accelerate launch --multi_gpu --num_processes=4 train.py --data_dir data/ffhq_49K/ --model_path model/ffhq_10m.pt --deg gaussian --t_ikl 400 --weight_con 0.5 --reg_coeff 0.25 --perturb_h 0.1 Releases No releases published
From the link, download the FFHQ checkpoint ffhq_10m.pt, rename to diffusion_ffhq_10m.pt, and paste it to ../model_zoo. To run guidance based on Analytic posterior covariance, download the precomputed Monte Carlo estimation from the link, and paste it to ./runs. To run guidance based on...
2. 通用框架 2.1 去噪扩散概率模型（DDPMs）2.2 噪声条件评分网络（NCSNs）2.3随机微分方程（SDE）...
# in DAPS folder mkdir checkpoint mv {DOWNLOAD_DIR}/ffqh_10m.pt checkpoint/ffhq256.pt (Optional) For nonlinear deblur task, we need the pretrained model from bkse at here: # in DAPS folder mv {DOWNLOAD_DIR}/GOPRO_wVAE.pth forward_operator/bkse/experiments/pretrained 3. (Optional)...
From the [link](https://drive.google.com/drive/folders/1jElnRoFv7b31fG0v6pTSQkelbSX3xGZh?usp=sharing), download the FFHQ checkpoint ```ffhq_10m.pt```, rename to ```diffusion_ffhq_10m.pt```, and paste it to ```../model_zoo```. To run guidance based on analytic posteri...
noise_schedule linear --num_channels 128 --num_head_channels 64 --num_res_blocks 1 --resblock_updown True --use_fp16 False --use_scale_shift_norm True --timestep_respacing 100 --model_path models/ffhq_10m.pt --base_samples ref_imgs/face --down_N 32 --range_t 20 --save_dir...
From thelink, download the FFHQ checkpointffhq_10m.pt, rename todiffusion_ffhq_10m.pt, and paste it to../model_zoo. To run guidance based onAnalyticposterior covariance, download the precomputed Monte Carlo estimation from thelink, and paste it to./runs. ...
noise_schedule linear --num_channels 128 --num_head_channels 64 --num_res_blocks 1 --resblock_updown True --use_fp16 False --use_scale_shift_norm True --timestep_respacing 100 --model_path models/ffhq_10m.pt --base_samples ref_imgs/face --down_N 32 --range_t 20 --save_dir...