Software Engineering Courses (SECourses)•3y ago

thanks. i tried a fresh install as well as a different commit, but i can't get my it/s back to where

thanks. i tried a fresh install as well as a different commit, but i can't get my it/s back to where it used to be. i have no clue how this happened. i haven't done any system updates or software installs recently, and this literally happened after doing the git pull

Furkan Gözükara SECourses•5/29/23, 9:10 PM

just came back

Ffelixsanz Click to see attachment

Furkan Gözükara SECourses•5/29/23, 9:11 PM

he probably did fine tuning

Furkan Gözükara SECourses•5/29/23, 9:11 PM

didnt train face or subject

Tthebean thanks. i tried a fresh install as well as a different commit, but i can't get m...

Furkan Gözükara SECourses•5/29/23, 9:11 PM

i can connect your pc and install if you are our patreon supporter

EEggo so this was it. i have new xformers now, and now my training in automatic 1111 ...

Furkan Gözükara SECourses•5/29/23, 9:11 PM

i can connect your pc and install if you are our patreon supporter

FFurkan Gözükara SECourses didnt train face or subject

felixsanz•5/29/23, 9:12 PM

but fine tuning is just training something else, isn't it?

felixsanz•5/29/23, 9:12 PM

dreambooth is fine-tuning

felixsanz•5/29/23, 9:12 PM

what is not lora, hypernetworks, textual inversion, etc..... is fine-tuning

Furkan Gözükara SECourses•5/29/23, 10:38 PM

by fine tuning i mean improving tokens

Furkan Gözükara SECourses•5/29/23, 10:38 PM

not teaching a subject

Furkan Gözükara SECourses•5/29/23, 10:39 PM

it turns out that this deepfake resolution is very low. i had to apply codeformer for fixing face

Furkan Gözükara SECourses•5/29/23, 10:39 PM

and it changes

Furkan Gözükara SECourses•5/29/23, 10:39 PM

however my face is very different than john wick so still decent

Ffelixsanz @Furkan Gözükara SECourses have you tried this? https://github.com/victorchall/E...

Haashim•5/29/23, 11:37 PM

You are looking to finetune a model?

Haashim•5/29/23, 11:37 PM

Do you know what everydream does vs Kohya ?

生

生死•5/30/23, 3:51 AM

if i have a own dataset and i want to train it into dolly 2.0 how should i do

Deleted User•5/30/23, 3:52 AM

Hey! does anyone know how i should go about making an ai generated video that flows smoothly? do i use controlnet1.1?

Sheng Antalpha•5/30/23, 4:25 AM

anyone know how to generate a photo with extremely detailed stripe and decoration like the photo above? I trained several model, they are good in looking, but far away from the detail level, anybody have a clue?

FFurkan Gözükara SECourses dev branch of automatic1111. they made a new release so do a git pull

Khang•5/30/23, 5:35 AM

Thank Dr.Furkan it works and I can compare different models now. But today I discover that XYZ plot still show the same images when I use different VAE or Clip Skip. The model can still make different images with different clip skip (attached images are clip skip 1 and 10) but XYZ is unable to do so.
I already do the git pull how should I fix this? (My model are trained by Kohya_ss)

SSheng Antalpha anyone know how to generate a photo with extremely detailed stripe and decoratio...

Khang•5/30/23, 5:39 AM

Have you tried the Detail Tweaker Lora on CivitAI yet? And some other upscale extensions of SD?

undelirium•5/30/23, 7:20 AM

@Furkan Gözükara SECourses as a follow up to information I offered about 2.X control nets last week, I found these today and thought maybe they'd be of interest to you as well https://huggingface.co/thibaud/controlnet-sd21/tree/main

thibaud/controlnet-sd21 at main

Cool And Normal•5/30/23, 8:43 AM

i made a thing

Cool And Normal•5/30/23, 8:45 AM

i think i needed to adjust a few things, but it was jsut to get a feel for the process, seem to be having problems with the python script to rename the files, so i didn't upscale it

Cool And Normal•5/30/23, 8:45 AM

which probably would have done wonders for the face

KKhang Have you tried the Detail Tweaker Lora on CivitAI yet? And some other upscale ex...

Sheng Antalpha•5/30/23, 8:48 AM

i tried upsclaing the effect is not obvious, i haven't try the Detail Tweaker Lora , do you have some insight?

Cool And Normal•5/30/23, 8:49 AM

i haven't tried to use Loras

Cool And Normal•5/30/23, 8:50 AM

but i get a feeling they would probably rein in styles a bit if used with this, maybe?

Cool And Normal•5/30/23, 8:51 AM

is restore faces a Lora?

Cool And Normal•5/30/23, 8:51 AM

but like just a built in one?

Cool And Normal•5/30/23, 8:52 AM

does anyone here use 2 gpus ?

Cool And Normal•5/30/23, 8:52 AM

or more i guess?

ashley•5/30/23, 9:05 AM

@Furkan Gözükara SECourses isn't Lion your preferred optimizer for Dreambooth training? They are removing it from the Dreambooth extension:

https://github.com/d8ahazard/sd_dreambooth_extension/pull/1244

GitHub

Replace some optimizers by ArrowM · Pull Request #1244 · d8ahazard/...

Optimizer changes:

remove Lion & AdanIP Adapt
replace Adan Adapt
Add SGD Adapt

A couple library updates
A couple linting changes

生生死 if i have a own dataset and i want to train it into dolly 2.0 how should i do

Furkan Gözükara SECourses•5/30/23, 9:15 AM

it is another story you should check their github

DDeleted User Hey! does anyone know how i should go about making an ai generated video that fl...

Furkan Gözükara SECourses•5/30/23, 9:16 AM

this is a very researched topic. i am also researching a good workflow

SSheng Antalpha anyone know how to generate a photo with extremely detailed stripe and decoratio...

Furkan Gözükara SECourses•5/30/23, 9:16 AM

this could be multiple times upscaled to add more details

KKhang Thank Dr.Furkan it works and I can compare different models now. But today I dis...

Furkan Gözükara SECourses•5/30/23, 9:17 AM

update auto1111 to latest version

Uundelirium @Furkan Gözükara SECourses as a follow up to information I offered about 2.X con...

Furkan Gözükara SECourses•5/30/23, 9:17 AM

nice thanks

Aashley @Furkan Gözükara SECourses isn't Lion your preferred optimizer for Dreambooth tr...

Furkan Gözükara SECourses•5/30/23, 9:19 AM

thanks i did a comment

FFurkan Gözükara SECourses thanks i did a comment

ashley•5/30/23, 9:20 AM

I also commented with the best settings per your video, he was probably using incorrect settings in his test.

Aashley I also commented with the best settings per your video, he was probably using in...

Furkan Gözükara SECourses•5/30/23, 9:21 AM

agreed

Furkan Gözükara SECourses•5/30/23, 9:22 AM

I need help with anyone that has reddit account. just posting into subreddit. need 2 people

FFurkan Gözükara SECourses agreed

ashley•5/30/23, 9:23 AM

I don't really know how Reddit works, must I just paste the YouTube link of the video to the subreddit link you sent me?

ashley•5/30/23, 9:23 AM

Apologies for replying to the incorrect message, I am still half asleep

Aashley I don't really know how Reddit works, must I just paste the YouTube link of the ...

Furkan Gözükara SECourses•5/30/23, 9:24 AM

so easy

Furkan Gözükara SECourses•5/30/23, 9:24 AM

like this

Furkan Gözükara SECourses•5/30/23, 9:24 AM

Furkan Gözükara SECourses•5/30/23, 9:24 AM

new post

ashley•5/30/23, 9:26 AM

Kallamamran•5/30/23, 9:39 AM

@Furkan Gözükara SECourses Would you mind me asking... I might have either missed it completely, missed it because of the flood of available info or I have not missed at all and there actually is no such video, BUT.... Finetuning... We have tutors for creating checkpoints/embeddings/LORA's from subjects/persons/styles, but where is the tutorials for creating complete finetuned checkpoints (Or embeddings/LORA's if possible). I mean like training them on thousands of images like some models are.
Most models on CivitAI seem to be just people mixing other models together and that feels more like a trial and error method that can turn out "how ever".

thanks. i tried a fresh install as well as a different commit, but i can't get my it/s back to where

Similar Threads