Software Engineering Courses (SECourses)•16mo ago

i think flux can do yes

dsienra•9/18/24, 5:19 PM

@Dr. Furkan Gözükara koha training is running much faster with the new version

dsienra•9/18/24, 5:19 PM

there is some news about this?

FFurkan Gözükara SECourses i think flux can do yes

LW24Ai•9/18/24, 5:19 PM

One user wrote to me that my workout has problems due to the weight of 1gb. For some reason, his models are 18 mb each. I have seen such models with a similar weight, but they did not bring me high-quality results and differences

LLW24Ai One user wrote to me that my workout has problems due to the weight of 1gb. For ...

Furkan Gözükara SECoursesOP•9/18/24, 5:31 PM

single layer

Furkan Gözükara SECoursesOP•9/18/24, 5:31 PM

reduces quality

Furkan Gözükara SECoursesOP•9/18/24, 5:31 PM

they dont test anything properly to compare

Ddsienra @Dr. Furkan Gözükara koha training is running much faster with the new version

Furkan Gözükara SECoursesOP•9/18/24, 5:31 PM

which version? torch 2.5?

LLW24Ai One user wrote to me that my workout has problems due to the weight of 1gb. For ...

Furkan Gözükara SECoursesOP•9/18/24, 5:34 PM

you can use lora resize of kohya and see quality difference

FFurkan Gözükara SECourses you can use lora resize of kohya and see quality difference

LW24Ai•9/18/24, 5:36 PM

thanks

Zono50•9/18/24, 5:40 PM

@Dr. Furkan Gözükara on supir app, if guidance scale needs to be 2, where do i set that? Text guidance scale?

FFurkan Gözükara SECourses which version? torch 2.5?

dsienra•9/18/24, 5:53 PM

I'm not sure, I updated to train individual layers but with not a good success, now I training all layers and is super fast

dsienra•9/18/24, 5:53 PM

rn_image_picker_lib_temp_2cd97714-df2b-4b42-93c9-2f5e4e32f5c0.jpg

Ddsienra Click to see attachment

dxqb•9/18/24, 5:54 PM

resolution? bs? card?

dsienra•9/18/24, 5:56 PM

dsienra•9/18/24, 5:56 PM

1024x1024

dsienra•9/18/24, 5:57 PM

dsienra•9/18/24, 5:58 PM

dsienra•9/18/24, 5:58 PM

1024x1024 rtx 3090

dsienra•9/18/24, 5:59 PM

this is normal?

dxqb•9/18/24, 6:01 PM

no, for 1024 that's fast

dsienra•9/18/24, 6:02 PM

the checkpoint is 1.18gb so I'm sure I'm training all the layers

dxqb•9/18/24, 6:04 PM

what did you have beefore the pytorch upgrade?

dsienra•9/18/24, 6:07 PM

false alarm sorry

dsienra•9/18/24, 6:07 PM

my dataset es low ress

dsienra•9/18/24, 6:07 PM

thats why

dsienra•9/18/24, 6:08 PM

I have bukets enable

dxqb•9/18/24, 6:08 PM

ah yeah, no upscale on kohya. resolution makes a major difference on flux

dxqb•9/18/24, 6:12 PM

what resolution was it actually?

dsienra•9/18/24, 6:14 PM

some 768 some 512

dsienra•9/18/24, 6:15 PM

not all 1:1 diferent aspect ratios

FFurkan Gözükara SECourses please ask there

Marrvelous•9/18/24, 6:18 PM

Thanks. I posted there but there is a long conversation already happening so probably missed. I'll try posting again later

ZZono50 @Dr. Furkan Gözükara on supir app, if guidance scale needs to be 2, where do i ...

Furkan Gözükara SECoursesOP•9/18/24, 8:10 PM

yes

Ddsienra Click to see attachment

Furkan Gözükara SECoursesOP•9/18/24, 8:10 PM

good speed

Ddsienra Click to see attachment

Furkan Gözükara SECoursesOP•9/18/24, 8:10 PM

nope

Ddsienra false alarm sorry

Furkan Gözükara SECoursesOP•9/18/24, 8:11 PM

yep i was gonna tell this :d

TheJayHole•9/18/24, 8:39 PM

New supporter here. Wanted to say thanks. Your info is great. Couple of questions for you. I'm doing LoRA training on Flux Really, really slow on my 3090 (Windows with latest torch 2.5 speedup), 51 images in LoRA. This may have been answered somewhere, but previously with this SDXL LoRA I would do 32 repeats, 51 images, 10 epoch (16320 steps). This was a little high admittedly. The config you had would be 10200 steps Are epochs better than repeats with FLUX1.dev? Does this lower the number required? Just looking to understand that a little better. As for the slowness I'm 4735 steps and running on 1.5-2 days. I did update torch. Pretty much the same speed for me. Might have to try on RunPod. Any thoughts/advice are appreciated.

TTheJayHole New supporter here. Wanted to say thanks. Your info is great. Couple of quest...

Furkan Gözükara SECoursesOP•9/18/24, 8:53 PM

welcome

Furkan Gözükara SECoursesOP•9/18/24, 8:53 PM

what is your step speed?

Furkan Gözükara SECoursesOP•9/18/24, 8:53 PM

you can use Rank_4_17250MB_4_85_Second_IT.json it is pretty decent speed

Furkan Gözükara SECoursesOP•9/18/24, 8:53 PM

200 epoch may be a lot for 51 image

Furkan Gözükara SECoursesOP•9/18/24, 8:53 PM

but you can compare checkpoints

xlads•9/18/24, 9:25 PM

Is nsfw possible with flux lora?

Xxlads Is nsfw possible with flux lora?

Furkan Gözükara SECoursesOP•9/18/24, 9:29 PM

i dont know i really dont into nsfw

Furkan Gözükara SECoursesOP•9/18/24, 9:29 PM

we also avoid discussing it here

xlads•9/18/24, 9:36 PM

Okay! Asking for someone else

dxqb•9/18/24, 9:39 PM

I've just trained a very bad dataset on flux. Blurry, mixed resolution down to 128px. For SDXL, I have carefully prepared it using SUPIR and more. With flux, it just works without preparation. This model is insane

Ddxqb I've just trained a very bad dataset on flux. Blurry, mixed resolution down to 1...

Furkan Gözükara SECoursesOP•9/18/24, 10:00 PM

so true

Furkan Gözükara SECoursesOP•9/18/24, 10:00 PM

with flux you can only get better

Furkan Gözükara SECoursesOP•9/18/24, 10:00 PM

i plan to do general fine tuning with our 20k images but i need at least 2x A100 like 1 week

i think flux can do yes

Similar Threads

Similar Threads

Similar Threads