about some small bug of prompt_tuning.py

[duxiaoyu555]

System Info

peft ==0.5.0
python == 3.9
transformers==4.33.1

Who can help?

No response

Information

• The official example scripts
• My own modified scripts

Tasks

• An officially supported task in the examples folder
• My own task or dataset (give details below)

Reproduction

from transformers import AutoTokenizer, AutoModelForCausalLM
from peft import PromptTuningConfig, get_peft_model, TaskType, PromptTuningInit
import torch
tokenizer = AutoTokenizer.from_pretrained("/upp/xgen/xgen-7b-8k-base", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("/upp/xgen/xgen-7b-8k-base", torch_dtype=torch.bfloat16,trust_remote_code=True)
config = PromptTuningConfig(task_type=TaskType.CAUSAL_LM,
                            prompt_tuning_init=PromptTuningInit.TEXT,
                            prompt_tuning_init_text="下面是一段人与机器人的对话。",
                            num_virtual_tokens=len(tokenizer("下面是一段人与机器人的对话。")["input_ids"]),
                            tokenizer_name_or_path="xxxxx")  #(local file)

model = get_peft_model(model, config)

Expected behavior

i have an advice of get_peft_model this method , in this function ,have an class PromptEmbedding in prompt_tuning.py
and line 112 tokenizer = AutoTokenizer.from_pretrained(config.tokenizer_name_or_path) should have an args trust_remote_code=True
i met an issue Tokenizer class xxxx does not exist or is not currently imported. because of it .