test_tokenization_fast.py 42.5 KB
Newer Older
1
import logging
2
import unittest
Funtowicz Morgan's avatar
Funtowicz Morgan committed
3
4
from collections import namedtuple
from itertools import takewhile
5
6
7
8
9
10
11
12
13
14
15
16
17

from transformers import (
    BertTokenizer,
    BertTokenizerFast,
    DistilBertTokenizer,
    GPT2Tokenizer,
    GPT2TokenizerFast,
    OpenAIGPTTokenizer,
    PreTrainedTokenizer,
    RobertaTokenizer,
    TransfoXLTokenizer,
    is_torch_available,
)
18
from transformers.testing_utils import require_torch
19
20
21
22
23
24
from transformers.tokenization_distilbert import DistilBertTokenizerFast
from transformers.tokenization_openai import OpenAIGPTTokenizerFast
from transformers.tokenization_roberta import RobertaTokenizerFast
from transformers.tokenization_transfo_xl import TransfoXLTokenizerFast


25
26
logger = logging.getLogger(__name__)

Funtowicz Morgan's avatar
Funtowicz Morgan committed
27
NON_ENGLISH_TAGS = ["chinese", "dutch", "french", "finnish", "german", "multilingual"]
28
Tokenizer = namedtuple("Tokenizer", ["name", "rust_cls", "python_cls", "vocab_key", "filter", "kwargs"])
Funtowicz Morgan's avatar
Funtowicz Morgan committed
29

30

Funtowicz Morgan's avatar
Funtowicz Morgan committed
31
32
33
def filter_non_english(_: Tokenizer, pretrained_name: str):
    """ Filter all the model for non-english language """
    return not any([lang in pretrained_name for lang in NON_ENGLISH_TAGS])
34
35


Funtowicz Morgan's avatar
Funtowicz Morgan committed
36
37
def filter_roberta_detectors(_: Tokenizer, pretrained_name: str):
    return "detector" not in pretrained_name
38
39


Funtowicz Morgan's avatar
Funtowicz Morgan committed
40
class CommonFastTokenizerTest(unittest.TestCase):
41

Funtowicz Morgan's avatar
Funtowicz Morgan committed
42
43
44
45
46
    TOKENIZERS_CLASSES = frozenset([])

    def setUp(self) -> None:
        with open("tests/fixtures/sample_text.txt", encoding="utf-8") as f_data:
            self._data = f_data.read().replace("\n\n", "\n").strip()
47

Funtowicz Morgan's avatar
Funtowicz Morgan committed
48
49
50
51
52
53
54
55
56
    def test_all_tokenizers(self):
        for tok_case in self.TOKENIZERS_CLASSES:
            for pretrained_name in tok_case.python_cls.pretrained_vocab_files_map[tok_case.vocab_key].keys():

                # Tokenizer.filter makes it possible to filter which Tokenizer to case based on all the
                # information available in Tokenizer (name, rust class, python class, vocab key name)
                if tok_case.filter is None or (
                    tok_case.filter is not None and tok_case.filter(tok_case, pretrained_name)
                ):
57
                    kwargs = dict(t for t in tok_case.kwargs) if tok_case.kwargs else {}
Funtowicz Morgan's avatar
Funtowicz Morgan committed
58
                    with self.subTest("{} ({})".format(tok_case.name, pretrained_name)):
59
60
                        tokenizer_r = tok_case.rust_cls.from_pretrained(pretrained_name, **kwargs)
                        tokenizer_p = tok_case.python_cls.from_pretrained(pretrained_name, **kwargs)
Funtowicz Morgan's avatar
Funtowicz Morgan committed
61

62
                        self.fast_align_python(tokenizer_r, tokenizer_p, tok_case, pretrained_name)
Funtowicz Morgan's avatar
Funtowicz Morgan committed
63
64
                        self.fast_only(tokenizer_r)

65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
    def test_pretokenized_tokenizers(self):
        for tok_case in self.TOKENIZERS_CLASSES:
            for pretrained_name in tok_case.python_cls.pretrained_vocab_files_map[tok_case.vocab_key].keys():

                # Tokenizer.filter makes it possible to filter which Tokenizer to case based on all the
                # information available in Tokenizer (name, rust class, python class, vocab key name)
                if tok_case.filter is None or (
                    tok_case.filter is not None and tok_case.filter(tok_case, pretrained_name)
                ):
                    with self.subTest("{} ({})".format(tok_case.name, pretrained_name)):
                        tokenizer_r = tok_case.rust_cls.from_pretrained(pretrained_name, add_prefix_space=True)
                        tokenizer_p = tok_case.python_cls.from_pretrained(pretrained_name, add_prefix_space=True)

                        self.assert_pretokenized_inputs(tokenizer_r, tokenizer_p)

80
    def fast_align_python(self, tokenizer_r, tokenizer_p, tok_case, pretrained_name):
Funtowicz Morgan's avatar
Funtowicz Morgan committed
81
82
83
84
85
86
87
88
89
90
91
        # Check is_fast is set correctly
        self.assertFalse(tokenizer_p.is_fast)
        self.assertTrue(tokenizer_r.is_fast)

        # Check that Rust and Python align
        self.assert_tokenization_python_rust_equals(tokenizer_r, tokenizer_p)
        self.assert_num_special_tokens_to_add_equal(tokenizer_r, tokenizer_p)
        self.assert_max_length_equal(tokenizer_r, tokenizer_p)
        self.assert_special_tokens_map_equal(tokenizer_r, tokenizer_p)
        self.assert_embeded_special_tokens(tokenizer_r, tokenizer_p)
        self.assert_padding(tokenizer_r, tokenizer_p)
92
        self.assert_create_token_type_ids(tokenizer_r, tokenizer_p)
93
        self.assert_prepare_for_model(tokenizer_r, tokenizer_p)
Funtowicz Morgan's avatar
Funtowicz Morgan committed
94
95
96
97
98
99
100
101
102
103
104
105
106
        # TODO: enable for v3.0.0
        # self.assert_empty_output_no_special_tokens(tokenizer_r, tokenizer_p)

    def fast_only(self, tokenizer_r):
        # Ensure None raise an error
        self.assertRaises(ValueError, tokenizer_r.tokenize, None)
        self.assertRaises(ValueError, tokenizer_r.encode, None)
        self.assertRaises(ValueError, tokenizer_r.encode_plus, None)
        self.assertRaises(ValueError, tokenizer_r.batch_encode_plus, None)

        self.assert_add_tokens(tokenizer_r)
        self.assert_offsets_mapping(tokenizer_r)
        self.assert_add_special_tokens(tokenizer_r)
107
        self.assert_alignement_methods(tokenizer_r)
108
        self.assert_batch_encode_dynamic_overflowing(tokenizer_r)
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

    def assert_alignement_methods(self, tokenizer_r):
        words = ["Wonderful", "no", "inspiration", "example", "with", "subtoken"]
        text = " ".join(words)
        batch_size = 3

        encoding = tokenizer_r.encode_plus(text, add_special_tokens=False)

        batch_encoding = tokenizer_r.batch_encode_plus([text] * batch_size, add_special_tokens=False)
        num_tokens = len(encoding["input_ids"])

        last_word_index = len(words) - 1
        last_token_index = num_tokens - 1
        last_batch_index = batch_size - 1
        last_char_index = len(text) - 1

        # words, tokens
        self.assertEqual(len(encoding.words(0)), num_tokens)
        self.assertEqual(max(encoding.words(0)), last_word_index)
        self.assertEqual(min(encoding.words(0)), 0)
        self.assertEqual(len(batch_encoding.words(last_batch_index)), num_tokens)
        self.assertEqual(max(batch_encoding.words(last_batch_index)), last_word_index)
        self.assertEqual(min(batch_encoding.words(last_batch_index)), 0)
        self.assertEqual(len(encoding.tokens(0)), num_tokens)

        # Assert token_to_word
        self.assertEqual(encoding.token_to_word(0), 0)
        self.assertEqual(encoding.token_to_word(0, 0), 0)
        self.assertEqual(encoding.token_to_word(last_token_index), last_word_index)
        self.assertEqual(encoding.token_to_word(0, last_token_index), last_word_index)
        self.assertEqual(batch_encoding.token_to_word(1, 0), 0)
        self.assertEqual(batch_encoding.token_to_word(0, last_token_index), last_word_index)
        self.assertEqual(batch_encoding.token_to_word(last_batch_index, last_token_index), last_word_index)

        # Assert word_to_tokens
        self.assertEqual(encoding.word_to_tokens(0).start, 0)
        self.assertEqual(encoding.word_to_tokens(0, 0).start, 0)
        self.assertEqual(encoding.word_to_tokens(last_word_index).end, last_token_index + 1)
        self.assertEqual(encoding.word_to_tokens(0, last_word_index).end, last_token_index + 1)
        self.assertEqual(batch_encoding.word_to_tokens(1, 0).start, 0)
        self.assertEqual(batch_encoding.word_to_tokens(0, last_word_index).end, last_token_index + 1)
        self.assertEqual(batch_encoding.word_to_tokens(last_batch_index, last_word_index).end, last_token_index + 1)

        # Assert token_to_chars
        self.assertEqual(encoding.token_to_chars(0).start, 0)
        self.assertEqual(encoding.token_to_chars(0, 0).start, 0)
        self.assertEqual(encoding.token_to_chars(last_token_index).end, last_char_index + 1)
        self.assertEqual(encoding.token_to_chars(0, last_token_index).end, last_char_index + 1)
        self.assertEqual(batch_encoding.token_to_chars(1, 0).start, 0)
        self.assertEqual(batch_encoding.token_to_chars(0, last_token_index).end, last_char_index + 1)
        self.assertEqual(batch_encoding.token_to_chars(last_batch_index, last_token_index).end, last_char_index + 1)

        # Assert char_to_token
        self.assertEqual(encoding.char_to_token(0), 0)
        self.assertEqual(encoding.char_to_token(0, 0), 0)
        self.assertEqual(encoding.char_to_token(last_char_index), last_token_index)
        self.assertEqual(encoding.char_to_token(0, last_char_index), last_token_index)
        self.assertEqual(batch_encoding.char_to_token(1, 0), 0)
        self.assertEqual(batch_encoding.char_to_token(0, last_char_index), last_token_index)
        self.assertEqual(batch_encoding.char_to_token(last_batch_index, last_char_index), last_token_index)

        # Assert char_to_word
        self.assertEqual(encoding.char_to_word(0), 0)
        self.assertEqual(encoding.char_to_word(0, 0), 0)
        self.assertEqual(encoding.char_to_word(last_char_index), last_word_index)
        self.assertEqual(encoding.char_to_word(0, last_char_index), last_word_index)
        self.assertEqual(batch_encoding.char_to_word(1, 0), 0)
        self.assertEqual(batch_encoding.char_to_word(0, last_char_index), last_word_index)
        self.assertEqual(batch_encoding.char_to_word(last_batch_index, last_char_index), last_word_index)

        # Assert word_to_chars
        self.assertEqual(encoding.word_to_chars(0).start, 0)
        self.assertEqual(encoding.word_to_chars(0, 0).start, 0)
        self.assertEqual(encoding.word_to_chars(last_word_index).end, last_char_index + 1)
        self.assertEqual(encoding.word_to_chars(0, last_word_index).end, last_char_index + 1)
        self.assertEqual(batch_encoding.word_to_chars(1, 0).start, 0)
        self.assertEqual(batch_encoding.word_to_chars(0, last_word_index).end, last_char_index + 1)
        self.assertEqual(batch_encoding.word_to_chars(last_batch_index, last_word_index).end, last_char_index + 1)
Funtowicz Morgan's avatar
Funtowicz Morgan committed
187

188
    def assert_tokenization_python_rust_equals(self, tokenizer_r, tokenizer_p):
189
190
191
192
193
        # Ensure basic input match
        input_p = tokenizer_p.encode_plus(self._data)
        input_r = tokenizer_r.encode_plus(self._data)

        for key in filter(lambda x: x in ["input_ids", "token_type_ids", "attention_mask"], input_p.keys()):
Funtowicz Morgan's avatar
Funtowicz Morgan committed
194
            self.assertSequenceEqual(input_p[key], input_r[key])
195
196
197
198
199

        input_pairs_p = tokenizer_p.encode_plus(self._data, self._data)
        input_pairs_r = tokenizer_r.encode_plus(self._data, self._data)

        for key in filter(lambda x: x in ["input_ids", "token_type_ids", "attention_mask"], input_p.keys()):
Funtowicz Morgan's avatar
Funtowicz Morgan committed
200
            self.assertSequenceEqual(input_pairs_p[key], input_pairs_r[key])
201
202

        # Ensure truncation match
203
204
        input_p = tokenizer_p.encode_plus(self._data, max_length=512, truncation=True)
        input_r = tokenizer_r.encode_plus(self._data, max_length=512, truncation=True)
205
206

        for key in filter(lambda x: x in ["input_ids", "token_type_ids", "attention_mask"], input_p.keys()):
Funtowicz Morgan's avatar
Funtowicz Morgan committed
207
            self.assertSequenceEqual(input_p[key], input_r[key])
208
209

        # Ensure truncation with stride match
210
211
212
213
214
215
        input_p = tokenizer_p.encode_plus(
            self._data, max_length=512, truncation=True, stride=3, return_overflowing_tokens=True
        )
        input_r = tokenizer_r.encode_plus(
            self._data, max_length=512, truncation=True, stride=3, return_overflowing_tokens=True
        )
216
217

        for key in filter(lambda x: x in ["input_ids", "token_type_ids", "attention_mask"], input_p.keys()):
218
            self.assertSequenceEqual(input_p[key], input_r[key][0])
Funtowicz Morgan's avatar
Funtowicz Morgan committed
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233

    def assert_num_special_tokens_to_add_equal(self, tokenizer_r, tokenizer_p):
        # Check we have the same number of added_tokens for both pair and non-pair inputs.
        self.assertEqual(tokenizer_r.num_special_tokens_to_add(False), tokenizer_p.num_special_tokens_to_add(False))
        self.assertEqual(tokenizer_r.num_special_tokens_to_add(True), tokenizer_p.num_special_tokens_to_add(True))

    def assert_max_length_equal(self, tokenizer_r, tokenizer_p):
        # Check we have the correct max_length for both pair and non-pair inputs.
        self.assertEqual(tokenizer_r.max_len_single_sentence, tokenizer_p.max_len_single_sentence)
        self.assertEqual(tokenizer_r.max_len_sentences_pair, tokenizer_p.max_len_sentences_pair)

    def assert_special_tokens_map_equal(self, tokenizer_r, tokenizer_p):
        # Assert the set of special tokens match.
        self.assertSequenceEqual(
            tokenizer_p.special_tokens_map.items(), tokenizer_r.special_tokens_map.items(),
234
235
        )

236
237
238
239
240
241
242
243
    def assert_add_tokens(self, tokenizer_r):
        vocab_size = tokenizer_r.vocab_size
        self.assertEqual(tokenizer_r.add_tokens(""), 0)
        self.assertEqual(tokenizer_r.add_tokens("testoken"), 1)
        self.assertEqual(tokenizer_r.add_tokens(["testoken1", "testtoken2"]), 2)
        self.assertEqual(len(tokenizer_r), vocab_size + 3)

        self.assertEqual(tokenizer_r.add_special_tokens({}), 0)
244
        self.assertEqual(tokenizer_r.add_special_tokens({"bos_token": "[BOS]", "eos_token": "[EOS]"}), 2)
245
246
247
248
249
250
251
        self.assertRaises(
            AssertionError, tokenizer_r.add_special_tokens, {"additional_special_tokens": "<testtoken1>"}
        )
        self.assertEqual(tokenizer_r.add_special_tokens({"additional_special_tokens": ["<testtoken2>"]}), 1)
        self.assertEqual(
            tokenizer_r.add_special_tokens({"additional_special_tokens": ["<testtoken3>", "<testtoken4>"]}), 2
        )
252
        self.assertEqual(len(tokenizer_r), vocab_size + 8)
253

Funtowicz Morgan's avatar
Funtowicz Morgan committed
254
    def assert_offsets_mapping(self, tokenizer_r):
255
256
257
258
        text = "Wonderful no inspiration example with subtoken"
        pair = "Along with an awesome pair"

        # No pair
Funtowicz Morgan's avatar
Funtowicz Morgan committed
259
260
261
262
        tokens_with_offsets = tokenizer_r.encode_plus(
            text, return_special_tokens_mask=True, return_offsets_mapping=True, add_special_tokens=True
        )
        added_tokens = tokenizer_r.num_special_tokens_to_add(False)
263
264
265
266
267
268
269
270
271
        offsets = tokens_with_offsets["offset_mapping"]

        # Assert there is the same number of tokens and offsets
        self.assertEqual(len(offsets), len(tokens_with_offsets["input_ids"]))

        # Assert there is online added_tokens special_tokens
        self.assertEqual(sum(tokens_with_offsets["special_tokens_mask"]), added_tokens)

        # Pairs
Funtowicz Morgan's avatar
Funtowicz Morgan committed
272
273
        tokens_with_offsets = tokenizer_r.encode_plus(
            text, pair, return_special_tokens_mask=True, return_offsets_mapping=True, add_special_tokens=True
274
        )
Funtowicz Morgan's avatar
Funtowicz Morgan committed
275
        added_tokens = tokenizer_r.num_special_tokens_to_add(True)
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
        offsets = tokens_with_offsets["offset_mapping"]

        # Assert there is the same number of tokens and offsets
        self.assertEqual(len(offsets), len(tokens_with_offsets["input_ids"]))

        # Assert there is online added_tokens special_tokens
        self.assertEqual(sum(tokens_with_offsets["special_tokens_mask"]), added_tokens)

    def assert_batch_encode_dynamic_overflowing(self, tokenizer: PreTrainedTokenizer):
        """
        When calling batch_encode with multiple sequence it can returns different number of
        overflowing encoding for each sequence:
        [
          Sequence 1: [Encoding 1, Encoding 2],
          Sequence 2: [Encoding 1],
          Sequence 3: [Encoding 1, Encoding 2, ... Encoding N]
        ]
        This needs to be padded so that it can represented as a tensor
        """
        returned_tensor = "pt" if is_torch_available() else "tf"

297
298
299
        if not tokenizer.pad_token or tokenizer.pad_token_id < 0:
            return

300
301
302
        tokens = tokenizer.encode_plus(
            "HuggingFace is solving NLP one commit at a time",
            max_length=6,
303
304
            padding=True,
            truncation=True,
305
306
307
308
309
310
311
312
313
314
315
            return_tensors=returned_tensor,
            return_overflowing_tokens=True,
        )

        for key in filter(lambda x: "overflow_to_sample_mapping" not in x, tokens.keys()):
            self.assertEqual(len(tokens[key].shape), 2)

        # Mono sample
        tokens = tokenizer.batch_encode_plus(
            ["HuggingFace is solving NLP one commit at a time"],
            max_length=6,
316
317
            padding=True,
            truncation="only_first",
318
319
320
321
322
323
324
325
326
327
328
329
            return_tensors=returned_tensor,
            return_overflowing_tokens=True,
        )

        for key in filter(lambda x: "overflow_to_sample_mapping" not in x, tokens.keys()):
            self.assertEqual(len(tokens[key].shape), 2)
            self.assertEqual(tokens[key].shape[-1], 6)

        # Multi sample
        tokens = tokenizer.batch_encode_plus(
            ["HuggingFace is solving NLP one commit at a time", "Very tiny input"],
            max_length=6,
330
331
            padding=True,
            truncation="only_first",
332
333
334
335
336
337
338
339
            return_tensors=returned_tensor,
            return_overflowing_tokens=True,
        )

        for key in filter(lambda x: "overflow_to_sample_mapping" not in x, tokens.keys()):
            self.assertEqual(len(tokens[key].shape), 2)
            self.assertEqual(tokens[key].shape[-1], 6)

340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
    def assert_pretokenized_inputs(self, tokenizer_r, tokenizer_p):
        # Input string
        pretokenized_input_simple = "This is a sample input".split()
        pretokenized_input_pair = "This is a sample pair".split()

        # Test encode for pretokenized inputs
        output_r = tokenizer_r.encode(pretokenized_input_simple, is_pretokenized=True)
        output_p = tokenizer_p.encode(pretokenized_input_simple, is_pretokenized=True)
        self.assertEqual(output_p, output_r)

        kwargs = {
            "is_pretokenized": True,
            "return_token_type_ids": True,
            "return_attention_mask": True,
            "return_overflowing_tokens": False,
            "return_special_tokens_mask": True,
            "return_offsets_mapping": False,  # Not implemented in python tokenizers
        }
358
359
360
361
362
363
364
365
        batch_kwargs = {
            "is_pretokenized": True,
            "return_token_type_ids": True,
            "return_attention_mask": True,  # we have an 's' here
            "return_overflowing_tokens": False,
            "return_special_tokens_mask": True,  # we have an 's' here
            "return_offsets_mapping": False,  # Not implemented in python tokenizers
        }
366
367
368
369
370
371
372
373
        # Test encode_plus for pretokenized inputs
        output_r = tokenizer_r.encode_plus(pretokenized_input_simple, **kwargs)
        output_p = tokenizer_p.encode_plus(pretokenized_input_simple, **kwargs)
        for key in output_p.keys():
            self.assertEqual(output_p[key], output_r[key])

        # Test batch_encode_plus for pretokenized inputs
        input_batch = ([pretokenized_input_simple] * 2) + [pretokenized_input_simple + pretokenized_input_pair]
374
375
        output_r = tokenizer_r.batch_encode_plus(input_batch, **batch_kwargs)
        output_p = tokenizer_p.batch_encode_plus(input_batch, **batch_kwargs)
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
        for key in output_p.keys():
            self.assertEqual(output_p[key], output_r[key])

        # Test encode for pretokenized inputs pairs
        output_r = tokenizer_r.encode(pretokenized_input_simple, pretokenized_input_pair, is_pretokenized=True)
        output_p = tokenizer_p.encode(pretokenized_input_simple, pretokenized_input_pair, is_pretokenized=True)
        self.assertEqual(output_p, output_r)

        # Test encode_plus for pretokenized inputs
        output_r = tokenizer_r.encode_plus(pretokenized_input_simple, pretokenized_input_pair, **kwargs)
        output_p = tokenizer_p.encode_plus(pretokenized_input_simple, pretokenized_input_pair, **kwargs)
        for key in output_p.keys():
            self.assertEqual(output_p[key], output_r[key])

        # Test batch_encode_plus for pretokenized inputs
        input_batch_pair = ([pretokenized_input_simple, pretokenized_input_pair] * 2) + [
            pretokenized_input_simple + pretokenized_input_pair,
            pretokenized_input_pair,
        ]
395
396
        output_r = tokenizer_r.batch_encode_plus(input_batch_pair, **batch_kwargs)
        output_p = tokenizer_p.batch_encode_plus(input_batch_pair, **batch_kwargs)
397
398
399
        for key in output_p.keys():
            self.assertEqual(output_p[key], output_r[key])

400
401
402
403
404
405
406
407
408
409
410
411
412
413
    def assert_create_token_type_ids(self, tokenizer_r, tokenizer_p):
        input_simple = [1, 2, 3]
        input_pair = [1, 2, 3]

        # Generate output
        output_r = tokenizer_r.create_token_type_ids_from_sequences(input_simple)
        output_p = tokenizer_p.create_token_type_ids_from_sequences(input_simple)
        self.assertEqual(output_p, output_r)

        # Generate pair output
        output_r = tokenizer_r.create_token_type_ids_from_sequences(input_simple, input_pair)
        output_p = tokenizer_p.create_token_type_ids_from_sequences(input_simple, input_pair)
        self.assertEqual(output_p, output_r)

414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
    def assert_build_inputs_with_special_tokens(self, tokenizer_r, tokenizer_p):
        # Input string
        input_simple = tokenizer_p.tokenize("This is a sample input")
        input_pair = tokenizer_p.tokenize("This is a sample pair")

        # Generate output
        output_r = tokenizer_r.build_inputs_with_special_tokens(input_simple)
        output_p = tokenizer_p.build_inputs_with_special_tokens(input_simple)
        self.assertEqual(output_p, output_r)

        # Generate pair output
        output_r = tokenizer_r.build_inputs_with_special_tokens(input_simple, input_pair)
        output_p = tokenizer_p.build_inputs_with_special_tokens(input_simple, input_pair)
        self.assertEqual(output_p, output_r)

        # Input tokens id
        input_simple = tokenizer_p.encode("This is a sample input")
        input_pair = tokenizer_p.encode("This is a sample pair")

        # Generate output
        output_r = tokenizer_r.build_inputs_with_special_tokens(input_simple)
        output_p = tokenizer_p.build_inputs_with_special_tokens(input_simple)
        self.assertEqual(output_p, output_r)

        # Generate pair output
        output_r = tokenizer_r.build_inputs_with_special_tokens(input_simple, input_pair)
        output_p = tokenizer_p.build_inputs_with_special_tokens(input_simple, input_pair)
        self.assertEqual(output_p, output_r)

Funtowicz Morgan's avatar
Funtowicz Morgan committed
443
444
    def assert_padding(self, tokenizer_r, tokenizer_p, max_length=15):
        def assert_padded_input_match(input_r: list, input_p: list, max_length: int):
445

Funtowicz Morgan's avatar
Funtowicz Morgan committed
446
            # Ensure we match max_length
447
448
            self.assertEqual(len(input_r), max_length)
            self.assertEqual(len(input_p), max_length)
449

Funtowicz Morgan's avatar
Funtowicz Morgan committed
450
451
452
453
            # Ensure the number of padded tokens is the same
            padded_tokens_r = list(takewhile(lambda i: i == tokenizer_r.pad_token_id, reversed(input_r)))
            padded_tokens_p = list(takewhile(lambda i: i == tokenizer_p.pad_token_id, reversed(input_p)))
            self.assertSequenceEqual(padded_tokens_r, padded_tokens_p)
454

455
        def assert_batch_padded_input_match(input_r: dict, input_p: dict, max_length: int):
Funtowicz Morgan's avatar
Funtowicz Morgan committed
456
            for i_r in input_r.values():
457
458
459
460
461
462
                self.assertEqual(len(i_r), 2), self.assertEqual(len(i_r[0]), max_length), self.assertEqual(
                    len(i_r[1]), max_length
                )
                self.assertEqual(len(i_r), 2), self.assertEqual(len(i_r[0]), max_length), self.assertEqual(
                    len(i_r[1]), max_length
                )
463

Funtowicz Morgan's avatar
Funtowicz Morgan committed
464
465
            for i_r, i_p in zip(input_r["input_ids"], input_p["input_ids"]):
                assert_padded_input_match(i_r, i_p, max_length)
466

Funtowicz Morgan's avatar
Funtowicz Morgan committed
467
468
            for i_r, i_p in zip(input_r["attention_mask"], input_p["attention_mask"]):
                self.assertSequenceEqual(i_r, i_p)
469

470
        # Encode - Simple input
Funtowicz Morgan's avatar
Funtowicz Morgan committed
471
472
473
        input_r = tokenizer_r.encode("This is a simple input", max_length=max_length, pad_to_max_length=True)
        input_p = tokenizer_p.encode("This is a simple input", max_length=max_length, pad_to_max_length=True)
        assert_padded_input_match(input_r, input_p, max_length)
474
475
476
        input_r = tokenizer_r.encode("This is a simple input", max_length=max_length, padding="max_length")
        input_p = tokenizer_p.encode("This is a simple input", max_length=max_length, padding="max_length")
        assert_padded_input_match(input_r, input_p, max_length)
477

478
479
480
481
482
        input_r = tokenizer_r.encode("This is a simple input", padding="longest")
        input_p = tokenizer_p.encode("This is a simple input", padding=True)
        assert_padded_input_match(input_r, input_p, len(input_r))

        # Encode - Pair input
Funtowicz Morgan's avatar
Funtowicz Morgan committed
483
484
485
486
487
488
489
        input_r = tokenizer_r.encode(
            "This is a simple input", "This is a pair", max_length=max_length, pad_to_max_length=True
        )
        input_p = tokenizer_p.encode(
            "This is a simple input", "This is a pair", max_length=max_length, pad_to_max_length=True
        )
        assert_padded_input_match(input_r, input_p, max_length)
490
491
492
493
494
495
496
497
498
499
        input_r = tokenizer_r.encode(
            "This is a simple input", "This is a pair", max_length=max_length, padding="max_length"
        )
        input_p = tokenizer_p.encode(
            "This is a simple input", "This is a pair", max_length=max_length, padding="max_length"
        )
        assert_padded_input_match(input_r, input_p, max_length)
        input_r = tokenizer_r.encode("This is a simple input", "This is a pair", padding=True)
        input_p = tokenizer_p.encode("This is a simple input", "This is a pair", padding="longest")
        assert_padded_input_match(input_r, input_p, len(input_r))
500

501
        # Encode_plus - Simple input
Funtowicz Morgan's avatar
Funtowicz Morgan committed
502
503
504
505
        input_r = tokenizer_r.encode_plus("This is a simple input", max_length=max_length, pad_to_max_length=True)
        input_p = tokenizer_p.encode_plus("This is a simple input", max_length=max_length, pad_to_max_length=True)
        assert_padded_input_match(input_r["input_ids"], input_p["input_ids"], max_length)
        self.assertSequenceEqual(input_r["attention_mask"], input_p["attention_mask"])
506
507
508
509
        input_r = tokenizer_r.encode_plus("This is a simple input", max_length=max_length, padding="max_length")
        input_p = tokenizer_p.encode_plus("This is a simple input", max_length=max_length, padding="max_length")
        assert_padded_input_match(input_r["input_ids"], input_p["input_ids"], max_length)
        self.assertSequenceEqual(input_r["attention_mask"], input_p["attention_mask"])
510

511
512
513
514
515
516
517
        input_r = tokenizer_r.encode_plus("This is a simple input", padding="longest")
        input_p = tokenizer_p.encode_plus("This is a simple input", padding=True)
        assert_padded_input_match(input_r["input_ids"], input_p["input_ids"], len(input_r["input_ids"]))

        self.assertSequenceEqual(input_r["attention_mask"], input_p["attention_mask"])

        # Encode_plus - Pair input
Funtowicz Morgan's avatar
Funtowicz Morgan committed
518
519
520
521
522
523
524
525
        input_r = tokenizer_r.encode_plus(
            "This is a simple input", "This is a pair", max_length=max_length, pad_to_max_length=True
        )
        input_p = tokenizer_p.encode_plus(
            "This is a simple input", "This is a pair", max_length=max_length, pad_to_max_length=True
        )
        assert_padded_input_match(input_r["input_ids"], input_p["input_ids"], max_length)
        self.assertSequenceEqual(input_r["attention_mask"], input_p["attention_mask"])
526
527
528
529
530
531
532
533
534
535
536
537
        input_r = tokenizer_r.encode_plus(
            "This is a simple input", "This is a pair", max_length=max_length, padding="max_length"
        )
        input_p = tokenizer_p.encode_plus(
            "This is a simple input", "This is a pair", max_length=max_length, padding="max_length"
        )
        assert_padded_input_match(input_r["input_ids"], input_p["input_ids"], max_length)
        self.assertSequenceEqual(input_r["attention_mask"], input_p["attention_mask"])
        input_r = tokenizer_r.encode_plus("This is a simple input", "This is a pair", padding="longest")
        input_p = tokenizer_p.encode_plus("This is a simple input", "This is a pair", padding=True)
        assert_padded_input_match(input_r["input_ids"], input_p["input_ids"], len(input_r["input_ids"]))
        self.assertSequenceEqual(input_r["attention_mask"], input_p["attention_mask"])
538

539
        # Batch_encode_plus - Simple input
Funtowicz Morgan's avatar
Funtowicz Morgan committed
540
541
542
543
544
545
        input_r = tokenizer_r.batch_encode_plus(
            ["This is a simple input 1", "This is a simple input 2"], max_length=max_length, pad_to_max_length=True
        )
        input_p = tokenizer_p.batch_encode_plus(
            ["This is a simple input 1", "This is a simple input 2"], max_length=max_length, pad_to_max_length=True
        )
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
        assert_batch_padded_input_match(input_r, input_p, max_length)

        input_r = tokenizer_r.batch_encode_plus(
            ["This is a simple input 1", "This is a simple input 2"], max_length=max_length, padding="max_length",
        )
        input_p = tokenizer_p.batch_encode_plus(
            ["This is a simple input 1", "This is a simple input 2"], max_length=max_length, padding="max_length",
        )
        assert_batch_padded_input_match(input_r, input_p, max_length)

        input_r = tokenizer_r.batch_encode_plus(
            ["This is a simple input 1", "This is a simple input 2"], max_length=max_length, padding="longest",
        )
        input_p = tokenizer_p.batch_encode_plus(
            ["This is a simple input 1", "This is a simple input 2"], max_length=max_length, padding=True,
        )
        assert_batch_padded_input_match(input_r, input_p, len(input_r["input_ids"][0]))

        input_r = tokenizer_r.batch_encode_plus(
            ["This is a simple input 1", "This is a simple input 2"], padding="longest"
        )
        input_p = tokenizer_p.batch_encode_plus(["This is a simple input 1", "This is a simple input 2"], padding=True)
        assert_batch_padded_input_match(input_r, input_p, len(input_r["input_ids"][0]))

        # Batch_encode_plus - Pair input
        input_r = tokenizer_r.batch_encode_plus(
            [
                ("This is a simple input 1", "This is a simple input 2"),
                ("This is a simple pair 1", "This is a simple pair 2"),
            ],
            max_length=max_length,
            truncation=True,
            padding="max_length",
        )
        input_p = tokenizer_p.batch_encode_plus(
            [
                ("This is a simple input 1", "This is a simple input 2"),
                ("This is a simple pair 1", "This is a simple pair 2"),
            ],
            max_length=max_length,
            truncation=True,
            padding="max_length",
        )
        assert_batch_padded_input_match(input_r, input_p, max_length)
590

Funtowicz Morgan's avatar
Funtowicz Morgan committed
591
592
593
594
595
        input_r = tokenizer_r.batch_encode_plus(
            [
                ("This is a simple input 1", "This is a simple input 2"),
                ("This is a simple pair 1", "This is a simple pair 2"),
            ],
596
            padding=True,
Funtowicz Morgan's avatar
Funtowicz Morgan committed
597
598
599
600
601
602
        )
        input_p = tokenizer_p.batch_encode_plus(
            [
                ("This is a simple input 1", "This is a simple input 2"),
                ("This is a simple pair 1", "This is a simple pair 2"),
            ],
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
            padding="longest",
        )
        assert_batch_padded_input_match(input_r, input_p, len(input_r["input_ids"][0]))

        # Using pad on single examples after tokenization
        input_r = tokenizer_r.encode_plus("This is a input 1")
        input_r = tokenizer_r.pad(input_r)

        input_p = tokenizer_r.encode_plus("This is a input 1")
        input_p = tokenizer_r.pad(input_p)

        assert_padded_input_match(input_r["input_ids"], input_p["input_ids"], len(input_r["input_ids"]))

        # Using pad on single examples after tokenization
        input_r = tokenizer_r.encode_plus("This is a input 1")
        input_r = tokenizer_r.pad(input_r, max_length=max_length, padding="max_length")

        input_p = tokenizer_r.encode_plus("This is a input 1")
        input_p = tokenizer_r.pad(input_p, max_length=max_length, padding="max_length")

        assert_padded_input_match(input_r["input_ids"], input_p["input_ids"], max_length)

        # Using pad after tokenization
        input_r = tokenizer_r.batch_encode_plus(
            ["This is a input 1", "This is a much longer input whilch should be padded"]
Funtowicz Morgan's avatar
Funtowicz Morgan committed
628
        )
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
        input_r = tokenizer_r.pad(input_r)

        input_p = tokenizer_r.batch_encode_plus(
            ["This is a input 1", "This is a much longer input whilch should be padded"]
        )
        input_p = tokenizer_r.pad(input_p)

        assert_batch_padded_input_match(input_r, input_p, len(input_r["input_ids"][0]))

        # Using pad after tokenization
        input_r = tokenizer_r.batch_encode_plus(
            ["This is a input 1", "This is a much longer input whilch should be padded"]
        )
        input_r = tokenizer_r.pad(input_r, max_length=max_length, padding="max_length")

        input_p = tokenizer_r.batch_encode_plus(
            ["This is a input 1", "This is a much longer input whilch should be padded"]
        )
        input_p = tokenizer_r.pad(input_p, max_length=max_length, padding="max_length")

        assert_batch_padded_input_match(input_r, input_p, max_length)
650

Funtowicz Morgan's avatar
Funtowicz Morgan committed
651
652
653
    def assert_save_pretrained(self, tokenizer_r, tokenizer_p):
        # Checks it save with the same files
        self.assertSequenceEqual(tokenizer_r.save_vocabulary("."), tokenizer_p.save_vocabulary("."))
654

Funtowicz Morgan's avatar
Funtowicz Morgan committed
655
656
        # Checks everything loads correctly in the same way
        tokenizer_rp, tokenizer_pp = tokenizer_r.from_pretrained("."), tokenizer_p.from_pretrained(".")
657

Funtowicz Morgan's avatar
Funtowicz Morgan committed
658
659
660
661
662
        # Check special tokens are set accordingly on Rust and Python
        for key in tokenizer_pp.special_tokens_map:
            self.assertTrue(hasattr(tokenizer_rp, key))
            # self.assertEqual(getattr(tokenizer_rp, key), getattr(tokenizer_pp, key))
            # self.assertEqual(getattr(tokenizer_rp, key + "_id"), getattr(tokenizer_pp, key + "_id"))
663

Funtowicz Morgan's avatar
Funtowicz Morgan committed
664
665
666
667
668
669
670
671
    def assert_embeded_special_tokens(self, tokenizer_r, tokenizer_p):
        sentence = "A, <mask> AllenNLP sentence."
        tokens_r = tokenizer_r.encode_plus(
            sentence, add_special_tokens=True, return_attention_mask=False, return_token_type_ids=True
        )
        tokens_p = tokenizer_p.encode_plus(
            sentence, add_special_tokens=True, return_attention_mask=False, return_token_type_ids=True
        )
672

Funtowicz Morgan's avatar
Funtowicz Morgan committed
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
        for key in tokens_p.keys():
            self.assertEqual(tokens_r[key], tokens_p[key])

        self.assertEqual(sum(tokens_r["token_type_ids"]), 0)
        self.assertEqual(sum(tokens_p["token_type_ids"]), 0)

        tokens_r = tokenizer_r.convert_ids_to_tokens(tokens_r["input_ids"])
        tokens_p = tokenizer_p.convert_ids_to_tokens(tokens_p["input_ids"])
        self.assertSequenceEqual(tokens_r, tokens_p)

    def assert_add_special_tokens(self, tokenizer_r):
        simple_num_special_tokens_to_add = tokenizer_r.num_special_tokens_to_add(pair=False)
        # pair_num_special_tokens_to_add = tokenizer_r.num_special_tokens_to_add(pair=True)

        for text in ["", " "]:
            # tokenize()
            no_special_tokens = tokenizer_r.tokenize(text, add_special_tokens=False)
            with_special_tokens = tokenizer_r.tokenize(text, add_special_tokens=True)
            self.assertEqual(len(no_special_tokens), len(with_special_tokens) - simple_num_special_tokens_to_add)

            # encode()
            no_special_tokens = tokenizer_r.encode(text, add_special_tokens=False)
            with_special_tokens = tokenizer_r.encode(text, add_special_tokens=True)
            self.assertEqual(len(no_special_tokens), len(with_special_tokens) - simple_num_special_tokens_to_add)

            # encode_plus()
            no_special_tokens = tokenizer_r.encode_plus(text, add_special_tokens=False)
            with_special_tokens = tokenizer_r.encode_plus(text, add_special_tokens=True)
            for key in no_special_tokens.keys():
                self.assertEqual(
                    len(no_special_tokens[key]), len(with_special_tokens[key]) - simple_num_special_tokens_to_add
                )

            # # batch_encode_plus
            no_special_tokens = tokenizer_r.batch_encode_plus([text, text], add_special_tokens=False)
            with_special_tokens = tokenizer_r.batch_encode_plus([text, text], add_special_tokens=True)
            for key in no_special_tokens.keys():
                for i_no, i_with in zip(no_special_tokens[key], with_special_tokens[key]):
                    self.assertEqual(len(i_no), len(i_with) - simple_num_special_tokens_to_add)

713
714
715
716
717
718
    def assert_prepare_for_model(self, tokenizer_r, tokenizer_p):
        string_sequence = "Asserting that both tokenizers are equal"
        python_output = tokenizer_p.prepare_for_model(tokenizer_p.encode(string_sequence))
        rust_output = tokenizer_r.prepare_for_model(tokenizer_r.encode(string_sequence))
        self.assertEqual(python_output, rust_output)

Funtowicz Morgan's avatar
Funtowicz Morgan committed
719
720
721
722
723
724
725
726

class WordPieceFastTokenizerTest(CommonFastTokenizerTest):
    """
    Override all the specific methods to test WordPiece behavior
    """

    TOKENIZERS_CLASSES = frozenset(
        [
727
728
729
730
            Tokenizer("Bert", BertTokenizerFast, BertTokenizer, "vocab_file", filter_non_english, None),
            Tokenizer(
                "DistilBert", DistilBertTokenizerFast, DistilBertTokenizer, "vocab_file", filter_non_english, None
            ),
Funtowicz Morgan's avatar
Funtowicz Morgan committed
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
        ]
    )

    def fast_only(self, tokenizer_r):
        super().fast_only(tokenizer_r)
        self.assert_offsets_with_special_characters(tokenizer_r)

    def assert_add_special_tokens(self, tokenizer_r):
        super().assert_add_special_tokens(tokenizer_r)

    def assert_offsets_with_special_characters(self, tokenizer_r):
        sentence = "A, na茂ve [MASK] AllenNLP sentence."
        tokens = tokenizer_r.encode_plus(
            sentence,
            return_attention_mask=False,
            return_token_type_ids=False,
            return_offsets_mapping=True,
            add_special_tokens=True,
        )
750

Funtowicz Morgan's avatar
Funtowicz Morgan committed
751
752
753
754
755
756
757
758
759
760
761
        expected_results = [
            ((0, 1), "A"),
            ((1, 2), ","),
            ((3, 8), "naive"),  # BERT normalizes this away
            # Append MASK here after lower-casing
            ((16, 21), "Allen"),
            ((22, 24), "##NL"),
            ((24, 25), "##P"),
            ((26, 34), "sentence"),
            ((35, 36), "."),
        ]
762

Funtowicz Morgan's avatar
Funtowicz Morgan committed
763
764
765
        # Check if the tokenizer is uncased
        if tokenizer_r.init_kwargs.get("do_lower_case"):
            expected_results = [(offset, token.lower()) for (offset, token) in expected_results]
766

Funtowicz Morgan's avatar
Funtowicz Morgan committed
767
768
769
770
        # Append the special tokens
        expected_results.insert(3, ((9, 15), "[MASK]"))
        expected_results.insert(0, (None, "[CLS]"))
        expected_results.append((None, "[SEP]"))
771

Funtowicz Morgan's avatar
Funtowicz Morgan committed
772
773
        self.assertEqual([e[1] for e in expected_results], tokenizer_r.convert_ids_to_tokens(tokens["input_ids"]))
        # self.assertEqual([e[0] for e in expected_results], tokens["offset_mapping"])
774
775


Funtowicz Morgan's avatar
Funtowicz Morgan committed
776
777
class RobertaFastTokenizerTest(CommonFastTokenizerTest):
    TOKENIZERS_CLASSES = frozenset(
778
779
780
781
782
783
784
785
786
787
        [
            Tokenizer(
                "Roberta",
                RobertaTokenizerFast,
                RobertaTokenizer,
                "vocab_file",
                filter_roberta_detectors,
                (("cls_token", "<s>"),),
            )
        ]
Funtowicz Morgan's avatar
Funtowicz Morgan committed
788
    )
789

Funtowicz Morgan's avatar
Funtowicz Morgan committed
790
791
792
793
    def assert_embeded_special_tokens(self, tokenizer_r, tokenizer_p):
        sentence = "A, <mask> AllenNLP sentence."
        tokens_r = tokenizer_r.encode_plus(sentence, add_special_tokens=True, return_token_type_ids=True)
        tokens_p = tokenizer_p.encode_plus(sentence, add_special_tokens=True, return_token_type_ids=True)
794

Funtowicz Morgan's avatar
Funtowicz Morgan committed
795
        # Rust correctly handles the space before the mask while python doesnt
796
797
        self.assertSequenceEqual(tokens_r["input_ids"], [0, 250, 6, 50264, 3823, 487, 21992, 3645, 4, 2])
        self.assertSequenceEqual(tokens_p["input_ids"], [0, 250, 6, 50264, 3823, 487, 21992, 3645, 4, 2])
798

Funtowicz Morgan's avatar
Funtowicz Morgan committed
799
800
        # token_type_ids should put 0 everywhere
        self.assertEquals(sum(tokens_r["token_type_ids"]), sum(tokens_p["token_type_ids"]))
801

Funtowicz Morgan's avatar
Funtowicz Morgan committed
802
803
804
805
806
        # attention_mask should put 1 everywhere, so sum over length should be 1
        self.assertEquals(
            sum(tokens_r["attention_mask"]) / len(tokens_r["attention_mask"]),
            sum(tokens_p["attention_mask"]) / len(tokens_p["attention_mask"]),
        )
807

Funtowicz Morgan's avatar
Funtowicz Morgan committed
808
        tokens_r = tokenizer_r.convert_ids_to_tokens(tokens_r["input_ids"])
809
810
811
        tokens_p = tokenizer_p.convert_ids_to_tokens(tokens_p["input_ids"])
        self.assertSequenceEqual(tokens_r, ["<s>", "A", ",", "<mask>", "臓Allen", "N", "LP", "臓sentence", ".", "</s>"])
        self.assertSequenceEqual(tokens_p, ["<s>", "A", ",", "<mask>", "臓Allen", "N", "LP", "臓sentence", ".", "</s>"])
812

813

Funtowicz Morgan's avatar
Funtowicz Morgan committed
814
815
class NoPaddingTokenFastTokenizerMatchingTest(CommonFastTokenizerTest):
    TOKENIZERS_CLASSES = [
816
817
        Tokenizer("OpenAI GPT", OpenAIGPTTokenizerFast, OpenAIGPTTokenizer, "vocab_file", None, None),
        Tokenizer("GPT2", GPT2TokenizerFast, GPT2Tokenizer, "vocab_file", None, [("add_prefix_space", True)]),
Funtowicz Morgan's avatar
Funtowicz Morgan committed
818
    ]
819

820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
    def fast_align_python(self, tokenizer_r, tokenizer_p, tok_case, pretrained_name):
        # Check is_fast is set correctly
        self.assertFalse(tokenizer_p.is_fast)
        self.assertTrue(tokenizer_r.is_fast)

        # Check that Rust and Python align
        self.assert_tokenization_python_rust_equals(tokenizer_r, tokenizer_p)
        self.assert_num_special_tokens_to_add_equal(tokenizer_r, tokenizer_p)
        self.assert_max_length_equal(tokenizer_r, tokenizer_p)
        self.assert_special_tokens_map_equal(tokenizer_r, tokenizer_p)
        self.assert_embeded_special_tokens(tokenizer_r, tokenizer_p)
        self.assert_padding(tokenizer_r, tokenizer_p)

        # Specific for
        kwargs = {}
        if tok_case.kwargs is not None:
            kwargs = dict(tok_case.kwargs)
        tokenizer_r = tok_case.rust_cls.from_pretrained(pretrained_name, **kwargs)
        self.assert_pretokenized_inputs(tokenizer_r, tokenizer_p)

Funtowicz Morgan's avatar
Funtowicz Morgan committed
840
841
842
843
844
845
846
847
848
    def assert_padding(self, tokenizer_r, tokenizer_p, max_length=15):
        # Simple input
        s = "This is a simple input"
        s2 = ["This is a simple input 1", "This is a simple input 2"]
        p = ("This is a simple input", "This is a pair")
        p2 = [
            ("This is a simple input 1", "This is a simple input 2"),
            ("This is a simple pair 1", "This is a simple pair 2"),
        ]
849

Funtowicz Morgan's avatar
Funtowicz Morgan committed
850
        # Simple input tests
851
        self.assertRaises(ValueError, tokenizer_r.encode, s, max_length=max_length, padding="max_length")
852

Funtowicz Morgan's avatar
Funtowicz Morgan committed
853
        # Simple input
854
        self.assertRaises(ValueError, tokenizer_r.encode_plus, s, max_length=max_length, padding="max_length")
855

Funtowicz Morgan's avatar
Funtowicz Morgan committed
856
        # Simple input
857
858
859
        self.assertRaises(
            ValueError, tokenizer_r.batch_encode_plus, s2, max_length=max_length, padding="max_length",
        )
860

Funtowicz Morgan's avatar
Funtowicz Morgan committed
861
        # Pair input
862
        self.assertRaises(ValueError, tokenizer_r.encode, p, max_length=max_length, padding="max_length")
863

Funtowicz Morgan's avatar
Funtowicz Morgan committed
864
        # Pair input
865
        self.assertRaises(ValueError, tokenizer_r.encode_plus, p, max_length=max_length, padding="max_length")
866

Funtowicz Morgan's avatar
Funtowicz Morgan committed
867
        # Pair input
868
869
870
        self.assertRaises(
            ValueError, tokenizer_r.batch_encode_plus, p2, max_length=max_length, padding="max_length",
        )
871

872

Funtowicz Morgan's avatar
Funtowicz Morgan committed
873
874
class TransfoXLFastTokenizerTest(NoPaddingTokenFastTokenizerMatchingTest):
    TOKENIZERS_CLASSES = frozenset(
875
        [Tokenizer("TransfoXL", TransfoXLTokenizerFast, TransfoXLTokenizer, "pretrained_vocab_file", None, None)]
Funtowicz Morgan's avatar
Funtowicz Morgan committed
876
    )
877

Funtowicz Morgan's avatar
Funtowicz Morgan committed
878
879
880
    @require_torch
    def test_all_tokenizers(self):
        super().test_all_tokenizers()
881
882
883
884

    @require_torch
    def test_pretokenized_tokenizers(self):
        super().test_pretokenized_tokenizers()