on wan 2.2 preset not working
on wan 2.2 preset not working
NThank you in advanced. I'm almost ashamed to post it for fear I forgot something simple. I put the read me into chatgpt to try and help me create the command line and the dataset.toml file: accelerate launch --mixed_precision bf16 --num_cpu_threads_per_process 1 flux_train_network.py ^
--pretrained_model_name_or_path "D:/APPS/KohyaSD/TestModels/Chroma.safetensors" ^
--clip_l "D:/APPS/KohyaSD/TestsModels/clip_l.safetensors" ^
--t5xxl "D:/APPS/KohyaSD/TestModels/model-00001-of-00002.safetensors" ^
--ae "D:/APPS/KohyaSD/TestModels/diffusion_pytorch_model.safetensors" ^
--model_type chroma ^
--apply_t5_attn_mask ^
--cache_latents_to_disk ^
--save_model_as safetensors ^
--sdpa ^
--persistent_data_loader_workers ^
--max_data_loader_n_workers 2 ^
--seed 42 ^
--gradient_checkpointing ^
--mixed_precision bf16 ^
--save_precision bf16 ^
--network_module networks.lora_flux ^
--network_dim 4 ^
--network_train_unet_only ^
--optimizer_type adamw8bit ^
--learning_rate 1e-4 ^
--cache_text_encoder_outputs ^
--cache_text_encoder_outputs_to_disk ^
--highvram ^
--max_train_epochs 4 ^
--save_every_n_epochs 1 ^
--dataset_config "D:/APPS/KohyaSD/TestCFG/C4TS_dataset_multi_bs.toml" ^
--output_dir "D:/APPS/KohyaSD/TestOutput" ^
--output_name chroma-lora-name ^
--timestep_sampling shift ^
--discrete_flow_shift 3.1582 ^
--model_prediction_type raw ^
--guidance_scale 1.0
--pretrained_model_name_or_path "D:/APPS/KohyaSD/TestModels/Chroma.safetensors" ^
--clip_l "D:/APPS/KohyaSD/TestsModels/clip_l.safetensors" ^
--t5xxl "D:/APPS/KohyaSD/TestModels/model-00001-of-00002.safetensors" ^
--ae "D:/APPS/KohyaSD/TestModels/diffusion_pytorch_model.safetensors" ^
--model_type chroma ^
--apply_t5_attn_mask ^
--cache_latents_to_disk ^
--save_model_as safetensors ^
--sdpa ^
--persistent_data_loader_workers ^
--max_data_loader_n_workers 2 ^
--seed 42 ^
--gradient_checkpointing ^
--mixed_precision bf16 ^
--save_precision bf16 ^
--network_module networks.lora_flux ^
--network_dim 4 ^
--network_train_unet_only ^
--optimizer_type adamw8bit ^
--learning_rate 1e-4 ^
--cache_text_encoder_outputs ^
--cache_text_encoder_outputs_to_disk ^
--highvram ^
--max_train_epochs 4 ^
--save_every_n_epochs 1 ^
--dataset_config "D:/APPS/KohyaSD/TestCFG/C4TS_dataset_multi_bs.toml" ^
--output_dir "D:/APPS/KohyaSD/TestOutput" ^
--output_name chroma-lora-name ^
--timestep_sampling shift ^
--discrete_flow_shift 3.1582 ^
--model_prediction_type raw ^
--guidance_scale 1.0
JWhats the error you get?
JI think the --t5xxl needs to be a single safetensor file or pth file, and not a partial file (00001 of 00002)
JAlso discrete_flow_shift maybe rounded number?
Jdid you manually edit the datset.toml? how does it look?
Fwhich preset exactly you using?
NYes, I edited the toml file. Also I am working off a 4090. It looks like it attempts to run. I get the error: File "D:\APPS\KohyaSD\sd-scripts-sd3\venv\lib\site-packages\torch\nn\modules\module.py", line 1362, in convert
raise NotImplementedError(
NotImplementedError: Cannot copy out of meta tensor; no data! Please use torch.nn.Module.to_empty() instead of torch.nn.Module.to() when moving module from meta to a different device.
Traceback (most recent call last):
File "D:\PROG\Python310\lib\runpy.py", line 196, in _run_module_as_main
return _run_code(code, main_globals, None, and Toml: [general]
color_aug = false
keep_tokens_separator = "|||"
caption_extension = ".txt"
[[datasets]]
enable_bucket = true
resolution = [1024, 1024]
[[datasets.subsets]]
image_dir = "D:/APPS/KohyaSD/TestIMG/10_C4TS cats"
num_repeats = 1 A suggestion from chatgpt was to move to bf16. (on the two lines in the bat file) I changes that is and looks like it attempts to load the models. But then another set of errors. It's like moving from one set of issues to another and I'm afraid I'm just working in the wrong direction.
raise NotImplementedError(
NotImplementedError: Cannot copy out of meta tensor; no data! Please use torch.nn.Module.to_empty() instead of torch.nn.Module.to() when moving module from meta to a different device.
Traceback (most recent call last):
File "D:\PROG\Python310\lib\runpy.py", line 196, in _run_module_as_main
return _run_code(code, main_globals, None, and Toml: [general]
Common settings for the dataset
flip_aug = truecolor_aug = false
keep_tokens_separator = "|||"
caption_extension = ".txt"
[[datasets]]
This is the dataset definition for training
batch_size = 2enable_bucket = true
resolution = [1024, 1024]
[[datasets.subsets]]
image_dir = "D:/APPS/KohyaSD/TestIMG/10_C4TS cats"
num_repeats = 1 A suggestion from chatgpt was to move to bf16. (on the two lines in the bat file) I changes that is and looks like it attempts to load the models. But then another set of errors. It's like moving from one set of issues to another and I'm afraid I'm just working in the wrong direction.
Fyou are trying to train flux?
NNo, I would ultimately like to train Chroma.
Fyou can use gui
Fand add some extra params
FGitHub
Thank you lodestones for the high quality model.
--model_type option is added to flux_train_network.py. For Chroma, --model_type chroma is required.
According to lodestones, --apply_t5_attn_mask --...
--model_type option is added to flux_train_network.py. For Chroma, --model_type chroma is required.
According to lodestones, --apply_t5_attn_mask --...
NThank you I will review that.
SHey Doc @Furkan Gözükara SECourses
Are there any plans for Fine-Tuning or LoRA for Wan2.2? If not in the near future, where would you recommend looking to learn more about that field?
Are there any plans for Fine-Tuning or LoRA for Wan2.2? If not in the near future, where would you recommend looking to learn more about that field?
Fyes i plan first qwen image then wan 2.2
Sthank you
Falso will publish results of flux dev vs flux krea dev
Ftoday . generating massive grid
F
Awan 2.2 lightx2 lora
Abut i want to switch to kijai new not the wan 2.1 old
Ayou need to update
Fwhich one here? also i will update wan 2.2 presets soon after testing new loras hopefully
F
Fi can also make chroma workflow i think
Athis
A
Aneeds update
Awith kijai lightning lora
Fwill check thanks
Fcan you give link
SVery interested in your feedbacks before trying a full kra finetune. How many images for your test ?
Djust went through your video on the sdxl lora training with one trainer.. im new to one trainer.. barely a week old... i wanna pay for your patreon and wanted to ask if you updated the presets you mentioned.. i have about 60 images i wanna train them for a person but i cant find regularization images for a female.. so im abit stuck there
FPatreon
Get more from SECourses: FLUX, Tutorials, Guides, Resources, Training, Scripts on Patreon
Fdataset published too
F
Fsdxl presets are updated at the moment
Fyou can see updates without subscription i just made that change : https://www.patreon.com/posts/96028218
Patreon
Get more from SECourses: FLUX, Tutorials, Guides, Resources, Training, Scripts on Patreon
Dawesome... will just follow the step by step instructions in the video
Dthanks so much.. this is a great help
Fnot a cherry pick
Fflux krea dev
F
Fat some prompts it works better than flux dev at some prompts worse
F
SMan thank you so much I got confused by the many negative feedbacks regarding krea training. From your tests it seems to be a huge improvement over flux.dev. Do you think it would do well with larger datasets as well (200 images ?)
Athere is
Adont listen to what they say about krea
Fwell it should i think. i trained flux dev with 256 images
