220 professional photos, cropped but not down sized, trained at 768, xformers, 8bit adam, bf16, 150
220 professional photos, cropped but not down sized, trained at 768, xformers, 8bit adam, bf16, 150 epochs
Ldreambooth?
Syep
Lwhat do you by "cropped but not down sized,"
Sso i will crop them in closer to the subject so the face stands out more but i don't make them 1080x1080 or anything like that
Lwhat setting did you use for "Save Model Frequency (Epochs)"?
Si just put 50 but i liked the whole 150 best
Sbut so some of my photos i used were closer to 4000x5000 pixels
Lyour models looks amazin
S@Dr. Furkan Gözükara I'm having thiis error "torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 20.00 MiB (GPU 0; 19.68 GiB total capacity; 12.33 GiB already allocated; 19.75 MiB free; 12.42 GiB reserved in total by PyTorc"
Si've been testing for fastest/ laziest way to make good models
SI already check "Gradient checkpointing"
S
Sthis is my runpod server
S50% cartoon model merge
Sdo you have the base picture?
Swhat model do you use?
Sbase picture of what?
Sdid you use images to train this model right?
S
Show many images did you use to train the model? I have a memory error and I could’t test the training 
S@Stefanoid9343 about 220
SWow
Sit's my wife so i had plenty haha
Son google colab I use to use like 17 and it works
SI train my wife image too but like 30
Sand it works
Si trained with about 20 for the model i made of my self but the poses and angles didn't seem nearly as diverse
SI try to follow the new tutorial but I run out or memory error
Syeah needs to be different
Sya i just don't have many professional photos of myself hanging around
Sat this point you don’t need that those models are taking care off
Sya but selfies are all mirror image not real image.
Sthat’s right
Smy working theory atm is that that higher quality images and resolution get better results
Si tried a few celebs to experiment and i didn't like those nearly as much
Sdo you have you own card? or you are using runpod or kaagle
Severything local
Si figured buying a new pc would be cheaper in the long run
Si need to try and experiment on product photography again but i didn't have much luck with products that had text on them in the past
ELooks really good, how many repeats did you do for 220 images?
Eor was it 1 repeat and the rest just epochs?
Si just did 150steps (epochs) not sure what repeat is
Eso no reguleraztion images at all, 220 images * 150 33000 ste[s. did you not get overtraining issues? or did you choose an earlier checkpoint?
NHello, long time ago I made models and already forgot how to customize them in colab,
I used to make initial descriptions in WD14 loras and edit them further. How can I do the same in colab?
If you can, can you share the config for better training accuracy
I want to put the training 768 x768 pixels, base model take anylora from lykon
I used to make initial descriptions in WD14 loras and edit them further. How can I do the same in colab?
If you can, can you share the config for better training accuracy
I want to put the training 768 x768 pixels, base model take anylora from lykon
Sno i didn't like the early checkpoints, used the 33k one
Sonce it's finished i usually merge it with another model i like at about 30% to help clean it up a bit
Every nice workflow 
looks really really good
looks really really goodS@Dr. Furkan Gözükara anyone know if gradio link is working again yet?
Nsdxl gives very good results training specific objects, very versatile
Snice
