AUTOMATIC
|
0ab0a50f9a
change formatting to match the main program in devices.py
|
2 年之前 |
源文雨
|
1130d5df66
Update devices.py
|
2 年之前 |
源文雨
|
76ab31e188
Fix wrong mps selection below MasOS 12.3
|
2 年之前 |
pepe10-gpu
|
62e9fec3df
actual better fix
|
2 年之前 |
pepe10-gpu
|
29eff4a194
terrible hack
|
2 年之前 |
pepe10-gpu
|
cd6c55c1ab
16xx card fix
|
2 年之前 |
brkirch
|
faed465a0b
MPS Upscalers Fix
|
2 年之前 |
brkirch
|
4c24347e45
Remove BSRGAN from --use-cpu, add SwinIR
|
2 年之前 |
AUTOMATIC
|
50b5504401
remove parsing command line from devices.py
|
2 年之前 |
Extraltodeus
|
57eb54b838
implement CUDA device selection by ID
|
2 年之前 |
brkirch
|
fdef8253a4
Add 'interrogate' and 'all' choices to --use-cpu
|
2 年之前 |
AUTOMATIC
|
7349088d32
--no-half-vae
|
2 年之前 |
brkirch
|
e9e2a7ec9a
Merge branch 'master' into cpu-cmdline-opt
|
2 年之前 |
AUTOMATIC
|
6c6ae28bf5
send all three of GFPGAN's and codeformer's models to CPU memory instead of just one for #1283
|
2 年之前 |
brkirch
|
27ddc24fde
Add BSRGAN to --add-cpu
|
2 年之前 |
brkirch
|
eeab7aedf5
Add --use-cpu command line option
|
2 年之前 |
brkirch
|
b88e4ea7d6
Merge branch 'master' into master
|
2 年之前 |
AUTOMATIC
|
820f1dc96b
initial support for training textual inversion
|
2 年之前 |
brkirch
|
bdaa36c844
When device is MPS, use CPU for GFPGAN instead
|
2 年之前 |
AUTOMATIC
|
9d40212485
first attempt to produce crrect seeds in batch
|
2 年之前 |
AUTOMATIC
|
c7e0e28ccd
changes for #294
|
2 年之前 |
AUTOMATIC
|
b70b51cc72
Allow TF32 in CUDA for increased performance #279
|
2 年之前 |
AUTOMATIC
|
8fb9c57ed6
add half() supporrt for CLIP interrogation
|
2 年之前 |
AUTOMATIC
|
f194457229
CLIP interrogator
|
2 年之前 |
Abdullah Barhoum
|
b5d1af11b7
Modular device management
|
2 年之前 |