 Aarni Koskela
|
f8fb74b93a
Bump Spandrel to 0.3.4; add spandrel-extra-arches for CodeFormer
|
hai 1 ano |
|
Aarni Koskela
|
777af661a2
Be more clear about Spandrel model nomenclature
|
hai 1 ano |
|
Aarni Koskela
|
4ad0c0c0a8
Verify architecture for loaded Spandrel models
|
hai 1 ano |
|
Aarni Koskela
|
b621a63cf6
Unify CodeFormer and GFPGAN restoration backends, use Spandrel for GFPGAN
|
hai 1 ano |
|
Aarni Koskela
|
b0f5934234
Use Spandrel for upscaling and face restoration architectures (aside from GFPGAN and LDSR)
|
hai 1 ano |
 avantcontra
|
236dd55dbe
fix Blank line contains whitespace
|
hai 1 ano |
|
avantcontra
|
443ca983ad
fix bug when using --gfpgan-models-path
|
hai 1 ano |
 AUTOMATIC1111
|
3cd4fd51ef
Merge pull request #10823 from akx/model-loady
|
%!s(int64=2) %!d(string=hai) anos |
|
Aarni Koskela
|
0afbc0c235
Fix up `if "http" in ...:` to be more sensible startswiths
|
%!s(int64=2) %!d(string=hai) anos |
|
Aarni Koskela
|
165ab44f03
Use os.makedirs(..., exist_ok=True)
|
%!s(int64=2) %!d(string=hai) anos |
|
AUTOMATIC
|
05933840f0
rename print_error to report, use it with together with package name
|
%!s(int64=2) %!d(string=hai) anos |
|
Aarni Koskela
|
00dfe27f59
Add & use modules.errors.print_error where currently printing exception info by hand
|
%!s(int64=2) %!d(string=hai) anos |
|
AUTOMATIC
|
4b854806d9
F401 fixes for ruff
|
%!s(int64=2) %!d(string=hai) anos |
 Max Audron
|
5eee2ac398
add data-dir flag and set all user data directories based on it
|
%!s(int64=2) %!d(string=hai) anos |
 brkirch
|
f4a488f585
Set device for facelib/facexlib and gfpgan
|
%!s(int64=2) %!d(string=hai) anos |
|
brkirch
|
e9e2a7ec9a
Merge branch 'master' into cpu-cmdline-opt
|
%!s(int64=2) %!d(string=hai) anos |
|
AUTOMATIC
|
6c6ae28bf5
send all three of GFPGAN's and codeformer's models to CPU memory instead of just one for #1283
|
%!s(int64=2) %!d(string=hai) anos |
|
brkirch
|
b88e4ea7d6
Merge branch 'master' into master
|
%!s(int64=2) %!d(string=hai) anos |
|
AUTOMATIC
|
6491b09c24
use existing function for gfpgan
|
%!s(int64=2) %!d(string=hai) anos |
|
brkirch
|
bdaa36c844
When device is MPS, use CPU for GFPGAN instead
|
%!s(int64=2) %!d(string=hai) anos |
|
AUTOMATIC
|
d1f098540a
remove unwanted formatting/functionality from the PR
|
%!s(int64=2) %!d(string=hai) anos |
 d8ahazard
|
0dce0df1ee
Holy $hit.
|
%!s(int64=2) %!d(string=hai) anos |
|
d8ahazard
|
740070ea9c
Re-implement universal model loading
|
%!s(int64=2) %!d(string=hai) anos |
|
AUTOMATIC
|
d4205e66fa
gfpgan: just download the damn model
|
%!s(int64=2) %!d(string=hai) anos |
|
AUTOMATIC
|
843b2b64fc
Instance of CUDA out of memory on a low-res batch, even with --opt-split-attention-v1 (found cause) #255
|
%!s(int64=2) %!d(string=hai) anos |
|
AUTOMATIC
|
6a9b33c848
codeformer support
|
%!s(int64=3) %!d(string=hai) anos |
|
AUTOMATIC
|
595c827bd3
option to unload GFPGAN after using
|
%!s(int64=3) %!d(string=hai) anos |
|
AUTOMATIC
|
345028099d
split codebase into multiple files; to anyone this affects negatively: sorry
|
%!s(int64=3) %!d(string=hai) anos |
