mirror of
https://github.com/huggingface/transformers.git
synced 2025-07-04 13:20:12 +06:00
866df66fe4
* First commit while I figure this out * make fixup * Remove unused method * Store prompt attrib * Fix prompt argument for tests * Make same changes in fast tokenizer * Remove global prompts from fast tokenizer too * stash commit * stash commit * Migrate PromptConfig to its True Final Location * Replace Conversation entirely with the new class * Import/dependency fixes * Import/dependency fixes * Change format for lots of default prompts * More default prompt fixups * Revert llama old methods so we can compare * Fix some default configs * Fix some default configs * Fix misspelled kwarg * Fixes for Blenderbot * make fixup * little rebase cleanup * Add basic documentation * Quick doc fix * Truncate docstring for now * Add handling for the case when messages is a single string * Quick llama merges * Update conversational pipeline and tests * Add a couple of legacy properties for backward compatibility * More legacy handling * Add docstring for build_conversation_input_ids * Restructure PromptConfig * Let's start T E M P L A T I N G * Refactor all default configs to use templates instead * Revert changes to the special token properties since we don't need them anymore * More class templates * Make the sandbox even sandier * Everything replaced with pure templating * Remove docs for PromptConfig * Add testing and optional requirement boilerplate * Fix imports and make fixup * Fix LLaMA tests and add Conversation docstring * Finally get LLaMA working with the template system * Finally get LLaMA working with the template system * make fixup * make fixup * fmt-off for the long lists of test tokens * Rename method to apply_chat_template for now * Start on documentation * Make chat_template a property that reads through to the default if it's not set * Expand docs * Expand chat templating doc some more * trim/lstrip blocks by default and update doc * Few doc tweaks * rebase cleanup * Clarify docstring * rebase cleanup * rebase cleanup * make fixup * Quick doc edit * Reformat the standard template to match ChatML * Re-add PEFT check * Update docs/source/en/chat_templating.md Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com> * Add apply_chat_template to the tokenizer doc * make fixup * Add doc links * Fix chat links * Fix chat links * Explain system messages in the doc * Add chat template test * Proper save-loading for chat template attribute * Add test skips for layout models * Remove _build_conversation_input_ids, add default_chat_template to code_llama * Make sure all LLaMA models are using the latest template * Remove default_system_prompt block in code_llama because it has no default prompt * Update ConversationPipeline preprocess * Add correct #Copied from links to the default_chat_templates * Remove unneeded type checking line * Add a dummy mark_processsed method * Reorganize Conversation to have **deprecated_kwargs * Update chat_templating.md * Quick fix to LLAMA tests * Small doc tweaks * Add proper docstrings and "copied from" statements to all default chat templates * Merge use_default_system_prompt support for code_llama too * Improve clarity around self.chat_template * Docstring fix * Fix blenderbot default template * More doctest fix * Break out some tokenizer kwargs * Update doc to explain default templates * Quick tweaks to tokenizer args * Cleanups for tokenizer args * Add note about cacheing * Quick tweak to the chat-templating doc * Update the LLaMA template with error checking and correct system message embedding * make fixup * make fixup * add requires_jinja * Cleanup to expected output formatting * Add cacheing * Fix typo in llama default template * Update LLaMA tests * Update documentation * Improved legacy handling in the Conversation class * Update Jinja template with proper error handling * Quick bugfix * Proper exception raising * Change cacheing behaviour so it doesn't try to pickle an entire Jinja env * make fixup * rebase cleanup --------- Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
155 lines
8.0 KiB
Python
155 lines
8.0 KiB
Python
# coding=utf-8
|
|
# Copyright 2022 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 GPTSw3Tokenizer
|
|
from transformers.testing_utils import get_tests_dir, require_jinja, require_sentencepiece, require_tokenizers, slow
|
|
|
|
from ...test_tokenization_common import TokenizerTesterMixin
|
|
|
|
|
|
SAMPLE_VOCAB = get_tests_dir("fixtures/test_sentencepiece_with_bytefallback.model")
|
|
|
|
|
|
@require_sentencepiece
|
|
@require_tokenizers
|
|
class GPTSw3TokenizationTest(TokenizerTesterMixin, unittest.TestCase):
|
|
tokenizer_class = GPTSw3Tokenizer
|
|
test_rust_tokenizer = False
|
|
test_sentencepiece = True
|
|
test_sentencepiece_ignore_case = False
|
|
|
|
def setUp(self):
|
|
super().setUp()
|
|
|
|
# We have a SentencePiece fixture for testing
|
|
tokenizer = GPTSw3Tokenizer(SAMPLE_VOCAB, eos_token="<unk>", bos_token="<unk>", pad_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 = "<s>"
|
|
token_id = 1
|
|
|
|
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], "<unk>")
|
|
self.assertEqual(vocab_keys[1], "<s>")
|
|
self.assertEqual(vocab_keys[-1], "j")
|
|
self.assertEqual(len(vocab_keys), 2_000)
|
|
|
|
def test_vocab_size(self):
|
|
self.assertEqual(self.get_tokenizer().vocab_size, 2_000)
|
|
|
|
def test_full_tokenizer(self):
|
|
tokenizer = GPTSw3Tokenizer(SAMPLE_VOCAB)
|
|
|
|
tokens = tokenizer.tokenize("This is a test")
|
|
self.assertListEqual(tokens, ["▁This", "▁is", "▁a", "▁t", "est"])
|
|
|
|
self.assertListEqual(tokenizer.convert_tokens_to_ids(tokens), [465, 287, 265, 631, 842])
|
|
|
|
tokens = tokenizer.tokenize("I was born in 92000, and this is falsé.")
|
|
# fmt: off
|
|
self.assertListEqual(
|
|
tokens,
|
|
["▁I", "▁was", "▁bor", "n", "▁in", "▁", "<0x39>", "2", "0", "0", "0", ",", "▁and", "▁this", "▁is", "▁f", "al", "s", "<0xC3>", "<0xA9>", "."],
|
|
)
|
|
# fmt: on
|
|
|
|
ids = tokenizer.convert_tokens_to_ids(tokens)
|
|
self.assertListEqual(
|
|
ids,
|
|
[262, 272, 1525, 286, 271, 268, 60, 916, 633, 633, 633, 259, 266, 301, 287, 384, 367, 263, 198, 172, 260],
|
|
)
|
|
|
|
back_tokens = tokenizer.convert_ids_to_tokens(ids)
|
|
# fmt: off
|
|
self.assertListEqual(
|
|
back_tokens,
|
|
["▁I", "▁was", "▁bor", "n", "▁in", "▁", "<0x39>", "2", "0", "0", "0", ",", "▁and", "▁this", "▁is", "▁f", "al", "s", "<0xC3>", "<0xA9>", "."]
|
|
)
|
|
# fmt: on
|
|
|
|
def test_fast_encode_decode(self):
|
|
tokenizer = GPTSw3Tokenizer(SAMPLE_VOCAB)
|
|
texts = ["This is a test", "I was born in 92000, and this is falsé."]
|
|
expected_ids_list = [
|
|
[465, 287, 265, 631, 842],
|
|
[262, 272, 1525, 286, 271, 268, 60, 916, 633, 633, 633, 259, 266, 301, 287, 384, 367, 263, 198, 172, 260],
|
|
]
|
|
|
|
# Test that encode_fast returns the same as tokenize + convert_tokens_to_ids
|
|
for text, expected_ids in zip(texts, expected_ids_list):
|
|
self.assertListEqual(tokenizer.encode_fast(text), expected_ids)
|
|
|
|
# Test that decode_fast returns the input text
|
|
for text, token_ids in zip(texts, expected_ids_list):
|
|
self.assertEqual(tokenizer.decode_fast(token_ids), text)
|
|
|
|
@slow
|
|
def test_tokenizer_integration(self):
|
|
sequences = [
|
|
"<|python|>def fibonacci(n)\n if n < 0:\n print('Incorrect input')",
|
|
"Hey there, how are you doing this fine day?",
|
|
"This is a text with a trailing spaces followed by a dot .",
|
|
"Häj sväjs lillebrör! =)",
|
|
"Det är inget fel på Mr. Cool",
|
|
]
|
|
|
|
# fmt: off
|
|
expected_encoding = {"input_ids": [[63423, 5, 6811, 14954, 282, 816, 3821, 63466, 63425, 63462, 18, 63978, 678, 301, 1320, 63423, 63455, 63458, 18, 63982, 4246, 3940, 1901, 47789, 5547, 18994], [19630, 1100, 63446, 1342, 633, 544, 4488, 593, 5102, 2416, 63495, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], [1652, 428, 268, 1936, 515, 268, 58593, 22413, 9106, 546, 268, 33213, 63979, 698, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], [55130, 63450, 924, 63449, 2249, 4062, 1558, 318, 63504, 21498, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], [509, 377, 2827, 2559, 332, 6575, 63443, 26801, 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]], "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, 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], [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], [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]]}
|
|
# fmt: on
|
|
self.tokenizer_integration_test_util(
|
|
expected_encoding=expected_encoding,
|
|
model_name="AI-Sweden/gpt-sw3-126m",
|
|
sequences=sequences,
|
|
)
|
|
|
|
@require_jinja
|
|
def test_tokenization_for_chat(self):
|
|
tokenizer = GPTSw3Tokenizer(SAMPLE_VOCAB)
|
|
# This is in English, but it's just here to make sure the chat control tokens are being added properly
|
|
test_chats = [
|
|
[{"role": "system", "content": "You are a helpful chatbot."}, {"role": "user", "content": "Hello!"}],
|
|
[
|
|
{"role": "system", "content": "You are a helpful chatbot."},
|
|
{"role": "user", "content": "Hello!"},
|
|
{"role": "assistant", "content": "Nice to meet you."},
|
|
],
|
|
[{"role": "assistant", "content": "Nice to meet you."}, {"role": "user", "content": "Hello!"}],
|
|
]
|
|
tokenized_chats = [tokenizer.apply_chat_template(test_chat) for test_chat in test_chats]
|
|
# fmt: off
|
|
expected_tokens = [
|
|
[268, 63, 127, 462, 276, 294, 348, 536, 797, 275, 127, 65, 63, 263, 65, 938, 541, 419, 530, 339, 265, 878, 708, 727, 275, 347, 541, 260, 63, 263, 65, 1256, 263, 314, 419, 366, 354, 294, 360, 63, 263, 65, 938, 541, 419, ],
|
|
[268, 63, 127, 462, 276, 294, 348, 536, 797, 275, 127, 65, 63, 263, 65, 938, 541, 419, 530, 339, 265, 878, 708, 727, 275, 347, 541, 260, 63, 263, 65, 1256, 263, 314, 419, 366, 354, 294, 360, 63, 263, 65, 938, 541, 419, 984, 429, 281, 264, 1261, 291, 260, 63, 263, 65, 938, 541, 419, ],
|
|
[268, 63, 127, 462, 276, 294, 348, 536, 797, 275, 127, 65, 63, 263, 65, 938, 541, 419, 984, 429, 281, 264, 1261, 291, 260, 63, 263, 65, 1256, 263, 314, 419, 366, 354, 294, 360, 63, 263, 65, 938, 541, 419, ]
|
|
]
|
|
# fmt: on
|
|
for tokenized_chat, expected_tokens in zip(tokenized_chats, expected_tokens):
|
|
self.assertListEqual(tokenized_chat, expected_tokens)
|