S
Y¿guys whats the best way to train faces? i got a 8gb vram ¿ is LoRA still best solution? or dreamboth with google colabs?, thanks 
Fnice
Fi will also test now
Fdreambooth colab
Fbest
FYouTubeSoftware Engineering Courses - SE Courses
Our Discord : https://discord.gg/HbqgGaZVmr. How to do free Stable Diffusion DreamBooth training on Google Colab for free.
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
Stable Diffusion Playlist: https://www.youtube.com/playlist?l...
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/SECoursesStable Diffusion Playlist: https://www.youtube.com/playlist?l...
FYouTubeSoftware Engineering Courses - SE Courses
Discord : https://discord.gg/HbqgGaZVmr. This is the video where you will learn how to use Google Colab for Stable Diffusion. 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
Google colab notebook link: https://colab.research.google....
https://www.patreon.com/SECoursesGoogle colab notebook link: https://colab.research.google....
LHello, I followed the Dreambooth Jan 22 tutorial with exact same SD & DB commit, but got all black starting from epoch 4, has anyone successfully trained dreambooth locally yet?
YouTubeSoftware Engineering Courses - SE Courses
Our Discord : https://discord.gg/HbqgGaZVmr. Newest update of DreamBooth extension of Automatic1111 brought huge quality and success improvement.
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 have explained how t...
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/SECoursesIn this video, I have explained how t...
Fbecause of xformers version
Fhere how to fix https://www.youtube.com/watch?v=O01BrQwOd-Q
YouTubeSoftware Engineering Courses - SE Courses
Our Discord : https://discord.gg/HbqgGaZVmr. This video I am showing how to downgrade CUDA version for proper training and I am showing how to do LoRA training with 8GB GPU.
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
This CUDA ...
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/SECoursesThis CUDA ...
LThank you for the reply, I'm using following xformer, is it the buggy version?
Python revision: 3.10.10 (main, Feb 8 2023, 14:50:01) [GCC 9.4.0]
Dreambooth revision: fd51c0b2ed20566c60affa853a32ebce1b0a1139
SD-WebUI revision: d8f8bcb821fa62e943eb95ee05b8a949317326fe
[+] bitsandbytes version 0.35.0 installed
[+] diffusers version 0.10.2 installed.
[+] transformers version 4.25.1 installed.
[+] xformers version 0.0.16rc425 installed.
[+] torch version 1.13.1+cu117 installed.
[+] torchvision version 0.14.1+cu117 installed.
Python revision: 3.10.10 (main, Feb 8 2023, 14:50:01) [GCC 9.4.0]
Dreambooth revision: fd51c0b2ed20566c60affa853a32ebce1b0a1139
SD-WebUI revision: d8f8bcb821fa62e943eb95ee05b8a949317326fe
[+] bitsandbytes version 0.35.0 installed
[+] diffusers version 0.10.2 installed.
[+] transformers version 4.25.1 installed.
[+] xformers version 0.0.16rc425 installed.
[+] torch version 1.13.1+cu117 installed.
[+] torchvision version 0.14.1+cu117 installed.
Fdreambooth still not supported in that xformers
F
YThank you so much friend, grettings from México, great tutorials been learning alot
Fty so much for feedback. if you like subscribe share that means a lot
Fand leave a comment
Fgold information
HWow, makes sense. The SD 2.1 model I use was made by someone who also trained the text encoder
HSo Controlnet seems like a more powerful Inpaint / Img2Img feature?
Fi think yes
Cwhat is controlnet?
HDr F new vid
Fye it is new amazing thing watch it i suggest
DSalam aleikum, I have an error
DFile "/usr/local/lib/python3.8/dist-packages/diffusers/models/attention.py", line 104, in set_use_memory_efficient_attention_xformers
raise ValueError(
ValueError: torch.cuda.is_available() should be True but is False. xformers' memory efficient attention is only available for GPU
Traceback (most recent call last):
File "/usr/local/bin/accelerate", line 8, in <module>
sys.exit(main())
File "/usr/local/lib/python3.8/dist-packages/accelerate/commands/accelerate_cli.py", line 45, in main
args.func(args)
File "/usr/local/lib/python3.8/dist-packages/accelerate/commands/launch.py", line 1097, in launch_command
simple_launcher(args)
File "/usr/local/lib/python3.8/dist-packages/accelerate/commands/launch.py", line 552, in simple_launcher
raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
subprocess.CalledProcessError: Command '['/usr/bin/python3', 'train_dreambooth.py', '--pretrained_model_name_or_path=runwayml/stable-diffusion-v1-5', '--pretrained_vae_name_or_path=stabilityai/sd-vae-ft-mse', '--output_dir=/content/drive/MyDrive/stable_diffusion_weights/ilshatgael', '--revision=fp16', '--with_prior_preservation', '--prior_loss_weight=1.0', '--seed=1337', '--resolution=512', '--train_batch_size=1', '--train_text_encoder', '--mixed_precision=fp16', '--use_8bit_adam', '--gradient_accumulation_steps=1', '--learning_rate=1e-6', '--lr_scheduler=constant', '--lr_warmup_steps=128', '--num_class_images=192', '--sample_batch_size=4', '--max_train_steps=1280', '--save_interval=10000', '--save_sample_prompt=ilshatgael', '--concepts_list=concepts_list.json']' returned non-zero exit status 1.
raise ValueError(
ValueError: torch.cuda.is_available() should be True but is False. xformers' memory efficient attention is only available for GPU
Traceback (most recent call last):
File "/usr/local/bin/accelerate", line 8, in <module>
sys.exit(main())
File "/usr/local/lib/python3.8/dist-packages/accelerate/commands/accelerate_cli.py", line 45, in main
args.func(args)
File "/usr/local/lib/python3.8/dist-packages/accelerate/commands/launch.py", line 1097, in launch_command
simple_launcher(args)
File "/usr/local/lib/python3.8/dist-packages/accelerate/commands/launch.py", line 552, in simple_launcher
raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
subprocess.CalledProcessError: Command '['/usr/bin/python3', 'train_dreambooth.py', '--pretrained_model_name_or_path=runwayml/stable-diffusion-v1-5', '--pretrained_vae_name_or_path=stabilityai/sd-vae-ft-mse', '--output_dir=/content/drive/MyDrive/stable_diffusion_weights/ilshatgael', '--revision=fp16', '--with_prior_preservation', '--prior_loss_weight=1.0', '--seed=1337', '--resolution=512', '--train_batch_size=1', '--train_text_encoder', '--mixed_precision=fp16', '--use_8bit_adam', '--gradient_accumulation_steps=1', '--learning_rate=1e-6', '--lr_scheduler=constant', '--lr_warmup_steps=128', '--num_class_images=192', '--sample_batch_size=4', '--max_train_steps=1280', '--save_interval=10000', '--save_sample_prompt=ilshatgael', '--concepts_list=concepts_list.json']' returned non-zero exit status 1.
DDid all step by step as in video (Dreambooth colab)
HYeah this is better than I thought.
With each new update now I see the benefit in all the time spent studying. Seems like things will begin to get even more advanced.
With each new update now I see the benefit in all the time spent studying. Seems like things will begin to get even more advanced.
DGuys somebody help me please
Yhi, i wanna install diferent models, wich version of anything v3.0 i have to download? fp16? fp32? or full?
THey dr Furkan, i saw your video and you had txt docs besides all of your sample images, is that necessary ? If not what would happen if we don't go through the hassle of creating a txt doc filled with tags for each image and just put them all into dreambooth training ? How does that work, and does Dreambooth recognise the subject and style the image contains by itself ? Wouldn't it need a description of what it's seeing ? I'm trying to train for a style and i probably have 100 images ready to be trained, so i want clarification on this one
DI solved my issue, thanks to everyone. But I didn't understand why in last step we skip conversion to .ckpt
Fi did run esterday it worked very well. probably your configuration error
Ffp16 . you will use on computer right
Fdont skip it if you will use in
Fyour pc
FYouTubeSoftware Engineering Courses - SE Courses
Discord : https://discord.gg/HbqgGaZVmr. This is the video where you will learn how to use Google Colab for Stable Diffusion. 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
Google colab notebook link: https://colab.research.google....
https://www.patreon.com/SECoursesGoogle colab notebook link: https://colab.research.google....
FYouTubeSoftware Engineering Courses - SE Courses
Our Discord : https://discord.gg/HbqgGaZVmr. Newest update of DreamBooth extension of Automatic1111 brought huge quality and success improvement.
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 have explained how t...
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/SECoursesIn this video, I have explained how t...
F@Kronos welcome
Fafter watching that dreambooth tutorial on colab
FYouTubeSoftware Engineering Courses - SE Courses
Discord : https://discord.gg/HbqgGaZVmr. This is the video where you will learn how to use Google Colab for Stable Diffusion. 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
Google colab notebook link: https://colab.research.google....
https://www.patreon.com/SECoursesGoogle colab notebook link: https://colab.research.google....
Fyou can generate ckpt and use on your computer with 4 GB vram
Fstart automatic1111 with --lowvram command
DMan, my photos are awful. Strong deformation and ugly(((( Put number of samples 10, CFG 7.5, Steps 50
SCan ControlNet be used to generate consistent animations?
Fprobably over trained. which videos you watched so far?
Fvery probably
Fit is able to keep shape very well
Falmost perfect