按照 README 微调太乙的 stable-diffusion 模型后，利用from_pretrained() 方法加载模型出错.

[hermanWu55]

https://github.com/IDEA-CCNL/Fengshenbang-LM/tree/main/fengshen/examples/finetune_taiyi_stable_diffusion 这百年按照了这个逻辑来运行了 finetune.sh 的脚本，没有修改逻辑，还是使用了 fengshen/examples/finetune_taiyi_stable_diffusion/demo_dataset/part_0 文件中提供的 数据拿来训练，我训练后，
在 workspace 中保存的模型，利用
pipe = StableDiffusionPipeline.from_pretrained(local_path).to("cuda")
这个脚本来执行，结果提示我如下报错：
You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.
Loading pipeline components...: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████| 7/7 [00:56<00:00, 8.01s/it]
Traceback (most recent call last):
File "", line 1, in
File "/home/herman.wu/miniconda3/envs/sd39/lib/python3.9/site-packages/diffusers/pipelines/pipeline_utils.py", line 884, in to
module.to(device, dtype)
File "/home/herman.wu/miniconda3/envs/sd39/lib/python3.9/site-packages/transformers/modeling_utils.py", line 2556, in to
return super().to(*args, **kwargs)
File "/home/herman.wu/miniconda3/envs/sd39/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1152, in to
return self._apply(convert)
File "/home/herman.wu/miniconda3/envs/sd39/lib/python3.9/site-packages/torch/nn/modules/module.py", line 802, in _apply
module._apply(fn)
File "/home/herman.wu/miniconda3/envs/sd39/lib/python3.9/site-packages/torch/nn/modules/module.py", line 802, in _apply
module._apply(fn)
File "/home/herman.wu/miniconda3/envs/sd39/lib/python3.9/site-packages/torch/nn/modules/module.py", line 825, in _apply
param_applied = fn(param)
File "/home/herman.wu/miniconda3/envs/sd39/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1150, in convert
return t.to(device, dtype if t.is_floating_point() or t.is_complex() else None, non_blocking)
NotImplementedError: Cannot copy out of meta tensor; no data!

可以看一下是什么原因导致的么？