tokenization_chatglm.py

import json
import os
import re
from typing import List, Optional, Union, Dict
from sentencepiece import SentencePieceProcessor
from transformers import PreTrainedTokenizer
from transformers.utils import logging, PaddingStrategy
from transformers.tokenization_utils_base import EncodedInput, BatchEncoding


logger = logging.get_logger(__name__)


class SPTokenizer:
    def __init__(self, model_path: str):
        # reload tokenizer
        assert os.path.isfile(model_path), model_path
        self.sp_model = SentencePieceProcessor(model_file=model_path)

        # BOS / EOS token IDs
        self.n_words: int = self.sp_model.vocab_size()
        self.bos_id: int = self.sp_model.bos_id()
        self.eos_id: int = self.sp_model.eos_id()
        self.pad_id: int = self.sp_model.unk_id()
        assert self.sp_model.vocab_size() == self.sp_model.get_piece_size()

        role_special_tokens = ["<|system|>", "<|user|>", "<|assistant|>", "<|observation|>"]
        special_tokens = ["[MASK]", "[gMASK]", "[sMASK]", "sop", "eop"] + role_special_tokens
        self.special_tokens = {}
        self.index_special_tokens = {}
        for token in special_tokens:
            self.special_tokens[token] = self.n_words
            self.index_special_tokens[self.n_words] = token
            self.n_words += 1
        self.role_special_token_expression = "|".join([re.escape(token) for token in special_tokens]) # for apply_chat_template

    def tokenize(self, s: str, encode_special_tokens=False):
        if encode_special_tokens:
            last_index = 0
            t = []
            for match in re.finditer(self.role_special_token_expression, s):
                if last_index < match.start():
                    t.extend(self.sp_model.EncodeAsPieces(s[last_index:match.start()]))
                t.append(s[match.start():match.end()])
                last_index = match.end()
            if last_index < len(s):
                t.extend(self.sp_model.EncodeAsPieces(s[last_index:]))
            return t
        else:
            return self.sp_model.EncodeAsPieces(s)

    def encode(self, s: str, bos: bool = False, eos: bool = False) -> List[int]:
        assert type(s) is str
        t = self.sp_model.encode(s)
        if bos:
            t = [self.bos_id] + t
        if eos:
            t = t + [self.eos_id]
        return t

    def decode(self, t: List[int]) -> str:
        text, buffer = "", []
        for token in t:
            if token in self.index_special_tokens:
                if buffer:
                    text += self.sp_model.decode(buffer)
                    buffer = []
                text += self.index_special_tokens[token]
            else:
                buffer.append(token)
        if buffer:
            text += self.sp_model.decode(buffer)
        return text

    def decode_tokens(self, tokens: List[str]) -> str:
        text = self.sp_model.DecodePieces(tokens)
        return text

    def convert_token_to_id(self, token):
        """ Converts a token (str) in an id using the vocab. """
        if token in self.special_tokens:
            return self.special_tokens[token]
        return self.sp_model.PieceToId(token)

    def convert_id_to_token(self, index):
        """Converts an index (integer) in a token (str) using the vocab."""
        if index in self.index_special_tokens:
            return self.index_special_tokens[index]
        if index in [self.eos_id, self.bos_id, self.pad_id] or index < 0 or index > self.sp_model.vocab_size():
            return ""
        return self.sp_model.IdToPiece(index)


class ChatGLMTokenizer(PreTrainedTokenizer):

    vocab_files_names = {"vocab_file": "tokenizer.model"}
    model_input_names = ["input_ids", "attention_mask", "position_ids"]

    def __init__(
        self,
        vocab_file,
        padding_side="left",
        clean_up_tokenization_spaces=False,
        encode_special_tokens=False,
        **kwargs
    ):
        self.name = "GLMTokenizer"
        self.vocab_file = vocab_file
        self.tokenizer = SPTokenizer(vocab_file)
        self.special_tokens = {
            "<bos>": self.tokenizer.bos_id,
            "<eos>": self.tokenizer.eos_id,
            "<unk>": self.tokenizer.pad_id,
            "<pad>": self.tokenizer.pad_id
        }
        self.encode_special_tokens = encode_special_tokens

        super().__init__(
            padding_side=padding_side,
            clean_up_tokenization_spaces=clean_up_tokenization_spaces,
            **kwargs
        )

    def get_command(self, token):
        if token in self.special_tokens:
            return self.special_tokens[token]
        assert token in self.tokenizer.special_tokens, f"{token} is not a special token for {self.name}"
        return self.tokenizer.special_tokens[token]

    @property
    def unk_token(self) -> str:
        return self.tokenizer.sp_model.IdToPiece(self.get_command("<unk>"))

    @property
    def pad_token(self) -> str:
        return self.tokenizer.sp_model.IdToPiece(self.get_command("<pad>"))

    @property
    def eos_token(self) -> str:
        return self.tokenizer.sp_model.IdToPiece(self.get_command("<eos>"))

    @property
    def unk_token_id(self) -> int:
        return self.get_command("<unk>")

    @property
    def pad_token_id(self) -> int:
        return self.get_command("<pad>")

    @property
    def eos_token_id(self):
        return self.get_command("<eos>")

    @unk_token.setter
    def unk_token(self, value):
        logger.warning("Setting unk_token is not supported, use the default one.")

    @pad_token.setter
    def pad_token(self, value):
        logger.warning("Setting pad_token is not supported, use the default one.")

    @eos_token.setter
    def eos_token(self, value):
        logger.warning("Setting eos_token is not supported, use the default one.")

    @property
    def vocab_size(self):
        return self.tokenizer.n_words

    def get_vocab(self):
        """ Returns vocab as a dict """
        vocab = {self._convert_id_to_token(i): i for i in range(self.vocab_size)}
        vocab.update(self.added_tokens_encoder)
        return vocab

    def _tokenize(self, text, **kwargs):
        return self.tokenizer.tokenize(text, encode_special_tokens=self.encode_special_tokens)

    def _convert_token_to_id(self, token):
        """ Converts a token (str) in an id using the vocab. """
        return self.tokenizer.convert_token_to_id(token)

    def _convert_id_to_token(self, index):
        """Converts an index (integer) in a token (str) using the vocab."""
        return self.tokenizer.convert_id_to_token(index)

    def convert_tokens_to_string(self, tokens: List[str]) -> str:
        return self.tokenizer.decode_tokens(tokens)

    def save_vocabulary(self, save_directory, filename_prefix=None):
        """
        Save the vocabulary and special tokens file to a directory.

        Args:
            save_directory (`str`):
                The directory in which to save the vocabulary.
            filename_prefix (`str`, *optional*):
                An optional prefix to add to the named of the saved files.

        Returns:
            `Tuple(str)`: Paths to the files saved.
        """
        if os.path.isdir(save_directory):
            vocab_file = os.path.join(
                save_directory, self.vocab_files_names["vocab_file"]
            )
        else:
            vocab_file = save_directory

        with open(self.vocab_file, 'rb') as fin:
            proto_str = fin.read()

        with open(vocab_file, "wb") as writer:
            writer.write(proto_str)

        return (vocab_file,)

    def get_prefix_tokens(self):
        prefix_tokens = [self.get_command("[gMASK]"), self.get_command("sop")]
        return prefix_tokens

    def build_single_message(self, role, metadata, message):
        assert role in ["system", "user", "assistant", "observation"], role
        role_tokens = [self.get_command(f"<|{role}|>")] + self.tokenizer.encode(f"{metadata}\n")
        message_tokens = self.tokenizer.encode(message)
        tokens = role_tokens + message_tokens
        return tokens

    def build_chat_input(self, query, history=None, role="user"):
        if history is None:
            history = []
        input_ids = []
        for item in history:
            content = item["content"]
            if item["role"] == "system" and "tools" in item:
                content = content + "\n" + json.dumps(item["tools"], indent=4, ensure_ascii=False)
            input_ids.extend(self.build_single_message(item["role"], item.get("metadata", ""), content))
        input_ids.extend(self.build_single_message(role, "", query))
        input_ids.extend([self.get_command("<|assistant|>")])
        return self.batch_encode_plus([input_ids], return_tensors="pt", is_split_into_words=True)

    def build_inputs_with_special_tokens(
        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
    ) -> List[int]:
        """
        Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating and
        adding special tokens. A BERT sequence has the following format:

        - single sequence: `[CLS] X [SEP]`
        - pair of sequences: `[CLS] A [SEP] B [SEP]`

        Args:
            token_ids_0 (`List[int]`):
                List of IDs to which the special tokens will be added.
            token_ids_1 (`List[int]`, *optional*):
                Optional second list of IDs for sequence pairs.

        Returns:
            `List[int]`: List of [input IDs](../glossary#input-ids) with the appropriate special tokens.
        """
        prefix_tokens = self.get_prefix_tokens()
        token_ids_0 = prefix_tokens + token_ids_0
        if token_ids_1 is not None:
            token_ids_0 = token_ids_0 + token_ids_1 + [self.get_command("<eos>")]
        return token_ids_0