diff options
| author | brkirch <brkirch@users.noreply.github.com> | 2023-01-25 00:23:10 -0500 |
|---|---|---|
| committer | brkirch <brkirch@users.noreply.github.com> | 2023-01-25 01:13:04 -0500 |
| commit | e3b53fd295aca784253dfc8668ec87b537a72f43 (patch) | |
| tree | 6fb26afd730c0561a2506ead2d2c8295d326de40 /extensions-builtin/Lora/scripts/lora_script.py | |
| parent | 84d9ce30cb427759547bc7876ed80ab91787d175 (diff) | |
Add UI setting for upcasting attention to float32
Adds "Upcast cross attention layer to float32" option in Stable Diffusion settings. This allows for generating images using SD 2.1 models without --no-half or xFormers.
In order to make upcasting cross attention layer optimizations possible it is necessary to indent several sections of code in sd_hijack_optimizations.py so that a context manager can be used to disable autocast. Also, even though Stable Diffusion (and Diffusers) only upcast q and k, unfortunately my findings were that most of the cross attention layer optimizations could not function unless v is upcast also.
Diffstat (limited to 'extensions-builtin/Lora/scripts/lora_script.py')
0 files changed, 0 insertions, 0 deletions