"""Tokenization classes for InternLM."""
import os
from shutil import copyfile
from typing import Any, Dict, List, Optional, Tuple
import sentencepiece as spm
from mindformers.models.tokenization_utils_base import AddedToken
from mindformers.tools.logger import logger
from mindformers.models.tokenization_utils import PreTrainedTokenizer
from mindformers.tools.register import MindFormerRegister, MindFormerModuleType
from mindformers.tools.utils import check_file
VOCAB_FILES_NAMES = {"vocab_file": "./tokenizer.model"}
PRETRAINED_VOCAB_FILES_MAP = {}
@MindFormerRegister.register(MindFormerModuleType.TOKENIZER)
class InternLM2Tokenizer(PreTrainedTokenizer):
"""
Construct a InternLM2 tokenizer. Based on byte-level Byte-Pair-Encoding.
Args:
vocab_file (`str`):
Path to the vocabulary file.
"""
vocab_files_names = VOCAB_FILES_NAMES
pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
model_input_names = ["input_ids", "attention_mask"]
_auto_class = "AutoTokenizer"
def __init__(
self,
vocab_file,
unk_token="<unk>",
bos_token="<s>",
eos_token="</s>",
pad_token="</s>",
sp_model_kwargs: Optional[Dict[str, Any]] = None,
add_bos_token=True,
add_eos_token=False,
decode_with_prefix_space=False,
clean_up_tokenization_spaces=False,
**kwargs,
):
self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
self.vocab_file = vocab_file
self.add_bos_token = add_bos_token
self.add_eos_token = add_eos_token
self.decode_with_prefix_space = decode_with_prefix_space
self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
check_file(vocab_file, "tokenizer")
self.sp_model.Load(vocab_file)
self._no_prefix_space_tokens = None
if kwargs.get("chat_template") is None:
kwargs["chat_template"] = "{{ bos_token }}" \
"{% for message in messages %}" \
"{{'<|im_start|>' + message['role'] + '\n' + " \
"message['content'] + '<|im_end|>' + '\n'}}" \
"{% endfor %}" \
"{% if add_generation_prompt %}" \
"{{ '<|im_start|>assistant\n' }}" \
"{% endif %}"
if kwargs.get("added_tokens_decoder") is None:
kwargs["added_tokens_decoder"] = {}
kwargs["added_tokens_decoder"].update({
0: AddedToken("<unk>", special=True),
1: AddedToken("<s>", special=True),
2: AddedToken("</s>", special=True),
92538: AddedToken("<|plugin|>", special=True),
92539: AddedToken("<|interpreter|>", special=True),
92540: AddedToken("<|action_end|>", special=True),
92541: AddedToken("<|action_start|>", special=True),
92542: AddedToken("<|im_end|>", special=True),
92543: AddedToken("<|im_start|>", special=True)})
super().__init__(
bos_token=bos_token,
eos_token=eos_token,
unk_token=unk_token,
pad_token=pad_token,
clean_up_tokenization_spaces=clean_up_tokenization_spaces,
**kwargs,
)
self.additional_special_tokens = ["<|im_start|>",
"<|im_end|>",
"<|action_start|>",
"<|action_end|>",
"<|interpreter|>",
"<|plugin|>"]
@property
def no_prefix_space_tokens(self):
if self._no_prefix_space_tokens is None:
vocab = self.convert_ids_to_tokens(list(range(self.vocab_size)))
self._no_prefix_space_tokens = {tok for tok in vocab if not tok.startswith("▁")}
return self._no_prefix_space_tokens
@property
def vocab_size(self):
"""Returns vocab size"""
return self.sp_model.get_piece_size()
@property
def bos_token_id(self) -> Optional[int]:
return self.sp_model.bos_id()
@property
def eos_token_id(self) -> Optional[int]:
return self.sp_model.eos_id()
def get_vocab(self):
"""Returns vocab as a dict"""
vocab = {self.convert_ids_to_tokens(i): i for i in range(self.vocab_size)}
vocab.update(self.added_tokens_encoder)
return vocab
def _tokenize(self, text):
"""Returns a tokenized string."""
return self.sp_model.encode(text, out_type=str)
def _convert_token_to_id(self, token):
"""Converts a token (str) in an id using the vocab."""
return self.sp_model.piece_to_id(token)
def _convert_id_to_token(self, index):
"""Converts an index (integer) in a token (str) using the vocab."""
token = self.sp_model.IdToPiece(index)
return token
def _maybe_add_prefix_space(self, tokens, decoded):
if tokens and tokens[0] not in self.no_prefix_space_tokens:
return " " + decoded
return decoded
def convert_tokens_to_string(self, tokens):
"""Converts a sequence of tokens (string) in a single string."""
current_sub_tokens = []
out_string = ""
for token in tokens:
if token in self.all_special_tokens:
out_string += self.sp_model.decode(current_sub_tokens) + token
current_sub_tokens = []
else:
current_sub_tokens.append(token)
out_string += self.sp_model.decode(current_sub_tokens)
out_string = self.clean_up_tokenization(out_string)
out_string = self._maybe_add_prefix_space(tokens=tokens, decoded=out_string)
return out_string
def save_vocabulary(self, save_directory, filename_prefix: Optional[str] = None) -> Tuple[str]:
"""
Save the vocabulary and special tokens file to a directory.
Args:
save_directory (`str`):
The directory in which to save the vocabulary.
Returns:
`Tuple(str)`: Paths to the files saved.
"""
if not os.path.isdir(save_directory):
logger.error(f"Vocabulary path ({save_directory}) should be a directory")
return ("",)
out_vocab_file = os.path.join(
save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"]
)
if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file) and os.path.isfile(self.vocab_file):
copyfile(self.vocab_file, out_vocab_file)
elif not os.path.isfile(self.vocab_file):
flags_ = os.O_WRONLY | os.O_CREAT | os.O_TRUNC
with os.fdopen(os.open(out_vocab_file, flags_, 0o750), "wb") as fi:
content_spiece_model = self.sp_model.serialized_model_proto()
fi.write(content_spiece_model)
return (out_vocab_file,)
def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
"""Build inputs with special tokens"""
if self.add_bos_token:
bos_token_ids = [self.bos_token_id]
else:
bos_token_ids = []
output = bos_token_ids + token_ids_0
if token_ids_1 is not None:
output = output + token_ids_1
if self.add_eos_token:
output = output + [self.eos_token_id]
return output
def get_special_tokens_mask(
self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None,
already_has_special_tokens: bool = False
) -> List[int]:
"""
Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding
special tokens using the tokenizer `prepare_for_model` method.
Args:
token_ids_0 (`List[int]`):
List of IDs.
token_ids_1 (`List[int]`, *optional*):
Optional second list of IDs for sequence pairs.
already_has_special_tokens (`bool`, *optional*, defaults to `False`):
Whether or not the token list is already formatted with special tokens for the model.
Returns:
`List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
"""
if already_has_special_tokens:
return super().get_special_tokens_mask(
token_ids_0=token_ids_0, token_ids_1=token_ids_1, already_has_special_tokens=True
)
if token_ids_1 is None:
return [1] + ([0] * len(token_ids_0)) + [1]
return [1] + ([0] * len(token_ids_0)) + [1, 1] + ([0] * len(token_ids_1)) + [1]
def create_token_type_ids_from_sequences(
self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
) -> List[int]:
"""
Create a mask from the two sequences passed to be used in a sequence-pair classification task. T5 does not make
use of token type ids, therefore a list of zeros is returned.
Args:
token_ids_0 (`List[int]`):
List of IDs.
token_ids_1 (`List[int]`, *optional*):
Optional second list of IDs for sequence pairs.
Returns:
`List[int]`: List of zeros.
"""
eos = [self.eos_token_id]
if token_ids_1 is None:
return len(token_ids_0 + eos) * [0]
return len(token_ids_0 + eos + token_ids_1 + eos) * [0]