Software Engineering Courses (SECourses)•10mo ago

@Dr. Furkan Gözükara with the help of swarmUi i will be able to make the consitency in the face with

@Dr. Furkan Gözükara with the help of swarmUi i will be able to make the consitency in the face with even flux as base model for the generation on my lora <segment.yolo-face_yolov9c.pt+10,0.7,0.5> with this text in the prompt

blackhatOP•3/6/25, 3:22 PM

do flux support this weight adjustment within promt ?

Bblackhat @Dr. Furkan Gözükara with the help of swarmUi i will be able to make the consite...

Furkan Gözükara SECourses•3/6/25, 3:23 PM

this means targeting 10th face

blackhatOP•3/6/25, 3:24 PM

actually i was altering and tesint it was -1

Furkan Gözükara SECourses•3/6/25, 3:24 PM

yes lora working

Bblackhat @Dr. Furkan Gözükara with the help of swarmUi i will be able to make the consite...

Furkan Gözükara SECourses•3/6/25, 3:24 PM

it wont make consistent face though

Furkan Gözükara SECourses•3/6/25, 3:24 PM

you need to train :d

FFurkan Gözükara SECourses https://github.com/huggingface/hf_transfer/issues/63

xlads•3/6/25, 3:25 PM

i got this error "pip install git+https://github.com/huggingface/hf_transfer@better_error_windows#egg=hf_transfer
Collecting hf_transfer
Cloning https://github.com/huggingface/hf_transfer (to revision better_error_windows) to c:\users\dl335\appdata\local\temp\pip-install-mobceim8\hf-transfer_2236cf29a2ae4c7899a2bbec52b452e9
Running command git clone --filter=blob:none --quiet https://github.com/huggingface/hf_transfer 'C:\Users\DL335\AppData\Local\Temp\pip-install-mobceim8\hf-transfer_2236cf29a2ae4c7899a2bbec52b452e9'
Running command git checkout -b better_error_windows --track origin/better_error_windows
branch 'better_error_windows' set up to track 'origin/better_error_windows'.
Switched to a new branch 'better_error_windows'
Resolved https://github.com/huggingface/hf_transfer to commit d330fc97c13e7b64e1b0bffb93820aa89ce72740
Installing build dependencies ... done
Getting requirements to build wheel ... done
Preparing metadata (pyproject.toml) ... error
error: subprocess-exited-with-error

× Preparing metadata (pyproject.toml) did not run successfully.
│ exit code: 1
╰─> [6 lines of output]

Cargo, the Rust package manager, is not installed or is not on PATH.
This package requires Rust and Cargo to compile extensions. Install it through
the system's package manager or via https://rustup.rs/

Checking for Rust toolchain....
[end of output]

note: This error originates from a subprocess, and is likely not a problem with pip.
error: metadata-generation-failed

× Encountered error while generating package metadata.
╰─> See above for output.

note: This is an issue with the package mentioned above, not pip.
hint: See above for details.

[notice] A new release of pip is available: 23.0.1 -> 25.0.1
[notice] To update, run: python.exe -m pip install --upgrade pip"

xlads•3/6/25, 3:25 PM

DO i just download rust into my c drive?

Xxlads i got this error "pip install git+http...

Furkan Gözükara SECourses•3/6/25, 3:26 PM

can you first run this

Furkan Gözükara SECourses•3/6/25, 3:26 PM

python.exe -m pip install --upgrade pip

Furkan Gözükara SECourses•3/6/25, 3:26 PM

then run that

Furkan Gözükara SECourses•3/6/25, 3:26 PM

wan 2.1 lora working i just verifieed

FFurkan Gözükara SECourses you need to train :d

blackhatOP•3/6/25, 3:26 PM

yes i trained it , i was going through your generations and the pomot and in that i found this text in many generations what this thing do ? if it doent manage consistency

blackhatOP•3/6/25, 3:26 PM

@Dr. Furkan Gözükara

FFurkan Gözükara SECourses can you first run this

xlads•3/6/25, 3:28 PM

i ran this then tried re-installing but still getting same rust error

Bblackhat yes i trained it , i was going through your generations and the pomot and in tha...

Furkan Gözükara SECourses•3/6/25, 3:29 PM

you must have something wrong in that case

Furkan Gözükara SECourses•3/6/25, 3:29 PM

during training

Furkan Gözükara SECourses•3/6/25, 3:29 PM

you are using your trained model right?

blackhatOP•3/6/25, 3:30 PM

yes

Xxlads i ran this then tried re-installing but still getting same rust error

Furkan Gözükara SECourses•3/6/25, 3:30 PM

ye i get too

Furkan Gözükara SECourses•3/6/25, 3:30 PM

let me report

Furkan Gözükara SECourses•3/6/25, 3:30 PM

so you need to edit bat file and remove hf_transfer line

FFurkan Gözükara SECourses you are using your trained model right?

blackhatOP•3/6/25, 3:32 PM

and its working well but i wanted to understand the use of the segment yolo part of your prompt to understand its usage in prompt
what is the use of it ?

FFurkan Gözükara SECourses so you need to edit bat file and remove hf_transfer line

xlads•3/6/25, 3:33 PM

okay i just reported in the github link you sent also

FFurkan Gözükara SECourses so you need to edit bat file and remove hf_transfer line

xlads•3/6/25, 3:37 PM

seems to be working

Bblackhat and its working well but i wanted to understand the use of the segment yolo part...

Furkan Gözükara SECourses•3/6/25, 3:42 PM

segment = manually masking and inpainting with prompt

Furkan Gözükara SECourses•3/6/25, 3:42 PM

that we are automating it

Xxlads seems to be working

Furkan Gözükara SECourses•3/6/25, 3:42 PM

great

Xxlads okay i just reported in the github link you sent also

Furkan Gözükara SECourses•3/6/25, 3:42 PM

thanks. i hope this can get fixed

xlads•3/6/25, 4:37 PM

is it possible to train with just one image and have consistency in character?

Xxlads is it possible to train with just one image and have consistency in character?

Furkan Gözükara SECourses•3/6/25, 5:59 PM

2 step. so train generate and train again

Furkan Gözükara SECourses•3/6/25, 5:59 PM

because single image usage not enough

xlads•3/6/25, 7:20 PM

I’m geting an error while downloading Training_model_files.The error message says: 'Could not set the permissions on the file... Error: [Errno 13] Permission denied.' The download continues, but I’m not sure if this will cause issues later.

julius•3/6/25, 7:26 PM

@Dr. Furkan Gözükara or anyone else have tried fine-tuning flux on a smartphone photo style? So that when afterwards doing a new fine-tuning on a person, the photos will look more casual/realistic?

I'm trying right now with 170 mixed photos, most from different Instagram profiles, maybe 2/3 of them includes people (not sure if that's smart or not) and otherwise just using Furkans default 48GB Batch size 7 config.

No idea how many steps/epoch is needed, or what the optimal learning rate is.

Using the word "style" as the Class Prompt, also not sure if that's smart.