Parameter of unet denoiser in the inference stage

[miaomiaomiao66]

Dear author, In the inference stage, the unet denoiser is defined in a straight forward way:
unet = create_model(**opt.unet_kwargs)
and the parameter is load from the pth file, for example 60_net_G.pth
however the following error occurred:
Missing key(s) in state_dict: "time_embed.0.weight", "time_embed.0.bias", "time_embed.2.weight", "time_embed.2.bias", "input_blocks.0.0.weight", "input_blocks.0.0.bias",
Unexpected key(s) in state_dict: "content_stream.res_0.conv_0.bias", "content_stream.res_0.conv_0.weight_orig", "content_stream.res_0.conv_0.weight_u",
After carefully go through the code and the essay, I found the checkpoint file is for DMTmodel, but not the unet denoiser,
Is this caused by parameter setting (**opt.unet_kwargs) or other factors?
I am looking forward to your reply and guidance!