How to generalize the personalize style photo

[YDyd11]

PASD is very nice job!
I was trying to make a personalize style(toonyou) photo, the original photo is 000020x2.png in the folder samples. However, I have tried toonyou.safetensors in civatai, with pasd_rrdb/checkpoint-100000, --use_personalized_model, others as default. I can not generalize the beautiful photo just like you paper show. Can you tell me what are the parameters in personalized_model?
Thank you!

[yangxy]

Try python test_pasd.py --pasd_model_path runs/pasd_light/checkpoint-120000 --personalized_model_path toonyou_beta3.safetensors --use_personalized_model --use_pasd_light --conditioning_scale 0.7 --image_path samples/000020x2.png --upscale 1

For stylization task. please use pasd_light which is more generative. Set smaller value of --conditioning_scale and --upscale to produce better stylization effect.

[YDyd11]

Thanks!!! It is work now!!! Also can work with other lora models which are comes from the civitai and only have 144m.

…

------------------ 原始邮件 ------------------ 发件人: ***@***.***>; 发送时间: 2023年9月13日(星期三) 下午4:20 收件人: ***@***.***>; 抄送: ***@***.***>; ***@***.***>; 主题: Re: [yangxy/PASD] How to generalize the personalize style photo (Issue #3) Try python test_pasd.py --pasd_model_path runs/pasd_light/checkpoint-120000 --personalized_model_path toonyou_beta3.safetensors --use_personalized_model --use_pasd_light --conditioning_scale 0.7 --image_path samples/000020x2.png --upscale 1 For stylization task. please use pasd_light which is more generative. Set smaller value of --conditioning_scale and --upscale to produce better stylization effect. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: ***@***.***>

[yangxy]

I am woking on it. Will be updated tomorrow.

[YDyd11]

Cool!!! I am wondering what is the meaning of rrdb? What are the difference between pasd and pasd_rrdb? It seems it wasn't explained in the paper. ------------------ 原始邮件 ------------------ 发件人: "yangxy/PASD" ***@***.***>; 发送时间: 2023年9月13日(星期三) 晚上8:04 ***@***.***>; ***@***.******@***.***>; 主题: Re: [yangxy/PASD] How to generalize the personalize style photo (Issue #3) I am woking on it. Will be updated tomorrow. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: ***@***.***>

[yangxy]

Updated. RRDB is widely used in SR task. pasd_rrdb is a version with heavier degradation module.

[zhanghongyong123456]

Try python test_pasd.py --pasd_model_path runs/pasd_light/checkpoint-120000 --personalized_model_path toonyou_beta3.safetensors --use_personalized_model --use_pasd_light --conditioning_scale 0.7 --image_path samples/000020x2.png --upscale 1

For stylization task. please use pasd_light which is more generative. Set smaller value of --conditioning_scale and --upscale to produce better stylization effect.

when i run cmd ： python test_pasd.py --pasd_model_path runs/pasd_light/checkpoint-120000 --personalized_model_path toonyou_beta3.safetensors --use_personalized_model --use_pasd_light --conditioning_scale 0.7 --image_path samples/000020x2.png --upscale 1
i get this error：Linear.forward() got an unexpected keyword argument 'scale'

[sunzhe09]

Hi，can I use personalize style photo with lcm lora？