wehub-resource-sync
200 lines
8.0 KiB
Python
200 lines
8.0 KiB
Python
# Copyright (c) 2023 PaddlePaddle Authors. All Rights Reserved.
|
|
# Copyright 2020 The HuggingFace Team. All rights reserved.
|
|
#
|
|
# 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 json
|
|
import os
|
|
import shutil
|
|
import tempfile
|
|
import unittest
|
|
|
|
from paddlenlp.transformers import LukeTokenizer
|
|
|
|
from ..test_tokenizer_common import TokenizerTesterMixin
|
|
|
|
VOCAB_FILES_NAMES = LukeTokenizer.resource_files_names
|
|
|
|
|
|
class TestTokenizationLuke(TokenizerTesterMixin, unittest.TestCase):
|
|
tokenizer_class = LukeTokenizer
|
|
test_offsets = False
|
|
|
|
def setUp(self):
|
|
super().setUp()
|
|
|
|
# Adapted from Sennrich et al. 2015 and https://github.com/rsennrich/subword-nmt
|
|
vocab = [
|
|
"l",
|
|
"o",
|
|
"w",
|
|
"e",
|
|
"r",
|
|
"s",
|
|
"t",
|
|
"i",
|
|
"d",
|
|
"n",
|
|
"\u0120",
|
|
"\u0120l",
|
|
"\u0120n",
|
|
"\u0120lo",
|
|
"\u0120low",
|
|
"er",
|
|
"\u0120lowest",
|
|
"\u0120newer",
|
|
"\u0120wider",
|
|
"<unk>",
|
|
"</s>",
|
|
"<pad>",
|
|
"<s>",
|
|
"<mask>",
|
|
]
|
|
vocab_tokens = dict(zip(vocab, range(len(vocab))))
|
|
merges = ["#version: 0.2", "\u0120 l", "\u0120l o", "\u0120lo w", "e r", ""]
|
|
self.special_tokens_map = {"unk_token": "<unk>"}
|
|
|
|
entity_vocab = {"[PAD]": 0, "[UNK]": 1, "[MASK]": 2, "[MASK2]": 3}
|
|
|
|
self.vocab_file = os.path.join(self.tmpdirname, VOCAB_FILES_NAMES["vocab_file"])
|
|
self.merges_file = os.path.join(self.tmpdirname, VOCAB_FILES_NAMES["merges_file"])
|
|
self.entity_file = os.path.join(self.tmpdirname, VOCAB_FILES_NAMES["entity_file"])
|
|
|
|
with open(self.vocab_file, "w", encoding="utf-8") as fp:
|
|
fp.write(json.dumps(vocab_tokens) + "\n")
|
|
with open(self.merges_file, "w", encoding="utf-8") as fp:
|
|
fp.write("\n".join(merges))
|
|
with open(self.entity_file, "w", encoding="utf-8") as fp:
|
|
fp.write(json.dumps(entity_vocab))
|
|
|
|
def test_add_special_tokens(self):
|
|
tokenizers = self.get_tokenizers(do_lower_case=False)
|
|
for tokenizer in tokenizers:
|
|
with self.subTest(f"{tokenizer.__class__.__name__}"):
|
|
input_text, ids = self.get_clean_sequence(tokenizer)
|
|
|
|
special_token = "[SPECIAL_TOKEN]"
|
|
|
|
tokenizer.add_special_tokens({"additional_special_tokens": special_token})
|
|
encoded_special_token = tokenizer.encode(
|
|
special_token, return_token_type_ids=None, add_special_tokens=False
|
|
)["input_ids"]
|
|
self.assertEqual(len(encoded_special_token), len(special_token))
|
|
|
|
text = tokenizer.decode(ids + encoded_special_token, clean_up_tokenization_spaces=False)
|
|
encoded = tokenizer.encode(text, return_token_type_ids=None, add_special_tokens=False)["input_ids"]
|
|
|
|
input_encoded = tokenizer.encode(input_text, return_token_type_ids=None, add_special_tokens=False)[
|
|
"input_ids"
|
|
]
|
|
special_token_id = tokenizer.encode(
|
|
special_token, return_token_type_ids=None, add_special_tokens=False
|
|
)["input_ids"]
|
|
self.assertEqual(encoded, input_encoded + special_token_id)
|
|
|
|
decoded = tokenizer.decode(encoded, skip_special_tokens=True)
|
|
self.assertTrue(special_token not in decoded)
|
|
|
|
def test_tokenize_special_tokens(self):
|
|
"""Test `tokenize` with special tokens."""
|
|
tokenizers = self.get_tokenizers(do_lower_case=True)
|
|
for tokenizer in tokenizers:
|
|
with self.subTest(f"{tokenizer.__class__.__name__}"):
|
|
SPECIAL_TOKEN_1 = "[SPECIAL_TOKEN_1]"
|
|
SPECIAL_TOKEN_2 = "[SPECIAL_TOKEN_2]"
|
|
tokenizer.add_tokens([SPECIAL_TOKEN_1], special_tokens=True)
|
|
tokenizer.add_special_tokens({"additional_special_tokens": [SPECIAL_TOKEN_2]})
|
|
|
|
token_1 = tokenizer.tokenize(SPECIAL_TOKEN_1)
|
|
token_2 = tokenizer.tokenize(SPECIAL_TOKEN_2)
|
|
|
|
self.assertEqual(len(token_1), len(SPECIAL_TOKEN_1))
|
|
self.assertEqual(len(token_2), 1)
|
|
self.assertEqual(token_1[0], SPECIAL_TOKEN_1[0])
|
|
self.assertEqual(token_2[0], SPECIAL_TOKEN_2)
|
|
|
|
def test_consecutive_unk_string(self):
|
|
tokenizers = self.get_tokenizers(fast=True, do_lower_case=True)
|
|
for tokenizer in tokenizers:
|
|
tokens = [tokenizer.unk_token for _ in range(2)]
|
|
string = tokenizer.convert_tokens_to_string(tokens)
|
|
encoding = tokenizer(
|
|
text=string,
|
|
)
|
|
self.assertEqual(len(encoding["input_ids"]), 4)
|
|
|
|
def test_save_and_load_tokenizer(self):
|
|
# safety check on max_len default value so we are sure the test works
|
|
tokenizers = self.get_tokenizers()
|
|
for tokenizer in tokenizers:
|
|
with self.subTest(f"{tokenizer.__class__.__name__}"):
|
|
self.assertNotEqual(tokenizer.model_max_length, 42)
|
|
|
|
# Now let's start the test
|
|
tokenizers = self.get_tokenizers()
|
|
for tokenizer in tokenizers:
|
|
with self.subTest(f"{tokenizer.__class__.__name__}"):
|
|
# Isolate this from the other tests because we save additional tokens/etc
|
|
tmpdirname = tempfile.mkdtemp()
|
|
|
|
sample_text = " He is very happy, UNwant\u00E9d,running"
|
|
before_tokens = tokenizer.encode(sample_text, add_special_tokens=False)
|
|
before_vocab = tokenizer.get_vocab()
|
|
tokenizer.save_pretrained(tmpdirname)
|
|
|
|
after_tokenizer = tokenizer.__class__.from_pretrained(tmpdirname)
|
|
after_tokens = after_tokenizer.encode(sample_text, add_special_tokens=False)
|
|
after_vocab = after_tokenizer.get_vocab()
|
|
self.assertListEqual(before_tokens["input_ids"], after_tokens["input_ids"])
|
|
self.assertEqual(before_vocab.keys(), after_vocab.keys())
|
|
|
|
shutil.rmtree(tmpdirname)
|
|
|
|
def test_conversion_reversible(self):
|
|
tokenizers = self.get_tokenizers(do_lower_case=False)
|
|
for tokenizer in tokenizers:
|
|
with self.subTest(f"{tokenizer.__class__.__name__}"):
|
|
vocab = tokenizer.get_vocab()
|
|
for word, ind in vocab.items():
|
|
if word == tokenizer.unk_token:
|
|
continue
|
|
self.assertEqual(tokenizer.encoder[word], ind)
|
|
self.assertEqual(tokenizer.convert_ids_to_tokens(ind), word)
|
|
|
|
def test_call(self):
|
|
self.skipTest("Direct call is not the same as encode")
|
|
|
|
def test_tokenizers_common_ids_setters(self):
|
|
self.skipTest("Add token not implement yet")
|
|
|
|
def test_add_tokens(self):
|
|
self.skipTest("Add token not implement yet")
|
|
|
|
def test_add_tokens_tokenizer(self):
|
|
self.skipTest("Add token not implement yet")
|
|
|
|
def test_added_token_serializable(self):
|
|
self.skipTest("Add token not implement yet")
|
|
|
|
def test_added_tokens_do_lower_case(self):
|
|
self.skipTest("Add token not implement yet")
|
|
|
|
def test_added_token_are_matched_longest_first(self):
|
|
self.skipTest("Add token not implement yet")
|
|
|
|
def test_special_tokens_initialization_with_non_empty_additional_special_tokens(self):
|
|
self.skipTest("Add token not implement yet")
|
|
|
|
def test_encode_decode_with_spaces(self):
|
|
self.skipTest("Add token not implement yet")
|