much faster for me
much faster for me
FYouTubeSECourses
Download scripts from here : https://www.patreon.com/posts/sota-very-best-90744385
Full length Twitter video here : https://twitter.com/GozukaraFurkan/status/1712665890259447831
SOTA (The Very Best) Image Captioning Models Script For Stable Diffusion And More
1 Click install and use SOTA image captioning models on your computer. Supports 8 bi...
Full length Twitter video here : https://twitter.com/GozukaraFurkan/status/1712665890259447831
SOTA (The Very Best) Image Captioning Models Script For Stable Diffusion And More
1 Click install and use SOTA image captioning models on your computer. Supports 8 bi...
Fit is still slow though it is accurate
Fi think can be used
Fnice
KYou're using VIT-L-14 in the video. Not laion/CLIP-ViT-bigG-14-laion2B
KRight?!
Fi think you need a good dataset and perhaps captions
KThe captions is easily created by CLIP interrogator and some manual labor. Question is what kind of dataset would be needed :/ All training videos (Including your fantastic videos) are of subject/object training or an art style. Not of a design
K
K200 seconds in and is has barely started interrogating :/
KIt takes forever
KI'm at 9% 430 seconds in
KJust slow
KViT-L-14/openai just takes a few seconde
K
Fi also have rtx 3090
Fwill try and let you know speed. mine working good
KThis is very strange
Fi will let you know in 30 min got to get my girl from school
Fstarting web ui
F
KWhat kind of GPU is that? 
K55.9GB
KNevermind!
KMine:
Ftakes lesser than 1 minute
F
KHow?!? 
KJust loading the model takes forever 
Fare you using batch process?
KNot for one image
F1 image loading will take time
Fi didnt count it
K
KLoading to RAM first :/
Fuse latest version
Fi changed it
Floads directly into vram
K
Fyours using shared vram
Fwhich is big no no
KIt is?
Fyes
Fyou have shared vram usage
KI'll remove it and install the new one
Fyep
KI have not done anything do get shared video memory. You mean my installation does, right?!
Fi am not sure
Fcould be related to your system
