4090 much stronger but not working very good on windows. probably 3090 has the best train / price pe
4090 much stronger but not working very good on windows. probably 3090 has the best train / price performance
Fon my todo list hopefully
Fyou can use free kaggle with our notebook
Fthis is something i plan to investigate too photo restoration
Fbut i didnt have much time yet
Fi think kohya also can read special captions toml files but i never bothered to learn
Z
Sthere is any command line to upload local files to runpod server? I download my model and I want to upload it is 1.7 gb
Fyes runpodctl
For upload to hugging face and use wget
MYou have misunderstood something. If a folder is named "2_ohwx man", it will only duplicate the images in it, not the txt files...
S@Dr. Furkan Gözükara do you use
no_half_vae
Fi dont use anymore
Fwith fp16 vae
SThanks!
Mupdate - I just updated from 3060 12 gb vram to 4090 24gb vram and WOW. It is amazingly fast and visibly better quality. Switched back to Windows 10 before upgrade so I don't have a comparison to W11. Amazing! btw I haven't tried training yet since I follow your runpod advice and use 48 gb vram to train.
Fi also noticed this
F--medvram produces different images
TPython is using 24GB of RAM when generating images with A1111. Is that normal?
TI also noticed the jump in quality. It has incredible details without hires fix
Fi think caching
Fso normal
EIf you put of a survey of most wanted tutorials i think itll maybe easier for you to see priorities (for patreon supporters)
Eso no bugs using 4090 with automatic1111\kohya ?
Vhello, all, for the training note book, i found out no place to like this part, tags
Vi mean on kaggle note book, where to input --caption_extension
Vi do not see
MHello, Started to follow your course on youtube for quality studio images,
when installed kohya SS gui I don't have dreambooth Lora but only LORA , should I continue ? @Dr. Furkan Gözükara
when installed kohya SS gui I don't have dreambooth Lora but only LORA , should I continue ? @Dr. Furkan Gözükara
Mis that normal ?
Vhello, all what is the trigger word of lora? instance name ? if i want to use revanimated as base model to train 1.5 lora, also huggingface repo id?
MHey Virginia, does Google Collab free again ?? it was recently changed to not free anymore
D
DDHow to make it public
DI am using mobile device
Vno colab cannot be used
SIt should only say Lora at the top
V1024 sdxl fail again ------------------------------------------------------------
Failures:
[1]:
time : 2023-10-10_15:03:15
host : 19fe5077aa6e
rank : 1 (local_rank: 1)
exitcode : 2 (pid: 1098)
error_file: <N/A>
traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html
------------------------------------------------------------
Root Cause (first observed failure):
[0]:
time : 2023-10-10_15:03:15
host : 19fe5077aa6e
rank : 0 (local_rank: 0)
exitcode : 2 (pid: 1097)
error_file: <N/A>
traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html
Failures:
[1]:
time : 2023-10-10_15:03:15
host : 19fe5077aa6e
rank : 1 (local_rank: 1)
exitcode : 2 (pid: 1098)
error_file: <N/A>
traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html
------------------------------------------------------------
Root Cause (first observed failure):
[0]:
time : 2023-10-10_15:03:15
host : 19fe5077aa6e
rank : 0 (local_rank: 0)
exitcode : 2 (pid: 1097)
error_file: <N/A>
traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html
Vtrain_network.py: error: argument --max_data_loader_n_workers: expected one argument
train_network.py: error: argument --max_data_loader_n_workers: expected one argument
Traceback (most recent call last):
File "/opt/conda/bin/accelerate", line 8, in <module>
sys.exit(main())
File "/opt/conda/lib/python3.10/site-packages/accelerate/commands/accelerate_cli.py", line 47, in main
args.func(args)
File "/opt/conda/lib/python3.10/site-packages/accelerate/commands/launch.py", line 977, in launch_command
multi_gpu_launcher(args)
File "/opt/conda/lib/python3.10/site-packages/accelerate/commands/launch.py", line 646, in multi_gpu_launcher
distrib_run.run(args)
File "/opt/conda/lib/python3.10/site-packages/torch/distributed/run.py", line 785, in run
elastic_launch(
File "/opt/conda/lib/python3.10/site-packages/torch/distributed/launcher/api.py", line 134, in call
return launch_agent(self._config, self._entrypoint, list(args))
File "/opt/conda/lib/python3.10/site-packages/torch/distributed/launcher/api.py", line 250, in launch_agent
raise ChildFailedError(
torch.distributed.elastic.multiprocessing.errors.ChildFailedError:
train_network.py: error: argument --max_data_loader_n_workers: expected one argument
Traceback (most recent call last):
File "/opt/conda/bin/accelerate", line 8, in <module>
sys.exit(main())
File "/opt/conda/lib/python3.10/site-packages/accelerate/commands/accelerate_cli.py", line 47, in main
args.func(args)
File "/opt/conda/lib/python3.10/site-packages/accelerate/commands/launch.py", line 977, in launch_command
multi_gpu_launcher(args)
File "/opt/conda/lib/python3.10/site-packages/accelerate/commands/launch.py", line 646, in multi_gpu_launcher
distrib_run.run(args)
File "/opt/conda/lib/python3.10/site-packages/torch/distributed/run.py", line 785, in run
elastic_launch(
File "/opt/conda/lib/python3.10/site-packages/torch/distributed/launcher/api.py", line 134, in call
return launch_agent(self._config, self._entrypoint, list(args))
File "/opt/conda/lib/python3.10/site-packages/torch/distributed/launcher/api.py", line 250, in launch_agent
raise ChildFailedError(
torch.distributed.elastic.multiprocessing.errors.ChildFailedError:
Ztrying to put my face on this image, but it's not going well, tried regular lora with positive prompt, roop and auto detailer, but never gets the face well
ZI assume it has to do with distance and darkness
AHi friends, I have this problem: Very slow image generation speed when using SDXL 1.0 on RTX 3060 Ti GPU and 16gb RAM. If i generate 1024x1024 with 20 steps then the generation time is 10 minutes, if for example 768x1024 then the generation speed is from 20 minutes. I have seen many people have generation speeds of only a couple minutes on the same GPU. I am using AUTOMATIC1111 version v1.6.0 and the start parameter --opt-sdp-attention. I don't get any errors about lack of VRAM, just a very slow generation speed. Maybe I should do a clean SD installation?
AI enabled sdxl.vae and got a reduction in image generation time. But I got an error. Anyway it's not a 10 minutes wait anymore:)
When generating at 40 steps, the generation time doubles.
When generating at 40 steps, the generation time doubles.
Mnone with A1111 so far and haven't tried kohya yet. Will in the next few days.
MWhat are other good rare tokens to use for training besides ohwx? I would like to train other subjects
MI also just noticed the change in processing units. Never paid attention before but I was getting 1.6 sec/Iteration on 3060 and now I'm getting 0.31 sec/iteration (or 3.21 iterations/sec). 4-5x faster and better quality
AI think I was able to fix the slow generation speed on the 3060 ti. I now generate a 768x1024 image in 30-40 seconds. The problem is with Nvidia drivers from version 531 and up. And also I used --xformers parameter instead of --opt-sdp-attention.
https://github.com/vladmandic/automatic/discussions/1285
https://github.com/vladmandic/automatic/discussions/1285
GitHub
it seems that nVidia changed memory management in the latest versions of drivers, specifically 532 and 535 new behavior is that once gpu vram is exhausted, it will actually use shared memory thus c...
LHey guys
I'm a bit lost, can I use kohya to train model for SD 1.5?
If yes, which one gives better result between Dreambooth and Kohya?
I'm a bit lost, can I use kohya to train model for SD 1.5?
If yes, which one gives better result between Dreambooth and Kohya?
EA6000 training sdxl kohya - 48gb config file, around 8000 steps, 3 hours makes sense?
