test_tokenization_deberta_v2.py

# 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 os
import unittest

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

from .test_tokenization_common import TokenizerTesterMixin


SAMPLE_VOCAB = os.path.join(os.path.dirname(os.path.abspath(__file__)), "fixtures/spiece.model")


@require_sentencepiece
@require_tokenizers
class DebertaV2TokenizationTest(TokenizerTesterMixin, unittest.TestCase):

    tokenizer_class = DebertaV2Tokenizer
    rust_tokenizer_class = None
    test_rust_tokenizer = False
    test_sentencepiece = True
    test_sentencepiece_ignore_case = True

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

        # We have a SentencePiece fixture for testing
        tokenizer = DebertaV2Tokenizer(SAMPLE_VOCAB)
        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_rust_and_python_full_tokenizers(self):
        if not self.test_rust_tokenizer:
            return

        tokenizer = self.get_tokenizer()
        rust_tokenizer = self.get_rust_tokenizer()

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

        tokens = tokenizer.tokenize(sequence)
        rust_tokens = rust_tokenizer.tokenize(sequence)
        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):
        tokenizer = DebertaV2Tokenizer(SAMPLE_VOCAB, keep_accents=True)

        tokens = tokenizer.tokenize("This is a test")
        self.assertListEqual(tokens, ["▁", "[UNK]", "his", "▁is", "▁a", "▁test"])

        self.assertListEqual(tokenizer.convert_tokens_to_ids(tokens), [13, 1, 4398, 25, 21, 1289])

        tokens = tokenizer.tokenize("I was born in 92000, and this is falsé.")
        # fmt: off
        self.assertListEqual(
            tokens,
            ["▁", "[UNK]", "▁was", "▁born", "▁in", "▁9", "2000", ",", "▁and", "▁this", "▁is", "▁fal", "s", "[UNK]", "."],
        )
        ids = tokenizer.convert_tokens_to_ids(tokens)
        self.assertListEqual(ids, [13, 1, 23, 386, 19, 561, 3050, 15, 17, 48, 25, 8256, 18, 1, 9])

        back_tokens = tokenizer.convert_ids_to_tokens(ids)
        self.assertListEqual(
            back_tokens,
            ["▁", "<unk>", "▁was", "▁born", "▁in", "▁9", "2000", ",", "▁and", "▁this", "▁is", "▁fal", "s", "<unk>", "."],
        )
        # fmt: on

    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):
        # fmt: off
        expected_encoding = {'input_ids': [[1, 32732, 36, 19390, 486, 27, 35052, 81436, 18, 60685, 1225, 7, 35052, 81436, 18, 9367, 16899, 18, 15937, 53, 594, 773, 18, 16287, 30465, 36, 69418, 6, 107805, 36979, 10993, 69418, 191, 6, 12692, 829, 6, 8655, 16555, 92459, 6, 12692, 9431, 20850, 14, 4184, 6369, 9875, 36, 1323, 23941, 53, 7, 4184, 6369, 11005, 36, 20582, 1186, 53, 19, 105, 3049, 1896, 1053, 16899, 1506, 11, 37978, 4243, 7, 1237, 31869, 200, 42754, 6, 19645, 45050, 3425, 7, 107535, 4, 2], [1, 448, 37132, 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], [1, 23, 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]], '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]], '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, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 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]]}  # noqa: E501
        # fmt: on

        self.tokenizer_integration_test_util(
            expected_encoding=expected_encoding,
            model_name="microsoft/deberta-v2-xlarge",
            revision="ad6e42c1532ddf3a15c39246b63f5559d558b670",
        )