Software Engineering Courses (SECourses)•2y ago

SUPIR seems to be amazing and the 1-click installer too! Great job! -Is SDXL model with Juggernaut X

SUPIR seems to be amazing and the 1-click installer too! Great job! -Is SDXL model with Juggernaut XL - V9 only used to generate the initial image that is then upscaled, so I do not need SDXL if I use other images to upscale ?

Rrafraf what did you name this metadata category/type? its specific to your Supir script...

Furkan Gözükara SECourses•3/2/24, 1:43 PM

it should work with all images that uses same kind of metadata

VVegan Apps SUPIR seems to be amazing and the 1-click installer too! Great job! -Is SDXL mod...

Furkan Gözükara SECourses•3/2/24, 1:43 PM

no it is used during upscale

Rrafraf what did you name this metadata category/type? its specific to your Supir script...

Furkan Gözükara SECourses•3/2/24, 1:43 PM

do you need other images metadata too? i may try to make it more robust

Furkan Gözükara SECourses•3/2/24, 1:44 PM

i am testing output folders parameter right now

rafraf•3/2/24, 1:44 PM

so, I use an called Allusion which sees the metadata from Auto1111 and Forge

rafraf•3/2/24, 1:44 PM

here's an example

rafraf•3/2/24, 1:44 PM

rafraf•3/2/24, 1:45 PM

Furkan Gözükara SECourses•3/2/24, 1:45 PM

ok i will try to make it more robust

rafraf•3/2/24, 1:45 PM

you get all the data from the generation

rafraf•3/2/24, 1:45 PM

i think this metadata that auto1111 saves is called Parameters

rafraf•3/2/24, 1:45 PM

So if you use the same, i think it will be universal

Furkan Gözükara SECourses•3/2/24, 1:57 PM

fixing errors

protector131090•3/2/24, 2:04 PM

trying to use Supor and at 2nd stage getting:

[Tiled VAE]: split to 4x2 = 8 tiles. Optimal tile size 480x448, original tile size 512x512
[Tiled VAE]: Executing Encoder Task Queue: 100%|████████████████████████████████████| 728/728 [00:01<00:00, 405.80it/s]
[Tiled VAE]: Done in 2.045s, max VRAM alloc 9757.875 MB
P:\SUPIR_v7\SUPIR\venv\lib\site-packages\torch\nn\functional.py:5476: UserWarning: 1Torch was not compiled with flash attention. (Triggered internally at ..\aten\src\ATen\native\transformers\cuda\sdp_utils.cpp:263.)
attn_output = scaled_dot_product_attention(q, k, v, attn_mask, dropout_p, is_causal)

Furkan Gözükara SECourses•3/2/24, 2:05 PM