train_dreambooth_lora.py BAAI/AltDiffusion-m9 报错KeyError: 'XLMRobertaTokenizer'

[zouxiaodong]

使用如下脚本，训练BAAI/AltDiffusion-m9 LORA :

export MODEL_NAME="BAAI/AltDiffusion-m9"
export INSTANCE_DIR="dog"
export OUTPUT_DIR="alt-lora-model"

python train_dreambooth_lora.py
--pretrained_model_name_or_path=$MODEL_NAME
--instance_data_dir=$INSTANCE_DIR
--output_dir=$OUTPUT_DIR
--instance_prompt="一张<鸣人>的图片"
--resolution=512
--train_batch_size=1
--gradient_accumulation_steps=1
--checkpointing_steps=100
--learning_rate=1e-4
--report_to="visualdl"
--lr_scheduler="constant"
--lr_warmup_steps=0
--max_train_steps=500
--validation_prompt="一张<鸣人>在篮子里面的图片"
--validation_epochs=50
--lora_rank=4
--seed=0

详细错误如下：
100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 564/564 [00:00<00:00, 2.49MB/s]
Traceback (most recent call last):
File "/home/ai/robert/PaddleMIX/ppdiffusers/examples/dreambooth/train_dreambooth_lora.py", line 1158, in
main()
File "/home/ai/robert/PaddleMIX/ppdiffusers/examples/dreambooth/train_dreambooth_lora.py", line 690, in main
tokenizer = AutoTokenizer.from_pretrained(
File "/home/ai/anaconda3/envs/paddlenlp/lib/python3.10/site-packages/paddlenlp/transformers/auto/tokenizer.py", line 363, in from_pretrained
tokenizer_class = cls._get_tokenizer_class_from_config(
File "/home/ai/anaconda3/envs/paddlenlp/lib/python3.10/site-packages/paddlenlp/transformers/auto/tokenizer.py", line 197, in _get_tokenizer_class_from_config
class_name = cls._name_mapping[init_class]
KeyError: 'XLMRobertaTokenizer'

[westfish]

请参考#32