what do you mean by that

what do you mean by that
104 Replies
goody
goody2y ago
Let me try to explain better 🙂 Example, starting from https://civitai.com/gallery/59757?modelId=5657&modelVersionId=6582&infinite=false&returnUrl=%2Fmodels%2F5657%2Fvinteprotogenmix-v10 using my merged checkpoint (that model -> my checkpoint -> sd15) keep the exact prompt and seed, and it's fairly close:
goody
goody2y ago
No description
Furkan Gözükara SECourses
wow very nice
goody
goody2y ago
BUT
Furkan Gözükara SECourses
but getting your face hard
goody
goody2y ago
yes! 🙂
Furkan Gözükara SECourses
that is whats up with those models
goody
goody2y ago
No description
Furkan Gözükara SECourses
they get generic things
goody
goody2y ago
it always wants to come back w/ basically that kind of face, the style is lost etc
Furkan Gözükara SECourses
i still see decent image though tried to generate hundreds of images to see? tested cfg prompt strenght multiple values?
goody
goody2y ago
Hmm. No, i always forget that! darn let me do so 🙂 i was trying to recreate something like:
goody
goody2y ago
No description
goody
goody2y ago
originally same model start let me do x/y also I couldn't figure this out from your video, now it's x/y/z and you get one image across all the grids and strength i get one image per vertical axis so it's very hard to compare
Furkan Gözükara SECourses
i have shown it in multiple videos x/y same x/y/z
goody
goody2y ago
i set x to S/R, y to cfg and it's one seed per column it seems, even if i set a seed. I don't know, it's not a huge deal anyway 🙂 starting a grid run now, let us see. oh I joined Patreon, not sure how to map it here but my initials are R.G. 🙂
Furkan Gözükara SECourses
awesome ty so much do you want me to give you patron role? i can help you x/y plot via anydesk if you want
goody
goody2y ago
sadly i can't share this computer, work stuff. But I will certainly take various questions answers in the future 🙂 Role, sure, is up to you. Happy to support also am running these on M1 Max Mac which works fine, not sure if it matters much if the computations complete successfully ? 🙂
Furkan Gözükara SECourses
it shouldnt matter as long as you are not using incorrect xformers i suppose you cant use it on mac
goody
goody2y ago
i can enable, but i sometimes forget to enable/disabel
goody
goody2y ago
No description
Furkan Gözükara SECourses
very good this is the test you need before merging checkpoint did you test best checkpoint
goody
goody2y ago
ah well this is from merged already 🙂 that being me my face
Furkan Gözükara SECourses
ye i know but before merging you should check which checkpoint works best
goody
goody2y ago
oh you mean best steps checkpoint ?
goody
goody2y ago
messing around with the checkopints it felt like my 3000 steps one was better than 2700 but let me check my memory against actual results 🙂
Furkan Gözükara SECourses
afk for a while
goody
goody2y ago
If i somewhat understand things, if my ckpt is too strong, it would be overtaking the style of the other model
Furkan Gözükara SECourses
you can reduce its combine strength or use lesser step checkpoint
goody
goody2y ago
this is my 2700 steps at .9 merge strength
goody
goody2y ago
No description
goody
goody2y ago
i don't think good enough, so perhaps I will do my 3000 steps at .75 instead of .9
goody
goody2y ago
No description
goody
goody2y ago
really the "right" way to do this is just train against the actual model right? 🙂 ie. train against protogen or whatever
Furkan Gözükara SECourses
your settings good except vae there is better vae on that model you can try training textual inversion on custom models it works better than dreambooth but maybe dreambooth better than before on custom models need testing :d
goody
goody2y ago
oh woah i didn't notice that was 56000 , i had 84000 also but didn't pick it
Furkan Gözükara SECourses
84k better
goody
goody2y ago
this is custom-> my 3000 step -> sd 15 strength .75 w/ the 56000 vae accidentally baked
No description
goody
goody2y ago
i think the ckpt is sort of good, 7.0 1.0/1.1 maybe 1.2 but i am open to whatever you recommend as best current way to train 🙂 20 images, you know my end goal, put my face into fancy styles like you and (i forget his name but I pasted the screen earlier)
Furkan Gözükara SECourses
looks like cfg 7 best
goody
goody2y ago
yeah
Furkan Gözükara SECourses
i think you can already generate good ones with this merged
goody
goody2y ago
to me that means probably it's "good", i.e. i don't have to push weights or cfg high?
Furkan Gözükara SECourses
prompts making huge difference
goody
goody2y ago
👍 Portrait of a handsome confident young (rgdw person:weight_val), rimlight, geometric pattern on the clothes, colorful background
Furkan Gözükara SECourses
let me test your prompt on my model can you type negative too
goody
goody2y ago
I am doing this one now, sorry i accidentally lost the previous: prompt: centered detailed portrait of a (rgdw person:weight_val), vibrant peacock feathers, intricate, elegant, highly detailed, digital painting, artstation, smooth, sharp focus, illustration, illuminated lines, outrun, vaporware, intricate venetian patterns, cyberpunk darksynth, by audrey kawasaki and ilya kuvshinov and alphonse mucha, award winning, ultra realistic, octane render negative: canvas frame, cartoon, 3d, ((disfigured)), ((bad art)), ((deformed)),((extra limbs)), ((extra barrel)),((close up)),((b&w)), weird colors, blurry, (((duplicate))), ((morbid)), ((mutilated)), [out of frame], extra fingers, mutated hands, ((poorly drawn hands)), ((poorly drawn face)), (((mutation))), (((deformed))), ((ugly)), blurry, ((bad anatomy)), (((bad proportions))), ((extra limbs)), cloned face, (((disfigured))), out of frame, ugly, extra limbs, (bad anatomy), gross proportions, (malformed limbs), ((missing arms)), ((missing legs)), (((extra arms))), (((extra legs))), mutated hands, (fused fingers), (too many fingers), (((long neck))), (((tripod))), (((tube))), Photoshop, video game, ugly, tiling, poorly drawn hands, poorly drawn feet, poorly drawn face, out of frame, mutation, mutated, extra limbs, extra legs, extra arms, disfigured, deformed, cross-eye, body out of frame, blurry, bad art, bad anatomy, 3d render, spots on neck, cropped, out of frame seed 2346951780 DDIM (b/c that's what I saw, but I don't have a real knowledge of which is right when) weight_val,1.0,1.1,1.2,1.3,1.4 7,8,9,10,11 20 steps
Furkan Gözükara SECourses
my results with same seed
Furkan Gözükara SECourses
i use eular a
goody
goody2y ago
and i don't remember what vae i baked in this one 😦 woah
Furkan Gözükara SECourses
random seed
Furkan Gözükara SECourses
which model you were merging i will try it
goody
goody2y ago
those are much ... i feel it "understands" you better your training
Furkan Gözükara SECourses
by the way this is 50 epoch training from my new video
goody
goody2y ago
VinteProtogenMix V1.0 | Stable Diffusion Checkpoint | Civitai
this model makes a colorful pictures. crisp sharp colors, character also looks very nice.was merged22h vintedois-diffusion-v0-1 + 1(protogenV22AnimeOffi_22 - v1-5-pruned-emaonly)+ vintedois-diffusion-v0-1+ Protogen v2.2 (Anime) Official ReleaseVAE not required.
goody
goody2y ago
wow i get confused on the epoch calculation but i think mine is more at any rate
Furkan Gözükara SECourses
explained it in new video
goody
goody2y ago
20 images * 3000 steps (i think...200 class)
Furkan Gözükara SECourses
clearly several times :d
goody
goody2y ago
how perfect! 🙂
goody
goody2y ago
Gist
rgdw man 50x classification
rgdw man 50x classification. GitHub Gist: instantly share code, notes, and snippets.
goody
goody2y ago
i already kicked off training but if you see something awful i would stop it maybe
goody
goody2y ago
No description
goody
goody2y ago
i might have to stop after class image generation anyway to drive to work 🙂 differences from past: 50x, bf16, set vae in main UI (not specified in training, not sure if matters) don't know if it's the vae but the class images look really good for a change! 🙂 I might write a script to capture as many details as one can about training env and model bc I’m always forgetting and would like to be able to just drop some file in and go. Unless something exist already.Ike model yaml and json, package versions, SD settings probably hmm
Furkan Gözükara SECourses
i think you can generate class images on txt2img tab it is faster then put them into a folder and give that folder path
goody
goody2y ago
i couldn't figure how to do all of them at once...there is batch (count? the bottom one) but honeslty i wasn't sure if I could just put 1700 there 🙂 i wonder if it's possible to make a docker image for all this to pull into runpod... something to play with later 🙂
Furkan Gözükara SECourses
dont have experience with dockers :d but i can connect your computer and show
goody
goody2y ago
no worries i have docker experience! 🙂 maybe can help everyone out. but that's just for getting the environment all set up, maybe inject default settings
goody
goody2y ago
No description
goody
goody2y ago
I love runpod, just keeps on chugging while i was off doing a work presentation!
Furkan Gözükara SECourses
pick higher ram machine helps a lot
goody
goody2y ago
No description
goody
goody2y ago
that's good yeah?
Furkan Gözükara SECourses
yep best one
goody
goody2y ago
well, it sorta went ok. 🙂 permission to dm? actually trying shivam trainer now again, high RAM premium GPU colab. 6000 steps (why not 🙂 ) 50x. bf16 seems to work, no 8bit adam. Pruned my training images to reduce similar. 🤞
goody
goody2y ago
errr fp16
No description
Furkan Gözükara SECourses
what is the model you used? if that branch dont exists you need to chnage
goody
goody2y ago
it was runway 1-5. Again very odd results. at 1000 steps even (28 instance, 1652 class, 1 size batch....58 steps per epoch? that would be 17 epoch?) some beautiful images then just regular photo of me. Or maybe that was 1400. At 1000 actually i think it's all images but it learned me as having big pink lips which maybe came form doing "person" not "man" 🙂 or is it just steps / instance, so 1000 / 28 = 35 epoch
goody
goody2y ago
No description
goody
goody2y ago
so i think I will go back to 800 model and remove the 1 or two photos that imply i had lip surgery 🙂 or maybe just start over w/o them. It's not bad, just 2 'photos' break through in that batch
Furkan Gözükara SECourses
i think this new incoming video will help you i have compared 10 different settings
goody
goody2y ago
oh wow that all sounds very exciting! btw unless it's not recommended highly, I think it's a wonderful hack to use output pics from overtrained model (where it is just a perfect made-up pic of me) for inference 🙂
Furkan Gözükara SECourses
yep some people prefer this too
Fabricatedgirls
Stable Diffusion 🎨 AI Art (@DiffusionPics)
Stable Diffusion XL preview is now on DreamStudio! Will be released after training is complete. #AIArt #StableDiffusion2 / #StableDiffusion
From Stable Diffusion 🎨 AI Art (@DiffusionPics)
Twitter
Fabricatedgirls
looks like stable diffusionXL is coming soon
Furkan Gözükara SECourses
wow nice
Fabricatedgirls
Joe Penna (@MysteryGuitarM)
@EMostaque all straight out of the model no tricks, no inits
Likes
116
From Joe Penna (@MysteryGuitarM)
Twitter
From An unknown user
From An unknown user
From An unknown user
Fabricatedgirls
SD XL images its able to do different aspect ratios some of these are super wide aspect in the thread looks like it may be resolution independent for training, which would be incredible
Fabricatedgirls
from the thread: super wide aspect
No description
Furkan Gözükara SECourses
amazing
FORTUNE VARO
FORTUNE VARO2y ago
Hi! I´m working on a project and need HD 16:9 Aspect, I´m working on a 640 x 384 resolution and upscaling to a X3 value in the output settings but the result is verry bad, very cartoonish and ultra stilyzed and I´m looking a more photorealistic vibe. Is there a way to ad Upscale Models to Deforum, by default there are just 3 models. THANKS !
Furkan Gözükara SECourses
use ultra sharp upscaler it is best is your 640 x 384 good?
FORTUNE VARO
FORTUNE VARO2y ago
Hi Dr. Thanks for answering! No, 640x384 is good for testing prompts and values but not for final export. I don´t see the Ultra Shar option on the Upscale model in Output : only this 3 optios appear: realesr-animevideov3 ---- realesr x4plus --- realesr x4plus-anime
Furkan Gözükara SECourses
you have to download it i explained in multiple videos :d
FORTUNE VARO
FORTUNE VARO2y ago
Ho! Thanks! I'll look for it Is there an specific video? I have seen almost all of them and don't remember learning about ultra sharp upscaler
Furkan Gözükara SECourses
this is a technique that you can use
FORTUNE VARO
FORTUNE VARO2y ago
WOW! Awesome! Thank you 💯 It appears that Ultrashar only works for still images and not for video. UltraSharp*
Furkan Gözükara SECourses
yes it works for images
Want results from more Discord servers?
Add your server