I've been training with Kohya SS GUI. why automatic 1111 dreambooth is preferred over that?

T

Hi guys,

Everything was working fine yesterday (created many images), but today after I created one single image successfully, all the other ones, they process (i can see the preview) but when they reach 99%->100% it gets CUDA out of memory (among other texts above that I can share) and the image doesn't process.

Tried restarting the PC and starting everything all over (and not using the PC for anything else).

Does anyone had a similar issue?

Laptop with rtx3080 8gb gddr6 vram

Message:

torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 1024.00 MiB (GPU 0; 8.00 GiB total capacity; 6.53 GiB already allocated; 0 bytes free; 6.63 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation.  See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF

torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 1024.00 MiB (GPU 0; 8.00 GiB total capacity; 6.53 GiB already allocated; 0 bytes free; 6.63 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation.  See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF

torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 1024.00 MiB (GPU 0; 8.00 GiB total capacity; 6.53 GiB already allocated; 0 bytes free; 6.63 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation.  See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF

torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 1024.00 MiB (GPU 0; 8.00 GiB total capacity; 6.53 GiB already allocated; 0 bytes free; 6.63 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation.  See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF

Arguments:

 --xformers --no-half --ckptfix

 --xformers --no-half --ckptfix

V

vgtchdwn•3/21/23, 3:17 AM

Hello, I am new here. I want to use dreambooth to train a personal model and inpainting to add it to an existing image. Is this possible? Are there any tutorials available?

S

Sinclair•3/21/23, 3:32 AM

EDIT: I had an error during the installation process, but I think I've fixed it! I'll pop back in, if another problems comes up.

EEssaBalls It's working now for me. I replied on their channel 🙂

A

AlxHsInf•3/21/23, 4:45 AM

@Furkan Gözükara SECourses so torch2 is working on 20XX series Nvidia Graphics cards now?

A

AlxHsInf•3/21/23, 4:46 AM

Do I just need to follow the same tutorial and try the same process again? I have a NVIDIA GeForce RTX 2080 SUPER

T

TheArtist•3/21/23, 5:19 AM

I followed the tutorial for installing dreambooth extension for webui on runpod. During the relaunch, I am getting this error: ModuleNotFoundError: No module named 'wrapt'

FFurkan Gözükara SECourses you cant use multiple gpu for a single training with automatic1111 or existing s...

T

Trimac•3/21/23, 5:50 AM

Sounds like youve already tested it. Im using pcie rsier connected in 16x slot and lets see if get something here.

FFurkan Gözükara SECourses if you could upvote this i would appreciate very much : https://www.reddit.com/r...

T

Trimac•3/21/23, 6:12 AM

How about you train a model like Mj? That would be next level for Sd

F

Fabricatedgirls•3/21/23, 8:49 AM

Hi i keep getting an error while compiling checkpoints: OSError: Error no file named diffusion_pytorch_model.bin found in directory C:\AI\stable-diffusion-webui\models\dreambooth\TEST_032123\checkpoints\checkpoint-19344.

F

Fabricatedgirls•3/21/23, 8:50 AM

xception compiling checkpoint: Error no file named diffusion_pytorch_model.bin found in directory C:\AI\stable-diffusion-webui\models\dreambooth\TEST_032123\checkpoints\checkpoint-19344.
Traceback (most recent call last):
File "C:\AI\stable-diffusion-webui\extensions\sd_dreambooth_extension\dreambooth\diff_to_sd.py", line 434, in compile_checkpoint
unet_model = UNet2DConditionModel().from_pretrained(os.path.dirname(unet_path))
File "C:\AI\stable-diffusion-webui\venv\lib\site-packages\diffusers\models\modeling_utils.py", line 527, in from_pretrained
model_file = _get_model_file(
File "C:\AI\stable-diffusion-webui\venv\lib\site-packages\diffusers\models\modeling_utils.py", line 821, in _get_model_file
raise EnvironmentError(

F

Fabricatedgirls•3/21/23, 8:51 AM

any idea what the problem might be? it's training but its not generating checkpoints

F

Fabricatedgirls•3/21/23, 8:51 AM

it gives an error then keeps training, but obviously unusable if i can't generate checkpoints!

ZZeroCool22 I can confirm that now it's working (1080 TI), i didn't do any extra steps, just...

F

Furkan Gözükara SECourses•3/21/23, 9:04 AM

it was torch 2 update

KKJ I've been training with Kohya SS GUI. why automatic 1111 dreambooth is preferred...

F

Furkan Gözükara SECourses•3/21/23, 9:04 AM

i say automatic1111 is better now but people have past bad experience

TThe Laughing Rottie Hi guys, Everything was working fine yesterday (created many images), but toda...

F

Furkan Gözükara SECourses•3/21/23, 9:05 AM

add --medvram and try

Vvgtchdwn Hello, I am new here. I want to use dreambooth to train a personal model and inp...

F

Furkan Gözükara SECourses•3/21/23, 9:05 AM

yes i did that in this video

F

Furkan Gözükara SECourses•3/21/23, 9:05 AM

https://youtu.be/m-UVVY_syP0

YouTubeSECourses - Software Engineering Courses

Training Midjourney Level Style And Yourself Into The SD 1.5 Model ...

RunPod: https://bit.ly/RunPodIO. Discord : https://bit.ly/SECoursesDiscord. Get your own #Midjourney level model for free via Style Teaching. If I have been of assistance to you and you would like to show your support for my work, please consider becoming a patron on

https://www.patreon.com/SECourses

Playlist of #StableDiffusion Tutorials, Au...

F

Furkan Gözükara SECourses•3/21/23, 9:05 AM

but if you are new you need to watch more to learn more

TTrimac How about you train a model like Mj? That would be next level for Sd

F

Furkan Gözükara SECourses•3/21/23, 9:06 AM

you can do that by following this but you need training images

FFabricatedgirls Hi i keep getting an error while compiling checkpoints: OSError: Error no file n...

F

Furkan Gözükara SECourses•3/21/23, 9:07 AM

hi welcome. how much vram and ram do you have?

F

Fabricatedgirls•3/21/23, 9:07 AM

24GB vram, 128GB ram

F

Fabricatedgirls•3/21/23, 9:07 AM

3090

FFabricatedgirls 3090

F

Furkan Gözükara SECourses•3/21/23, 9:07 AM

then this is not vram or ram problem

F

Furkan Gözükara SECourses•3/21/23, 9:07 AM

could be installation problem

F

Fabricatedgirls•3/21/23, 9:08 AM

no i think something must have got broken updating

F

Furkan Gözükara SECourses•3/21/23, 9:08 AM

are you using torch 2?

F

Fabricatedgirls•3/21/23, 9:08 AM

not sure which torch version, i think i may have had to roll it back before a couple weeks agp

F

Fabricatedgirls•3/21/23, 9:08 AM

ago

F

Furkan Gözükara SECourses•3/21/23, 9:08 AM

i suggest to do fresh install

F

Fabricatedgirls•3/21/23, 9:08 AM

should i be using latest?

F

Furkan Gözükara SECourses•3/21/23, 9:08 AM

i explained in this video

F

Furkan Gözükara SECourses•3/21/23, 9:09 AM

https://youtu.be/pom3nQejaTs

YouTubeSECourses - Software Engineering Courses

How To Install New DREAMBOOTH & Torch 2 On Automatic1111 Web UI PC ...

Discord : https://bit.ly/SECoursesDiscord. Torch 2 / Pytorch 2 is now supported along with new DreamBooth Automatic1111 Web UI extension. If I have been of assistance to you and you would like to show your support for my work, please consider becoming a patron on

https://www.patreon.com/SECourses

In this video, I am explaining how you can man...

F

Furkan Gözükara SECourses•3/21/23, 9:09 AM

first make it work default then you can try torch 2

F

Fabricatedgirls•3/21/23, 9:09 AM

fresh install of all of A1111

F

Fabricatedgirls•3/21/23, 9:09 AM

?

FFabricatedgirls fresh install of all of A1111

F

Furkan Gözükara SECourses•3/21/23, 9:09 AM

yes both that and dreambooth usually solves many problems

F

Furkan Gözükara SECourses•3/21/23, 9:09 AM

dont install any extension

F

Fabricatedgirls•3/21/23, 9:09 AM

damn, so many added extensions and models and extra stuff

F

Furkan Gözükara SECourses•3/21/23, 9:09 AM

until you make it work

F

Furkan Gözükara SECourses•3/21/23, 9:09 AM

extensions can also conflict

F

Fabricatedgirls•3/21/23, 9:10 AM

installed it sucks to ditch everything

F

Furkan Gözükara SECourses•3/21/23, 9:10 AM

give it a try. dont delete other one

F

Furkan Gözükara SECourses•3/21/23, 9:10 AM

also do you guys prefer faster videos where i cut waiting or you prefer longer ones? e.g. this is a fast one : https://youtu.be/m-UVVY_syP0

YouTubeSECourses - Software Engineering Courses

Training Midjourney Level Style And Yourself Into The SD 1.5 Model ...

RunPod: https://bit.ly/RunPodIO. Discord : https://bit.ly/SECoursesDiscord. Get your own #Midjourney level model for free via Style Teaching. If I have been of assistance to you and you would like to show your support for my work, please consider becoming a patron on

https://www.patreon.com/SECourses

Playlist of #StableDiffusion Tutorials, Au...

F

Furkan Gözükara SECourses•3/21/23, 9:10 AM

this is a longer one : https://youtu.be/Bdl-jWR3Ukc

YouTubeSECourses - Software Engineering Courses

Zero To Hero Stable Diffusion DreamBooth Tutorial By Using Automati...

Our Discord : https://discord.gg/HbqgGaZVmr. The most advanced tutorial of Stable Diffusion Dreambooth Training. If I have been of assistance to you and you would like to show your support for my work, please consider becoming a patron on

https://www.patreon.com/SECourses

Playlist of Stable Diffusion Tutorials, Automatic1111 and Google Colab ...

F

Fabricatedgirls•3/21/23, 9:11 AM

is there a torch2 only install video?

FFabricatedgirls is there a torch2 only install video?

F

Furkan Gözükara SECourses•3/21/23, 9:11 AM

yes i have here. https://youtu.be/pom3nQejaTs

YouTubeSECourses - Software Engineering Courses

How To Install New DREAMBOOTH & Torch 2 On Automatic1111 Web UI PC ...

Discord : https://bit.ly/SECoursesDiscord. Torch 2 / Pytorch 2 is now supported along with new DreamBooth Automatic1111 Web UI extension. If I have been of assistance to you and you would like to show your support for my work, please consider becoming a patron on

https://www.patreon.com/SECourses

In this video, I am explaining how you can man...

F

Fabricatedgirls•3/21/23, 9:12 AM

thank you

F

Fabricatedgirls•3/21/23, 9:39 AM

well updating torch alone didn't fix the checkpoint creation problem, damn was really hoping it would be easy lol

FFabricatedgirls well updating torch alone didn't fix the checkpoint creation problem, damn was r...

F

Furkan Gözükara SECourses•3/21/23, 9:39 AM

if you are our patreon subscriber i can manually install it on your computer via google remote desktop

F

Fabricatedgirls•3/21/23, 10:02 AM

damn, deleting dreambooth and reinstalling it didnt help, updating torch/cuda didnt help. They dont make it easy

I've been training with Kohya SS GUI. why automatic 1111 dreambooth is preferred over that?

Similar Threads