StableSR
StableSR copied to clipboard
IceClear
[IJCV2024] Exploiting Diffusion Prior for Real-World Image Super-Resolution
The effect of using display pictures is very poor Result picture run command： python scripts/sr_val_ddim_text_T_negativeprompt_canvas_tile.py \ --config configs/stableSRNew/v2-finetune_text_T_768v.yaml --ckpt ./models/stablesr_768v_000139.ckpt \ --vqgan_ckpt ./models/vqgan_cfw_00011.ckpt --init-img ./test --outdir ../OUT_PATH/ \ --ddim_steps 20...
SD-Turbo
It seems that it could finish inference in 4-8 steps, right? Please update about SD-turbo! @IceClear
When training the Time-aware encoder with SFT, the training loss does not decrease at all. How should I address this situation? The training statement I executed is: python main.py --train...
Dear StableSR team, I try to test the demo colab StableSR_inference.ipynb [1], and met an issue about RuntimeError: PytorchStreamReader. May you support me this? brgds, ---- [1] demo colab, https://colab.research.google.com/drive/11SE2_oDvbYtcuHDbaLAxsKk_o3flsO1T?usp=sharing#scrollTo=Cj2YQGg3J0TQ...
Hi, I changed to my dataset to train SFT and it freezes at validation sanity check Could you please tell me how to solve this issue? thanks
"Global exception Input type (struct c10::Half) and bias type (float) should be the same" who can help me please.
hello ，I am encountering some issues: AttributeError: module 'ldm.models.diffusion.ddpm' has no attribute 'LatentDiffusionSRTextWT'
When i run this code: python scripts/sr_val_ddpm_text_T_vqganfin_old.py --config configs/stableSRNew/v2-finetune_text_T_512.yaml --ckpt "./model/vqgan_cfw_00011.ckpt1" --vqgan_ckpt "./model/vqgan_cfw_00011.ckpt1" --init-img ./inputs/test_example/ --outdir ./outputs/ --ddpm_steps 200 --dec_w 0.5 --colorfix_type adain input is 128 * 128
hi, thanks for this excellent work. Could you please share the code for finetuning stable diffusion to get the pretrained stable diffusion checkpoint? Thanks!
Hello, I finetuned my own model with additional data following the README; Finally i got the ckpt, and i tested my images, but i found using "scripts/sr_val_ddpm_text_T_vqganfin_oldcanvas.py" with "--upscale 4"...
