预处理器 depth_marigold 错误 diffusers.models.unets.unet_2d_condition.UNet2DConditionModel'>

File "E:\forge-aki-v1.0\python\Lib\site-packages\diffusers\models\modeling_utils.py", line 848, in from_pretrained
raise ValueError(
ValueError: Cannot load <class 'diffusers.models.unets.unet_2d_condition.UNet2DConditionModel'> from Bingxin/Marigold because the following keys are missing:
down_blocks.0.attentions.0.proj_in.dummy, up_blocks.2.attentions.0.transformer_blocks.0.attn1.to_k.dummy, down_blocks.2.attentions.0.proj_in.dummy, up_blocks.3.attentions.1.transformer_blocks.0.ff.net.0.proj.dummy, up_blocks.1.attentions.0.proj_out.dummy, up_blocks.2.attentions.1.transformer_blocks.0.ff.net.2.dummy, mid_block.attentions.0.transformer_blocks.0.attn1.to_q.dummy, mid_block.attentions.0.transformer_blocks.0.attn1.to_k.dummy, down_blocks.2.attentions.0.transformer_blocks.0.attn1.to_q.dummy, up_blocks.1.attentions.0.transformer_blocks.0.attn1.to_v.dummy, up_blocks.2.attentions.0.transformer_blocks.0.attn2.to_v.dummy, up_blocks.2.resnets.2.time_emb_proj.dummy, mid_block.attentions.0.transformer_blocks.0.attn2.to_out.0.dummy, down_blocks.2.attentions.1.transformer_blocks.0.attn2.to_q.dummy, up_blocks.0.resnets.1.time_emb_proj.dummy, up_blocks.2.attentions.2.transformer_blocks.0.attn1.to_v.dummy, up_blocks.1.attentions.1.transformer_blocks.0.ff.net.2.dummy, up_blocks.1.attentions.2.transformer_blocks.0.ff.net.2.dummy, up_blocks.3.attentions.2.transformer_blocks.0.attn2.to_v.dummy, mid_block.attentions.0.transformer_blocks.0.attn2.to_v.dummy, down_blocks.2.attentions.0.transformer_blocks.0.attn2.to_v.dummy, up_blocks.3.attentions.1.transformer_blocks.0.attn1.to_q.dummy, down_blocks.1.attentions.0.transformer_blocks.0.attn1.to_q.dummy, up_blocks.3.attentions.1.transformer_blocks.0.attn1.to_out.0.dummy, time_embedding.linear_2.dummy, up_blocks.1.attentions.2.transformer_blocks.0.attn2.to_k.dummy, up_blocks.2.attentions.1.transformer_blocks.0.attn2.to_out.0.dummy, down_blocks.1.attentions.0.transformer_blocks.0.attn1.to_k.dummy, up_blocks.1.attentions.2.transformer_blocks.0.attn2.to_out.0.dummy, down_blocks.0.attentions.1.transformer_blocks.0.ff.net.0.proj.dummy, up_blocks.2.attentions.2.transformer_blocks.0.ff.net.2.dummy, up_blocks.3.attentions.2.transformer_blocks.0.ff.net.0.proj.dummy, down_blocks.1.attentions.1.proj_out.dummy, up_blocks.1.attentions.1.transformer_blocks.0.ff.net.0.proj.dummy, up_blocks.1.attentions.1.transformer_blocks.0.attn1.to_v.dummy, down_blocks.0.attentions.1.transformer_blocks.0.attn1.to_k.dummy, up_blocks.3.attentions.1.proj_out.dummy, up_blocks.2.attentions.0.transformer_blocks.0.ff.net.0.proj.dummy, up_blocks.3.attentions.2.transformer_blocks.0.attn2.to_out.0.dummy, down_blocks.1.resnets.0.time_emb_proj.dummy, up_blocks.3.attentions.0.proj_out.dummy, down_blocks.1.attentions.0.transformer_blocks.0.attn1.to_out.0.dummy, down_blocks.2.attentions.0.transformer_blocks.0.attn1.to_k.dummy, up_blocks.1.attentions.1.transformer_blocks.0.attn1.to_out.0.dummy, up_blocks.2.attentions.2.transformer_blocks.0.attn2.to_v.dummy, down_blocks.0.resnets.0.time_emb_proj.dummy, up_blocks.1.attentions.0.transformer_blocks.0.attn2.to_out.0.dummy, up_blocks.2.attentions.0.transformer_blocks.0.attn2.to_out.0.dummy, down_blocks.1.attentions.1.transformer_blocks.0.attn2.to_k.dummy, up_blocks.1.attentions.2.transformer_blocks.0.attn1.to_out.0.dummy, up_blocks.3.resnets.1.time_emb_proj.dummy, down_blocks.2.resnets.0.time_emb_proj.dummy, down_blocks.0.attentions.1.transformer_blocks.0.attn1.to_q.dummy, up_blocks.0.resnets.0.time_emb_proj.dummy, down_blocks.0.attentions.1.transformer_blocks.0.attn1.to_v.dummy, up_blocks.2.attentions.1.transformer_blocks.0.attn1.to_out.0.dummy, up_blocks.2.attentions.2.proj_in.dummy, down_blocks.1.attentions.1.transformer_blocks.0.attn1.to_v.dummy, up_blocks.2.attentions.1.proj_in.dummy, down_blocks.0.attentions.0.transformer_blocks.0.attn2.to_q.dummy, up_blocks.2.attentions.1.transformer_blocks.0.attn2.to_v.dummy, up_blocks.2.attentions.0.transformer_blocks.0.attn1.to_out.0.dummy, up_blocks.1.attentions.2.transformer_blocks.0.attn1.to_k.dummy, up_blocks.2.attentions.0.transformer_blocks.0.attn2.to_q.dummy, down_blocks.2.attentions.0.transformer_blocks.0.attn1.to_v.dummy, up_blocks.2.resnets.0.time_emb_proj.dummy, mid_block.attentions.0.transformer_blocks.0.attn1.to_v.dummy, up_blocks.3.attentions.2.proj_out.dummy, up_blocks.2.attentions.2.proj_out.dummy, up_blocks.3.attentions.2.transformer_blocks.0.attn1.to_v.dummy, mid_block.resnets.0.time_emb_proj.dummy, down_blocks.0.attentions.0.transformer_blocks.0.attn1.to_q.dummy, up_blocks.1.attentions.1.transformer_blocks.0.attn2.to_v.dummy, up_blocks.2.attentions.2.transformer_blocks.0.ff.net.0.proj.dummy, time_embedding.linear_1.dummy, down_blocks.0.attentions.0.transformer_blocks.0.ff.net.0.proj.dummy, up_blocks.1.attentions.1.transformer_blocks.0.attn2.to_out.0.dummy, up_blocks.3.resnets.2.time_emb_proj.dummy, down_blocks.1.attentions.1.transformer_blocks.0.attn2.to_q.dummy, up_blocks.2.attentions.1.transformer_blocks.0.attn2.to_k.dummy, up_blocks.2.attentions.0.transformer_blocks.0.ff.net.2.dummy, up_blocks.1.attentions.1.transformer_blocks.0.attn2.to_k.dummy, up_blocks.3.attentions.1.proj_in.dummy, up_blocks.2.attentions.2.transformer_blocks.0.attn1.to_out.0.dummy, up_blocks.2.attentions.1.transformer_blocks.0.attn1.to_v.dummy, down_blocks.2.attentions.1.transformer_blocks.0.attn1.to_v.dummy, up_blocks.1.attentions.0.transformer_blocks.0.attn2.to_q.dummy, down_blocks.2.attentions.0.transformer_blocks.0.attn2.to_out.0.dummy, up_blocks.2.resnets.1.time_emb_proj.dummy, down_blocks.1.attentions.0.transformer_blocks.0.attn1.to_v.dummy, up_blocks.3.attentions.0.transformer_blocks.0.ff.net.0.proj.dummy, down_blocks.0.attentions.1.proj_in.dummy, down_blocks.0.attentions.0.transformer_blocks.0.attn1.to_v.dummy, down_blocks.1.attentions.0.transformer_blocks.0.ff.net.2.dummy, up_blocks.1.attentions.1.transformer_blocks.0.attn1.to_k.dummy, down_blocks.2.attentions.1.transformer_blocks.0.ff.net.2.dummy, up_blocks.1.attentions.0.transformer_blocks.0.attn1.to_out.0.dummy, up_blocks.3.attentions.2.transformer_blocks.0.attn1.to_q.dummy, down_blocks.3.resnets.1.time_emb_proj.dummy, up_blocks.1.resnets.1.time_emb_proj.dummy, up_blocks.3.attentions.1.transformer_blocks.0.attn2.to_v.dummy, up_blocks.0.resnets.2.time_emb_proj.dummy, up_blocks.1.attentions.2.transformer_blocks.0.attn2.to_q.dummy, up_blocks.3.attentions.0.transformer_blocks.0.attn1.to_k.dummy, down_blocks.2.attentions.1.transformer_blocks.0.attn2.to_k.dummy, mid_block.attentions.0.transformer_blocks.0.ff.net.2.dummy, down_blocks.0.attentions.1.transformer_blocks.0.attn2.to_out.0.dummy, up_blocks.2.attentions.2.transformer_blocks.0.attn2.to_k.dummy, up_blocks.1.attentions.0.transformer_blocks.0.attn1.to_q.dummy, down_blocks.1.attentions.1.transformer_blocks.0.ff.net.2.dummy, up_blocks.1.attentions.0.proj_in.dummy, down_blocks.2.attentions.0.transformer_blocks.0.attn2.to_q.dummy, up_blocks.2.attentions.1.transformer_blocks.0.attn2.to_q.dummy, up_blocks.3.attentions.1.transformer_blocks.0.attn1.to_v.dummy, up_blocks.1.attentions.2.transformer_blocks.0.attn1.to_v.dummy, up_blocks.3.attentions.1.transformer_blocks.0.attn2.to_q.dummy, up_blocks.3.attentions.0.transformer_blocks.0.ff.net.2.dummy, up_blocks.1.resnets.0.time_emb_proj.dummy, up_blocks.3.attentions.2.transformer_blocks.0.ff.net.2.dummy, down_blocks.2.attentions.0.transformer_blocks.0.attn2.to_k.dummy, up_blocks.1.attentions.2.proj_out.dummy, up_blocks.1.attentions.0.transformer_blocks.0.attn2.to_v.dummy, down_blocks.0.resnets.1.time_emb_proj.dummy, down_blocks.1.resnets.1.time_emb_proj.dummy, down_blocks.2.attentions.0.transformer_blocks.0.ff.net.0.proj.dummy, up_blocks.1.attentions.0.transformer_blocks.0.ff.net.0.proj.dummy, up_blocks.2.attentions.0.proj_in.dummy, up_blocks.3.attentions.2.proj_in.dummy, up_blocks.2.attentions.2.transformer_blocks.0.attn2.to_q.dummy, up_blocks.1.resnets.2.time_emb_proj.dummy, down_blocks.0.attentions.0.transformer_blocks.0.attn2.to_v.dummy, up_blocks.2.attentions.0.transformer_blocks.0.attn1.to_q.dummy, up_blocks.2.attentions.1.transformer_blocks.0.attn1.to_k.dummy, down_blocks.1.attentions.0.transformer_blocks.0.attn2.to_k.dummy, up_blocks.3.attentions.0.transformer_blocks.0.attn2.to_v.dummy, up_blocks.3.attentions.0.transformer_blocks.0.attn1.to_out.0.dummy, down_blocks.2.attentions.1.proj_in.dummy, down_blocks.1.attentions.0.proj_out.dummy, up_blocks.3.attentions.1.transformer_blocks.0.attn1.to_k.dummy, up_blocks.3.attentions.0.transformer_blocks.0.attn1.to_v.dummy, up_blocks.1.attentions.1.proj_in.dummy, up_blocks.1.attentions.2.transformer_blocks.0.attn2.to_v.dummy, mid_block.attentions.0.transformer_blocks.0.attn1.to_out.0.dummy, down_blocks.2.attentions.1.transformer_blocks.0.ff.net.0.proj.dummy, down_blocks.0.attentions.1.transformer_blocks.0.attn2.to_v.dummy, up_blocks.3.attentions.1.transformer_blocks.0.attn2.to_out.0.dummy, mid_block.attentions.0.proj_out.dummy, up_blocks.2.attentions.0.transformer_blocks.0.attn2.to_k.dummy, down_blocks.1.attentions.0.proj_in.dummy, down_blocks.2.attentions.1.transformer_blocks.0.attn1.to_k.dummy, up_blocks.1.attentions.1.transformer_blocks.0.attn2.to_q.dummy, up_blocks.1.attentions.0.transformer_blocks.0.attn2.to_k.dummy, down_blocks.3.resnets.0.time_emb_proj.dummy, down_blocks.1.attentions.1.transformer_blocks.0.ff.net.0.proj.dummy, down_blocks.1.attentions.1.proj_in.dummy, down_blocks.2.resnets.1.time_emb_proj.dummy, up_blocks.3.attentions.1.transformer_blocks.0.attn2.to_k.dummy, down_blocks.2.attentions.1.transformer_blocks.0.attn1.to_q.dummy, down_blocks.1.attentions.0.transformer_blocks.0.attn2.to_v.dummy, up_blocks.1.attentions.2.proj_in.dummy, up_blocks.1.attentions.1.transformer_blocks.0.attn1.to_q.dummy, up_blocks.2.attentions.2.transformer_blocks.0.attn1.to_q.dummy, down_blocks.1.attentions.0.transformer_blocks.0.attn2.to_q.dummy, up_blocks.1.attentions.0.transformer_blocks.0.attn1.to_k.dummy, down_blocks.1.attentions.1.transformer_blocks.0.attn1.to_out.0.dummy, up_blocks.2.attentions.0.transformer_blocks.0.attn1.to_v.dummy, down_blocks.0.attentions.1.proj_out.dummy, down_blocks.0.attentions.0.transformer_blocks.0.attn2.to_k.dummy, down_blocks.1.attentions.0.transformer_blocks.0.ff.net.0.proj.dummy, down_blocks.0.attentions.1.transformer_blocks.0.ff.net.2.dummy, up_blocks.2.attentions.2.transformer_blocks.0.attn2.to_out.0.dummy, down_blocks.1.attentions.1.transformer_blocks.0.attn1.to_q.dummy, up_blocks.3.attentions.0.transformer_blocks.0.attn2.to_q.dummy, down_blocks.2.attentions.1.proj_out.dummy, down_blocks.1.attentions.1.transformer_blocks.0.attn2.to_out.0.dummy, down_blocks.2.attentions.1.transformer_blocks.0.attn2.to_v.dummy, down_blocks.2.attentions.0.proj_out.dummy, up_blocks.3.attentions.0.proj_in.dummy, mid_block.attentions.0.transformer_blocks.0.attn2.to_q.dummy, down_blocks.1.attentions.1.transformer_blocks.0.attn1.to_k.dummy, up_blocks.2.attentions.0.proj_out.dummy, down_blocks.0.attentions.0.transformer_blocks.0.attn2.to_out.0.dummy, down_blocks.0.attentions.0.transformer_blocks.0.attn1.to_k.dummy, down_blocks.1.attentions.1.transformer_blocks.0.attn2.to_v.dummy, up_blocks.3.attentions.2.transformer_blocks.0.attn1.to_out.0.dummy, down_blocks.2.attentions.0.transformer_blocks.0.ff.net.2.dummy, down_blocks.1.attentions.0.transformer_blocks.0.attn2.to_out.0.dummy, up_blocks.3.attentions.0.transformer_blocks.0.attn2.to_k.dummy, up_blocks.1.attentions.2.transformer_blocks.0.ff.net.0.proj.dummy, up_blocks.3.attentions.0.transformer_blocks.0.attn2.to_out.0.dummy, up_blocks.2.attentions.1.proj_out.dummy, up_blocks.2.attentions.2.transformer_blocks.0.attn1.to_k.dummy, up_blocks.1.attentions.0.transformer_blocks.0.ff.net.2.dummy, up_blocks.3.attentions.2.transformer_blocks.0.attn2.to_k.dummy, up_blocks.3.attentions.1.transformer_blocks.0.ff.net.2.dummy, up_blocks.2.attentions.1.transformer_blocks.0.attn1.to_q.dummy, up_blocks.1.attentions.2.transformer_blocks.0.attn1.to_q.dummy, down_blocks.2.attentions.1.transformer_blocks.0.attn2.to_out.0.dummy, down_blocks.0.attentions.1.transformer_blocks.0.attn1.to_out.0.dummy, down_blocks.0.attentions.0.proj_out.dummy, mid_block.attentions.0.proj_in.dummy, down_blocks.2.attentions.1.transformer_blocks.0.attn1.to_out.0.dummy, mid_block.attentions.0.transformer_blocks.0.ff.net.0.proj.dummy, mid_block.resnets.1.time_emb_proj.dummy, up_blocks.3.attentions.0.transformer_blocks.0.attn1.to_q.dummy, mid_block.attentions.0.transformer_blocks.0.attn2.to_k.dummy, up_blocks.3.resnets.0.time_emb_proj.dummy, down_blocks.0.attentions.0.transformer_blocks.0.attn1.to_out.0.dummy, up_blocks.2.attentions.1.transformer_blocks.0.ff.net.0.proj.dummy, down_blocks.0.attentions.0.transformer_blocks.0.ff.net.2.dummy, down_blocks.2.attentions.0.transformer_blocks.0.attn1.to_out.0.dummy, down_blocks.0.attentions.1.transformer_blocks.0.attn2.to_k.dummy, down_blocks.0.attentions.1.transformer_blocks.0.attn2.to_q.dummy, up_blocks.3.attentions.2.transformer_blocks.0.attn1.to_k.dummy, up_blocks.1.attentions.1.proj_out.dummy, up_blocks.3.attentions.2.transformer_blocks.0.attn2.to_q.dummy.
Please make sure to pass low_cpu_mem_usage=False and device_map=None if you want to randomly initialize those weights or else make sure your checkpoint file is correct.

Reproduce the depth_marigold preprocessing failure and inspect the loading path at python\Lib\site-packages\diffusers\models\modeling_utils.py, line 848. Compare the installed diffusers model-loading behavior with the Bingxin/Marigold checkpoint; done means the checkpoint loads without the reported missing-key error.