replit-dolly / adapt_tokenizer.py
lentan's picture
Upload folder using huggingface_hub
c52e29e
from typing import Union
from transformers import AutoTokenizer, PreTrainedTokenizer, PreTrainedTokenizerFast
Tokenizer = Union[PreTrainedTokenizer, PreTrainedTokenizerFast]
NUM_SENTINEL_TOKENS: int = 100
def adapt_tokenizer_for_denoising(tokenizer: Tokenizer):
"""Adds sentinel tokens and padding token (if missing).
Expands the tokenizer vocabulary to include sentinel tokens
used in mixture-of-denoiser tasks as well as a padding token.
All added tokens are added as special tokens. No tokens are
added if sentinel tokens and padding token already exist.
"""
sentinels_to_add = [f'<extra_id_{i}>' for i in range(NUM_SENTINEL_TOKENS)]
tokenizer.add_tokens(sentinels_to_add, special_tokens=True)
if tokenizer.pad_token is None:
tokenizer.add_tokens('<pad>', special_tokens=True)
tokenizer.pad_token = '<pad>'
assert tokenizer.pad_token_id is not None
sentinels = ''.join([f'<extra_id_{i}>' for i in range(NUM_SENTINEL_TOKENS)])
_sentinel_token_ids = tokenizer(sentinels, add_special_tokens=False).input_ids
tokenizer.sentinel_token_ids = _sentinel_token_ids
class AutoTokenizerForMOD(AutoTokenizer):
"""AutoTokenizer + Adaptation for MOD.
A simple wrapper around AutoTokenizer to make instantiating
an MOD-adapted tokenizer a bit easier.
MOD-adapted tokenizers have sentinel tokens (e.g., <extra_id_0>),
a padding token, and a property to get the token ids of the
sentinel tokens.
"""
@classmethod
def from_pretrained(cls, *args, **kwargs):
"""See `AutoTokenizer.from_pretrained` docstring."""
tokenizer = super().from_pretrained(*args, **kwargs)
adapt_tokenizer_for_denoising(tokenizer)
return tokenizer