step2 faile

[FourTest]

def load_wonder3d_pipeline(config): pipeline = DiffusionPipeline.from_pretrained(

config.pretrained_model_name_or_path,#这里被墙了,我从其他电脑下载后存放到/home/test/wonder3d这个目录了

'/home/test/wonder3d',
#custom_pipeline='flamehaze1115/wonder3d-pipeline', #放开这里又从网上下
torch_dtype=weight_dtype
)

Traceback (most recent call last): File "mv.py", line 172, in pipeline = load_wonder3d_pipeline(config) File "mv.py", line 30, in load_wonder3d_pipeline pipeline = DiffusionPipeline.from_pretrained( File "/home/test/py3.8_for_DrawingSpinUp/lib/python3.8/site-packages/diffusers/pipelines/pipeline_utils.py", line 949, in from_pretrained pipeline_class = _get_pipeline_class( File "/home/test/py3.8_for_DrawingSpinUp/lib/python3.8/site-packages/diffusers/pipelines/pipeline_utils.py", line 346, in _get_pipeline_class pipeline_cls = getattr(diffusers_module, config["_class_name"]) AttributeError: module 'diffusers' has no attribute 'MVDiffusionImagePipeline'

[FourTest]

解决了,下载flamehaze1115/wonder3d-pipeline,然后custom_pipeline参数指定对应的目录