FThe Dreambooth implementations out there typically introduce generated images out of the model itself into the training set. I do not suggest that, and that's why EveryDream is not a Dreambooth implementation. I instead suggest you can gather other alternative "ground truth" data and "mix it in" with your training data to "preserve" the model and avoid "forgetting" or other artifacts. The tools repo has a Laion data scraper, or you can use things datasets like FFHQ photo set.
I recommend "ground truth" images as opposed to the Dreambooth implementations which use generated images out of the model itself for preservation. Dreambooth training on generated images can reinforce the error already present in the model. Ex. 6 fingers, bad limbs, etc.
Fthis is what i've been saying!
Fvery interesting
Fgonna try it right now!
Fi think dreambooth extension without classification images does what everdreamer trainer does but i havent tested
D@Furkan Gözükara SECourses read my dm's
Feverdreamer uses classification images. but not generated ones
AExcuse my ignorance, but what is everdreamer?
F
Fthis guys says that reg images are useless
Fand captions is better
Fthe opposite of what people like Furkan says
Fthis is so confusing!! xD
Fit's another method for fine-tuning, like kohya_ss, etc
Eso this was it. i have new xformers now, and now my training in automatic 1111 dreambooth extension is working, but only in the sample images dreambooth folder shows me. , when i go to text to image, it is not recognizing my lora input at all. is there a known issue with stable diffusion and lora? theres always something 
(kohya trained lora seems to be working, I may have to just default to kohya instead of dreambooth extension )
(kohya trained lora seems to be working, I may have to just default to kohya instead of dreambooth extension )
MValueError: height and width must be > 0
It happens when I do with Inpaint. Why ?
It happens when I do with Inpaint. Why ?
Fi'm liking everdreamer, it's easy, you just spawn pod and execute a simple command
Fand results looks good, but i'm still on epoch 20, let's see
Tthanks. i tried a fresh install as well as a different commit, but i can't get my it/s back to where it used to be. i have no clue how this happened. i haven't done any system updates or software installs recently, and this literally happened after doing the git pull
Fjust came back
Fhe probably did fine tuning
Fdidnt train face or subject
Fi can connect your pc and install if you are our patreon supporter
Fi can connect your pc and install if you are our patreon supporter
Fbut fine tuning is just training something else, isn't it?
Fdreambooth is fine-tuning
Fwhat is not lora, hypernetworks, textual inversion, etc..... is fine-tuning
Fby fine tuning i mean improving tokens
Fnot teaching a subject
FFit turns out that this deepfake resolution is very low. i had to apply codeformer for fixing face
Fand it changes
Fhowever my face is very different than john wick so still decent
HYou are looking to finetune a model?
HDo you know what everydream does vs Kohya ?
生if i have a own dataset and i want to train it into dolly 2.0 how should i do
DHey! does anyone know how i should go about making an ai generated video that flows smoothly? do i use controlnet1.1?
Sanyone know how to generate a photo with extremely detailed stripe and decoration like the photo above? I trained several model, they are good in looking, but far away from the detail level, anybody have a clue?
KThank Dr.Furkan it works and I can compare different models now. But today I discover that XYZ plot still show the same images when I use different VAE or Clip Skip. The model can still make different images with different clip skip (attached images are clip skip 1 and 10) but XYZ is unable to do so.
I already do the git pull how should I fix this? (My model are trained by Kohya_ss)
I already do the git pull how should I fix this? (My model are trained by Kohya_ss)
KHave you tried the Detail Tweaker Lora on CivitAI yet? And some other upscale extensions of SD?
U@Furkan Gözükara SECourses as a follow up to information I offered about 2.X control nets last week, I found these today and thought maybe they'd be of interest to you as well https://huggingface.co/thibaud/controlnet-sd21/tree/main
Ci made a thing
Ci think i needed to adjust a few things, but it was jsut to get a feel for the process, seem to be having problems with the python script to rename the files, so i didn't upscale it
Cwhich probably would have done wonders for the face
Si tried upsclaing the effect is not obvious, i haven't try the Detail Tweaker Lora , do you have some insight?
Ci haven't tried to use Loras
Cbut i get a feeling they would probably rein in styles a bit if used with this, maybe?
Cis restore faces a Lora?
Cbut like just a built in one?