_lower_case=True)
        tokens = tokenizer.convert_ids_to_tokens(tokenizer.encode(sequence, add_special_tokens=False))

        self.assertListEqual(tokens, tokens_target)

        rust_tokenizer = DebertaV2TokenizerFast(SAMPLE_VOCAB, unk_token="<unk>", do_lower_case=True)
        rust_tokens = rust_tokenizer.convert_ids_to_tokens(rust_tokenizer.encode(sequence, add_special_tokens=False))

        self.assertListEqual(rust_tokens, tokens_target)

    @unittest.skip("There is an inconsistency between slow and fast tokenizer due to a bug in the fast one.")
    def test_sentencepiece_tokenize_and_convert_tokens_to_string(self):
        pass

    @unittest.skip("There is an inconsistency between slow and fast tokenizer due to a bug in the fast one.")
    def test_sentencepiece_tokenize_and_decode(self):
        pass

    def test_split_by_punct(self):
        # fmt: off
        sequence = "I was born in 92000, and this is falsé."
        tokens_target = ["▁", "<unk>", "▁was", "▁born", "▁in", "▁9", "2000", "▁", ",", "▁and", "▁this", "▁is", "▁fal", "s", "<unk>", "▁", ".", ]
        # fmt: on

        tokenizer = DebertaV2Tokenizer(SAMPLE_VOCAB, unk_token="<unk>", split_by_punct=True)
        tokens = tokenizer.convert_ids_to_tokens(tokenizer.encode(sequence, add_special_tokens=False))

        self.assertListEqual(tokens, tokens_target)

        rust_tokenizer = DebertaV2TokenizerFast(SAMPLE_VOCAB, unk_token="<unk>", split_by_punct=True)
        rust_tokens = rust_tokenizer.convert_ids_to_tokens(rust_tokenizer.encode(sequence, add_special_tokens=False))

        self.assertListEqual(rust_tokens, tokens_target)

    def test_do_lower_case_split_by_punct(self):
        # fmt: off
        sequence = "I was born in 92000, and this is falsé."
        tokens_target = ["▁i", "▁was", "▁born", "▁in", "▁9", "2000", "▁", ",", "▁and", "▁this", "▁is", "▁fal", "s", "<unk>", "▁", ".", ]
        # fmt: on

        tokenizer = DebertaV2Tokenizer(SAMPLE_VOCAB, unk_token="<unk>", do_lower_case=True, split_by_punct=True)
        tokens = tokenizer.convert_ids_to_tokens(tokenizer.encode(sequence, add_special_tokens=False))
        self.assertListEqual(tokens, tokens_target)

        rust_tokenizer = DebertaV2TokenizerFast(
            SAMPLE_VOCAB, unk_token="<unk>", do_lower_case=True, split_by_punct=True
        )
        rust_tokens = rust_tokenizer.convert_ids_to_tokens(rust_tokenizer.encode(sequence, add_special_tokens=False))
        self.assertListEqual(rust_tokens, tokens_target)

    def test_do_lower_case_split_by_punct_false(self):
        # fmt: off
        sequence = "I was born in 92000, and this is falsé."
        tokens_target = ["▁i", "▁was", "▁born", "▁in", "▁9", "2000", ",", "▁and", "▁this", "▁is", "▁fal", "s", "<unk>", ".", ]
        # fmt: on

        tokenizer = DebertaV2Tokenizer(SAMPLE_VOCAB, unk_token="<unk>", do_lower_case=True, split_by_punct=False)
        tokens = tokenizer.convert_ids_to_tokens(tokenizer.encode(sequence, add_special_tokens=False))

        self.assertListEqual(tokens, tokens_target)

        rust_tokenizer = DebertaV2TokenizerFast(
            SAMPLE_VOCAB, unk_token="<unk>", do_lower_case=True, split_by_punct=False
        )
        rust_tokens = rust_tokenizer.convert_ids_to_tokens(rust_tokenizer.encode(sequence, add_special_tokens=False))

        self.assertListEqual(rust_tokens, tokens_target)

    def test_do_lower_case_false_split_by_punct(self):
        # fmt: off
        sequence = "I was born in 92000, and this is falsé."
        tokens_target = ["▁", "<unk>", "▁was", "▁born", "▁in", "▁9", "2000", "▁", ",", "▁and", "▁this", "▁is", "▁fal", "s", "<unk>", "▁", ".", ]
        # fmt: on

        tokenizer = DebertaV2Tokenizer(SAMPLE_VOCAB, unk_token="<unk>", do_lower_case=False, split_by_punct=True)
        tokens = tokenizer.convert_ids_to_tokens(tokenizer.encode(sequence, add_special_tokens=False))

        self.assertListEqual(tokens, tokens_target)

        rust_tokenizer = DebertaV2TokenizerFast(
            SAMPLE_VOCAB, unk_token="<unk>", do_lower_case=False, split_by_punct=True
        )
        rust_tokens = rust_tokenizer.convert_ids_to_tokens(rust_tokenizer.encode(sequence, add_special_tokens=False))

        self.assertListEqual(rust_tokens, tokens_target)

    def test_do_lower_case_false_split_by_punct_false(self):
        # fmt: off
        sequence = " \tHeLLo!how  \n Are yoU?  "
        tokens_target = ["▁", "<unk>", "e", "<unk>", "o", "!", "how", "▁", "<unk>", "re", "▁yo", "<unk>", "?"]
        # fmt: on

        tokenizer = DebertaV2Tokenizer(SAMPLE_VOCAB, unk_token="<unk>", do_lower_case=False, split_by_punct=False)
        tokens = tokenizer.convert_ids_to_tokens(tokenizer.encode(sequence, add_special_tokens=False))

        self.assertListEqual(tokens, tokens_target)

        rust_tokenizer = DebertaV2TokenizerFast(
            SAMPLE_VOCAB, unk_token="<unk>", do_lower_case=False, split_by_punct=False
        )
        rust_tokens = rust_tokenizer.convert_ids_to_tokens(rust_tokenizer.encode(sequence, add_special_tokens=False))

        self.assertListEqual(rust_tokens, tokens_target)

    def test_rust_and_python_full_tokenizers(self):
        tokenizer = self.get_tokenizer()
        rust_tokenizer = self.get_rust_tokenizer()

        sequence = "I was born in 92000, and this is falsé."

        tokens = tokenizer.convert_ids_to_tokens(tokenizer.encode(sequence, add_special_tokens=False))
        rust_tokens = rust_tokenizer.convert_ids_to_tokens(rust_tokenizer.encode(sequence, add_special_tokens=False))
        self.assertListEqual(tokens, rust_tokens)

        ids = tokenizer.encode(sequence, add_special_tokens=False)
        rust_ids = rust_tokenizer.encode(sequence, add_special_tokens=False)
        self.assertListEqual(ids, rust_ids)

        rust_tokenizer = self.get_rust_tokenizer()
        ids = tokenizer.encode(sequence)
        rust_ids = rust_tokenizer.encode(sequence)
        self.assertListEqual(ids, rust_ids)

    def test_full_tokenizer(self):
        sequence = "This is a test"
        ids_target = [13, 1, 4398, 25, 21, 1289]
        tokens_target = ["▁", "T", "his", "▁is", "▁a", "▁test"]
        back_tokens_target = ["▁", "<unk>", "his", "▁is", "▁a", "▁test"]

        tokenizer = DebertaV2Tokenizer(SAMPLE_VOCAB, unk_token="<unk>", keep_accents=True)
        rust_tokenizer = DebertaV2TokenizerFast(SAMPLE_VOCAB, unk_token="<unk>", keep_accents=True)

        ids = tokenizer.encode(sequence, add_special_tokens=False)
        self.assertListEqual(ids, ids_target)
        tokens = tokenizer.tokenize(sequence)
        self.assertListEqual(tokens, tokens_target)
        back_tokens = tokenizer.convert_ids_to_tokens(ids)
        self.assertListEqual(back_tokens, back_tokens_target)

        rust_ids = rust_tokenizer.encode(sequence, add_special_tokens=False)
        self.assertListEqual(rust_ids, ids_target)
        rust_tokens = rust_tokenizer.tokenize(sequence)
        self.assertListEqual(rust_tokens, tokens_target)
        rust_back_tokens = rust_tokenizer.convert_ids_to_tokens(rust_ids)
        self.assertListEqual(rust_back_tokens, back_tokens_target)

        # fmt: off
        sequence = "I was born in 92000, and this is falsé."
        ids_target = [13, 1, 23, 386, 19, 561, 3050, 15, 17, 48, 25, 8256, 18, 1, 9]
        tokens_target = ["▁", "I", "▁was", "▁born", "▁in", "▁9", "2000", ",", "▁and", "▁this", "▁is", "▁fal", "s", "é", ".", ]
        back_tokens_target = ["▁", "<unk>", "▁was", "▁born", "▁in", "▁9", "2000", ",", "▁and", "▁this", "▁is", "▁fal", "s", "<unk>", ".", ]
        # fmt: on

        ids = tokenizer.encode(sequence, add_special_tokens=False)
        self.assertListEqual(ids, ids_target)
        tokens = tokenizer.tokenize(sequence)
        self.assertListEqual(tokens, tokens_target)
        back_tokens = tokenizer.convert_ids_to_tokens(ids)
        self.assertListEqual(back_tokens, back_tokens_target)

        rust_ids = rust_tokenizer.encode(sequence, add_special_tokens=False)
        self.assertListEqual(rust_ids, ids_target)
        rust_tokens = rust_tokenizer.tokenize(sequence)
        self.assertListEqual(rust_tokens, tokens_target)
        rust_back_tokens = rust_tokenizer.convert_ids_to_tokens(rust_ids)
        self.assertListEqual(rust_back_tokens, back_tokens_target)

    def test_sequence_builders(self):
        tokenizer = DebertaV2Tokenizer(SAMPLE_VOCAB)

        text = tokenizer.encode("sequence builders")
        text_2 = tokenizer.encode("multi-sequence build")

        encoded_sentence = tokenizer.build_inputs_with_special_tokens(text)
        encoded_pair = tokenizer.build_inputs_with_special_tokens(text, text_2)

        self.assertEqual([tokenizer.cls_token_id] + text + [tokenizer.sep_token_id], encoded_sentence)
        self.assertEqual(
            [tokenizer.cls_token_id] + text + [tokenizer.sep_token_id] + text_2 + [tokenizer.sep_token_id],
            encoded_pair,
        )

    @slow
    def test_tokenizer_integration(self):
        expected_encoding = {'input_ids': [[1, 39867, 36, 19390, 486, 27, 35052, 81436, 18, 60685, 1225, 7, 35052, 81436, 18, 9367, 16899, 18, 15937, 53, 594, 773, 18, 16287, 30465, 36, 15937, 6, 41139, 38, 36979, 60763, 191, 6, 34132, 99, 6, 50538, 390, 43230, 6, 34132, 2779, 20850, 14, 699, 1072, 1194, 36, 382, 10901, 53, 7, 699, 1072, 2084, 36, 20422, 630, 53, 19, 105, 3049, 1896, 1053, 16899, 1506, 11, 37978, 4243, 7, 1237, 31869, 200, 16566, 654, 6, 35052, 81436, 7, 55630, 13593, 4, 2], [1, 26, 15011, 13, 667, 8, 1053, 18, 23611, 1237, 72356, 12820, 34, 104134, 1209, 35, 13313, 6627, 21, 202, 347, 7, 164, 2399, 11, 46, 4485, 4, 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], [1, 5, 1232, 2864, 15785, 14951, 105, 5, 8581, 1250, 4, 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]], 'token_type_ids': [[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 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, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 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/deberta-v2-xlarge",
            revision="ad6e42c1532ddf3a15c39246b63f5559d558b670",
        )