Test Image

[ChengxiHAN]

Hi bro, I want to test whu dataset when I run:

python ddpm_cd.py --config config/whu_test.json --phase test -enable_wandb -log_eval

but there is an error, could you tell me how to fix it? thanks, bro

22-10-10 16:44:17.159 - INFO: Model [DDPM] is created.
22-10-10 16:44:17.159 - INFO: Initial Diffusion Model Finished
22-10-10 16:44:17.591 - INFO: Loading pretrained model for CD model [experiments/ddpm-RS-CDHead-WHU_221008_144806/checkpoint/cd_model_E79] ...
Traceback (most recent call last):
File "ddpm_cd.py", line 96, in
change_detection = Model.create_CD_model(opt)
File "/data/chengxi.han/Sigma124/ddpm-cd/model/init.py", line 13, in create_CD_model
m = M(opt)
File "/data/chengxi.han/Sigma124/ddpm-cd/model/cd_model.py", line 49, in init
self.load_network()
File "/data/chengxi.han/Sigma124/ddpm-cd/model/cd_model.py", line 161, in load_network
network.load_state_dict(torch.load(
File "/data/chengxi.han/anaconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1604, in load_state_dict
raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
RuntimeError: Error(s) in loading state_dict for cd_head_v2:
size mismatch for decoder.0.block.0.weight: copying a param with shape torch.Size([1024, 3072, 1, 1]) from checkpoint, the shape in cur model is torch.Size([1024, 2048, 1, 1]).
size mismatch for decoder.2.block.0.weight: copying a param with shape torch.Size([1024, 3072, 1, 1]) from checkpoint, the shape in cur model is torch.Size([1024, 2048, 1, 1]).
size mismatch for decoder.4.block.0.weight: copying a param with shape torch.Size([512, 1536, 1, 1]) from checkpoint, the shape in currmodel is torch.Size([512, 1024, 1, 1]).
size mismatch for decoder.6.block.0.weight: copying a param with shape torch.Size([256, 768, 1, 1]) from checkpoint, the shape in curreodel is torch.Size([256, 512, 1, 1]).
size mismatch for decoder.8.block.0.weight: copying a param with shape torch.Size([128, 384, 1, 1]) from checkpoint, the shape in curreodel is torch.Size([128, 256, 1, 1]).
wandb: Waiting for W&B process to finish... (failed 1). Press Control-C to abort syncing.
wandb: Synced jolly-bee-43: https://wandb.ai/sigmahan/ddpm-RS-CDHead/runs/3h7lld6h
wandb: Synced 6 W&B file(s), 0 media file(s), 0 artifact file(s) and 0 other file(s)
wandb: Find logs at: ./experiments/wandb/run-20221010_164400-3h7lld6h/logs

[wgcban]

Hi @ChengxiHAN
Did you specify the correct time steps (t) corresponding to the downloaded pre-trained model in the config file here