yes i will do detailed test
yes i will do detailed test
Flots of shared vrma
Fit must be very slow
Fostris still doesnt support block swap
Finteresting did he add block swap?
Fit is not as slow as much i would say
Fwhat is your training resolution?
Fi would say make all images exact resolution
Fyou probably can train with 2 different aspect ratio
Fmake sure to enable bucketing
Dthank you. would i need both sets of reg images? so in the end 2 image concepts and 2 reg image concepts? or would you just throw them all in one folder (1 image folder, 1 reg folder)
Fgood question. probably having different aspect ratio would help
Fyou training sdxl right?
Dyes sdxl
Fyes it should work
Dcurrently i have two folders, one with 1024x1024 and one with 896x1152. so the question is: should i create one concept in onetrainer with both resolutions in one folder, or separate them and have two concepts
Dand would i add 2 reg image concepts for either resolution
Dor one concept with both reg resolutions inside
Fin one trainer
Fjust add them as 2 concepts
Fand properly set their repeat count
Dso 2 concepts for the images (1024x1024 and 896x1152). i have 50 images each. since it's 2 concepts do i use 50 or 100?
Flogic simple
Fat 1 epoch you have 100 train images
Fthen you should use 100 reg images
F50 from each in that case
Jdue to block swap and vae_cache_cpu enabled.
interestingly enough it made no difference in my case for speed whether i turn it on or not for that 480z360 dataset , tested speeds with and without. (without it fits in dedicated vram without shared mem as well)
(though usualy vram bottleneck does limit speeds in other instances)
interestingly enough it made no difference in my case for speed whether i turn it on or not for that 480z360 dataset , tested speeds with and without. (without it fits in dedicated vram without shared mem as well)
(though usualy vram bottleneck does limit speeds in other instances)
Ji am still messing around with what barely fits in vram, apparently as long as there is about 1gb vram dedicated available it will run full speed (400 watt at 70% power limit)
JFor Wan2.1 T2V using (25 videos) 480x360@61 and (23) images 960x720 fits just perfectly in an RTX 5090. Wondering about the results, but this is the most I was sable to cram into the VRAM without going overboard and not using any offloading / shared vram
J
CYesterday, I extracted a LORA from my Dreambooth model (of my character) that I trained in Kohya GUI on the FLUX.dev model. I wanted to see how the LORA would perform with the Flux Krea model, but unfortunately, it doesn't look great.
I guess I'll have to train a new LORA specifically on the Flux Krea model, or train an entire Dreambooth model from scratch.
I guess I'll have to train a new LORA specifically on the Flux Krea model, or train an entire Dreambooth model from scratch.
JMakes sense I guess
AI can confirm. Lora trained on Dev inferenced on krea is not that good
ABut a checkpoint with krea Lora works
CI'm currently training a test LORA on the wan22 low noise model, specifically a LORA for a person or character. From what I understand, it's best to train it on a low noise model. It's a fast training, only 30 images in diffusion pipe wsl. It should take about 1.5 hours.
Almk how it went
AI tried wan 2.2 t2i and I was very disappointed
Atext was all wrong and textures / quality same as krea
CI will
Ais this musubi
CDiffusion-Pipe - best for me. using many times and created many nice Loras for wan2.1
Fi think kohya lora extraction is not working accurate
Janyone know how to upload finetuned flux dev model ( dreambooth style ) to replicate so i can call it with a api?
Fcheck this
F
Fwe had requested base model support i am not sure if he added
Dcan i run multiple instances of onetrainer(gui) or at least cmd? because i don't seem to find any batch processing in the interface
Dlike if i wanted to train the same dataset with the same settings but different base models (for a lora) it would be nice if it would do them after another over night. right now i run it, wait around 1-2hours and then start a new one