KeyError: 'model.diffusion_model.input_blocks.0.0.weight'

[David Helen]

hello
@justinpinkney
when i load the finetune model sd-clip-vit-l14-img-embed_ema_only.ckpt
it alert: KeyError: 'model.diffusion_model.input_blocks.0.0.weight'
so how to solve this problem?
Thanks!!!

my yaml is:
`model:
base_learning_rate: 1.0e-05
target: ldm.models.diffusion.ddpm.LatentDiffusion
params:
linear_start: 0.00085
linear_end: 0.0120
num_timesteps_cond: 1
log_every_t: 200
timesteps: 1000
first_stage_key: image
cond_stage_key: rgb_image
image_size: 64
channels: 4
cond_stage_trainable: False # Note: different from the one we trained before
# unet_trainable: attn
# unet_trainable: "attn"
conditioning_key: crossattn
monitor: val/loss_simple_ema
scale_factor: 0.18215

scheduler_config: # 10000 warmup steps
  target: ldm.lr_scheduler.LambdaLinearScheduler
  params:
    warm_up_steps: [ 1000 ]
    cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases
    f_start: [ 1.e-6 ]
    f_max: [ 1. ]
    f_min: [ 1. ]

unet_config:
  target: ldm.modules.diffusionmodules.openaimodel.UNetModel
  params:
    image_size: 32 # unused
    in_channels: 4
    out_channels: 4
    model_channels: 320
    attention_resolutions: [ 4, 2, 1 ]
    num_res_blocks: 2
    channel_mult: [ 1, 2, 4, 4 ]
    num_heads: 8
    use_spatial_transformer: True
    transformer_depth: 1
    context_dim: 768
    use_checkpoint: True
    legacy: False

first_stage_config:
  target:  ldm.models.autoencoder.AutoencoderKL #ldm.models.autoencoder.VQModelInterface    #ldm.models.autoencoder.AutoencoderKL
  params:
    ckpt_path: /home/dailongquan/110.014/wsm/stable-diffusion-main/models/first_stage_models/kl-f8/model.ckpt
    embed_dim: 4
    #n_embed: 8192
    monitor: val/rec_loss
    ddconfig:
      double_z: True
      z_channels: 4
      resolution: 256
      in_channels: 3
      out_ch: 3
      ch: 128
      ch_mult:
      - 1
      - 2
      - 4
      - 4
      num_res_blocks: 2
      attn_resolutions: []
      dropout: 0.0
    lossconfig:
      target: torch.nn.Identity

cond_stage_config:
  target: ldm.modules.encoders.modules.FrozenCLIPImageEmbedder

data:
target: main.DataModuleFromConfig
params:
batch_size: 2
num_workers: 8
num_val_workers: 0 # Avoid a weird val dataloader issue
train:
target: ldm.data.simple.FangHuaData
# image_key: image
params:
root_dir: /home/share/movie_dataset/fanghua/png
ext: jpg
image_transforms:
- target: torchvision.transforms.Resize
params:
size: 256
interpolation: 3
- target: torchvision.transforms.RandomCrop
params:
size: 256

validation:
  target: ldm.data.simple.FangHuaData
  # image_key: image
  params:
    root_dir: /home/share/movie_dataset/fanghua/png
    ext: jpg
    image_transforms:
    - target: torchvision.transforms.Resize
      params:
        size: 256
        interpolation: 3
    - target: torchvision.transforms.RandomCrop
      params:
        size: 256

lightning:
find_unused_parameters: false
modelcheckpoint:
params:
every_n_train_steps: 500
callbacks:
image_logger:
target: main.ImageLogger
params:
batch_frequency: 100
max_images: 8
increase_log_steps: False
log_first_step: True
log_images_kwargs:
use_ema_scope: False
inpaint: False
plot_progressive_rows: False
plot_diffusion_rows: False
N: 8
unconditional_guidance_scale: 3.0
unconditional_guidance_label: [""]

trainer:
benchmark: True
# val_check_interval: 5000000 # really sorry
num_sanity_val_steps: 0
accumulate_grad_batches: 2
`