and do you recommend masked training over "normal" training?
and do you recommend masked training over "normal" training?
Dalso as far as I remember your dataset is images without a detailed background anyway, right?
Njust had amazing results training a flux fine tune , thank u
Nalso how do I switch fp16 models in flux without this message
XHello, sent you a request
X@BrennenRB
XInteresting!
JI have a dataset of 25 images for this character, and I’m trying to train a LoRA for FLUX. I used the best config with Rank 4 and set the epoch to 100, which resulted in 2500 steps. However, the results from the prompt don’t look like the character I want to portray. Should I increase the epoch to 200? Also, I’m using captions generated by ChatGPT like this: 'GDZHBo_flx, a cartoon character with a large mustache and glasses, sitting with a playful expression as he appears to be gesturing something with his hands, adding a humorous and relaxed vibe.' Any advice?
JThis is an example of the results I’m getting after training the character. The outputs don’t quite match the character I want
DThis works (Saved me tons of space on home PC):
You should be able to change ModelRoot to whatever place on your system you have your models stored and, if the subdirs vary from what SwarmUI uses, modify them here. That should capture unet. It will look in {ModelRoot}/unet or {ModelRoot}/diffusion_models as needed.
You should be able to change ModelRoot to whatever place on your system you have your models stored and, if the subdirs vary from what SwarmUI uses, modify them here. That should capture unet. It will look in {ModelRoot}/unet or {ModelRoot}/diffusion_models as needed.
Fsounds too much but can be tested
Fnice
Fi dont recommend
Fget your training set better :d
Fbut i plan to test
Fon flux too
Fyou can convert model to fp8
FPatreon
Get more from SECourses: FLUX, Tutorials, Guides, Resources, Training, Scripts on Patreon
Fdont caption in details
Fit reduces likeliness
Fand you already dont have many images
Jthank, I will try with more images about 40 images, should I keep epoch 200 or lower epoch?
Fso you can be sure it is trained enough
SI created a pod from a network volume and now want to store files in the volume
Swhere do I find the network volume inside the pod when I'm ssh'd in ?
Fyou can ask this in runpod server i never done
FFi just found a new error of bmaltais
Fnever ending :d
FGitHub
Hugging face hub has to be 0.25.2 or doesnt work man these errors always happens to me when recording a tutorial video @kohya-ss @bmaltais
Ti would disagree partially. u wanna caption clothing, background etc.
but not for what the lora should learn from the images
but not for what the lora should learn from the images
Fwell flux has internal captioning mechanism
Fthat is why i dont recommned for flux
Tit's just.... i trained same 10img lora of myself once without captions and once with minimal captions. and it's so hard to say which is better side by side... subjective.
i need some more xy grids
i need some more xy grids
Tdid BFL tell you? ^^
Fwell it is discovered by several scripts developers
Fby analyzing its architecture
Fit generates image embeddings
Fwhich makes captioning effect
Tinteresting. thanks for sharing
P@Dr. Furkan Gözükara Hello everyone, I have a question about custom lora training on kohya using flux
I trained my face with 21 photos and 142 epochs as you mention on the equation on your video, which is... 3.000 / (number of photos) = number of epochs.
When I write the prompt on SwarmUI using Flux.1-dev-fp8. with 1024x1024 resolution the picture looks with low quality, also when the size is 1024+ seems to appear some artifact vertical lines on the picture (see the image) . I'm thinking to train again from zero a new lora with more images or use finetuning. What do you recommend?
I trained my face with 21 photos and 142 epochs as you mention on the equation on your video, which is... 3.000 / (number of photos) = number of epochs.
When I write the prompt on SwarmUI using Flux.1-dev-fp8. with 1024x1024 resolution the picture looks with low quality, also when the size is 1024+ seems to appear some artifact vertical lines on the picture (see the image) . I'm thinking to train again from zero a new lora with more images or use finetuning. What do you recommend?
Ffine tuning prevents this
Fbut this usually happens when you generate high resolution
Fah yes you said 1024+
Fyes fine tuning reduces this a lot
Falso you can use SUPIR to upscale
Fsupir face enhance difference
F
Foriginal vs supir https://imgsli.com/MzA4NzQx/0/2
