Software Engineering Courses (SECourses)•3y ago

https://github.com/AUTOMATIC1111/stable-diffusion-webui/releases/tag/v1.7.0

GitHub

Release 1.7.0 · AUTOMATIC1111/stable-diffusion-webui

Features:

settings tab rework: add search field, add categories, split UI settings page into many
add altdiffusion-m18 support (#13364)
support inference with LyCORIS GLora networks (#13610)
add l...

JJMPA You mean I should use this instead of SDXL base 1.0 on Kohya training ?

Xylber•12/16/23, 6:11 PM

Isn't that the new president of Argentina?

krigeta•12/16/23, 7:27 PM

I have a lora for lineart style, an image(simple lineart + colored lineart) both then how can we change the style of the little lineart only? Using controlnet or ipadapters? Or change the style of the plane colored image by some anime style lora?

Oomv421 https://github.com/AUTOMATIC1111/stable-diffusion-webui/releases/tag/v1.7.0

mikemenders•12/16/23, 7:49 PM

Someone explain this to me in a simple, understandable way, what is it for in 1.7.0 version?
https://github.com/AUTOMATIC1111/stable-diffusion-webui/pull/13568

GitHub

Add lora-embedding bundle system by KohakuBlueleaf · Pull Request #...

Description
Some Trainer (like hcp-diffusion, naifu) have implemented a method of training called "pivtoal tuning", which is basically trained Embedding and LoRA(DreamBooth) at the same t...

Oomv421 https://github.com/AUTOMATIC1111/stable-diffusion-webui/releases/tag/v1.7.0

Pelolisu•12/16/23, 7:56 PM

the funny thing is that in the dev branch there are still 132 commits that are not in this release, a lot of things are still to come and be adjusted.

Mmikemenders Someone explain this to me in a simple, understandable way, what is it for in 1....

Fire 🔥•12/16/23, 10:32 PM

It say right there in the issue:

We can now train Embeddings (aka Textual Inversion) and LoRA at the same time
However, multi-concept models produce many embedding files, one for each concept.
This adds a new feature to read "bundle" files that bundle many embeddings together into one

Kkrigeta Hello @Dr. Furkan Gözükara , may you help me with the comfyUI workflow? I want t...

Furkan Gözükara SECourses•12/16/23, 10:50 PM

sadly i am not that expert

JJMPA After 20 hours of training,… it says this. Please help

Furkan Gözükara SECourses•12/16/23, 10:50 PM

your space full

Furkan Gözükara SECourses•12/16/23, 10:50 PM

you can download existing checkpoints

Furkan Gözükara SECourses•12/16/23, 10:51 PM

but it wont continue

Furkan Gözükara SECourses•12/16/23, 10:51 PM

you need to calculate number of checkpoints

DDeleted User Is this is sd 1.5

Furkan Gözükara SECourses•12/16/23, 10:51 PM

yes

JJMPA You mean I should use this instead of SDXL base 1.0 on Kohya training ?

Furkan Gözükara SECourses•12/16/23, 10:55 PM

i see this was on runpod

Furkan Gözükara SECourses•12/16/23, 10:55 PM

you can delete older models

Furkan Gözükara SECourses•12/16/23, 10:55 PM

or turn off pod

Furkan Gözükara SECourses•12/16/23, 10:55 PM

expand storafge

Furkan Gözükara SECourses•12/16/23, 10:55 PM

and use last checkpoint as a base model

Furkan Gözükara SECourses•12/16/23, 10:55 PM

and continue

Furkan Gözükara SECourses•12/16/23, 10:56 PM

for delete use rm -r file_path

Iiamannakogan I have a question. For some people, the results of SDXL Dreamboots look OK in ge...

Furkan Gözükara SECourses•12/16/23, 10:56 PM

are you using adetailer to improve?

Mmagical Hello my kaggle account got blocked can i create a new account using the same ph...

Furkan Gözükara SECourses•12/16/23, 10:57 PM

no you need a new number

Mmikemenders Someone explain this to me in a simple, understandable way, what is it for in 1....

Furkan Gözükara SECourses•12/16/23, 10:59 PM

looks like just another lora method

Kallamamran•12/16/23, 11:22 PM

So.... Anyone else stuck in Check for Updates in A1111

Kallamamran•12/16/23, 11:22 PM

And counting

Kallamamran•12/16/23, 11:27 PM

KKallamamran So.... Anyone else stuck in Check for Updates in A1111

Furkan Gözükara SECourses•12/16/23, 11:29 PM

newest update could have bug

Furkan Gözükara SECourses•12/16/23, 11:31 PM

UDEMY banned my course and it is too much bothersome. i wonder if anyone among us could help me?

Kallamamran•12/16/23, 11:42 PM

Well... I removed some extensions I don't use anyway and now it works

Kallamamran•12/16/23, 11:42 PM

Banned!?!?! Fu**ing why?

Kallamamran•12/16/23, 11:53 PM

now I remember why I used Roop and not Reactor

JMPA•12/17/23, 12:06 AM

Does anyone know why the photos generated when the face is close up are perfect, and when it is shown further away, like with the whole body, it looks bad?

KKallamamran Banned!?!?! Fu**ing why?

Furkan Gözükara SECourses•12/17/23, 12:09 AM

because of a tweet .

JJMPA Does anyone know why the photos generated when the face is close up are perfect,...

Furkan Gözükara SECourses•12/17/23, 12:09 AM

yes

Furkan Gözükara SECourses•12/17/23, 12:09 AM

because of the resolution

Furkan Gözükara SECourses•12/17/23, 12:09 AM

for distant shots resolution of models being low

JMPA•12/17/23, 12:09 AM

How can I solve it?

JJMPA How can I solve it?

Furkan Gözükara SECourses•12/17/23, 12:11 AM

you can adetailer extension face

Furkan Gözükara SECourses•12/17/23, 12:11 AM

it inpaints

JMPA•12/17/23, 12:12 AM

I put it but not sure how config it do you have a tutorial?

JJMPA I put it but not sure how config it do you have a tutorial?

Furkan Gözükara SECourses•12/17/23, 12:13 AM

yes i show in my tutorials

Furkan Gözükara SECourses•12/17/23, 12:13 AM

did you do a specific training

Furkan Gözükara SECourses•12/17/23, 12:13 AM

like ohwx man?

Furkan Gözükara SECourses•12/17/23, 12:13 AM

what is your prompt

JMPA•12/17/23, 12:18 AM

"ohwx man 1.1" as superman with 50 years old, no marked muscles, on Argentina Obelisc street, cinematic 4k epic detailed 4k epic detailed photograph shot on kodak detailed bokeh cinematic hbo dark moody,fat belly, 80kg body, aggressive screaming face, natural real hair, without big muscles, very detailed face skin real, real light, face a little fat without perfect markings,

JMPA•12/17/23, 12:18 AM

Negative: muscular, (semi-realistic, cgi, 3d, render, sketch, cartoon, drawing, anime:1.4), fat, text, cropped, out of frame, worst quality, low quality, jpeg artifacts, ugly, duplicate, morbid, mutilated, extra fingers, mutated hands, poorly drawn hands, poorly drawn face, mutation, deformed, blurry, dehydrated, bad anatomy, bad proportions, extra limbs, cloned face, disfigured, gross proportions, malformed limbs, missing arms, missing legs, extra arms, extra legs, fused fingers, too many fingers, long neck, cap, freckles, pilot clothes, imperfect letters, perfect skin, perfect body, perfect hair , young, muscular, muscles

JMPA•12/17/23, 12:18 AM

——

JMPA•12/17/23, 12:18 AM

i use DPM++ 2S a Karras, DPM++ 3M SDE Exponential, DPM++ SDE Karras

JMPA•12/17/23, 12:25 AM

If you see this is perfect

JMPA•12/17/23, 12:26 AM

And this is not the face person….

JJMPA "ohwx man 1.1" as superman with 50 years old, no marked muscles, on Argentina O...

Furkan Gözükara SECourses•12/17/23, 12:32 AM

then in adetailer prompt

https://github.com/AUTOMATIC1111/stable-diffusion-webui/releases/tag/v1.7.0

Similar Threads

Similar Threads

Similar Threads