InvokeAI/invokeai/backend
gogurtenjoyer 233869b56a Mac MPS FP16 fixes
This PR is to allow FP16 precision to work on Macs with MPS. In addition, it centralizes the torch fixes/workarounds
required for MPS into a new backend utility file `mps_fixes.py`. This is conditionally imported in `api_app.py`/`cli_app.py`.

Many MANY thanks to StAlKeR7779 for patiently working to debug and fix these issues.
2023-07-04 18:10:53 -04:00
..
generator Mac MPS FP16 fixes 2023-07-04 18:10:53 -04:00
image_util fix potential race condition in config system 2023-05-25 20:41:26 -04:00
install add the import model router 2023-07-04 14:35:47 +10:00
model_management add the import model router 2023-07-04 14:35:47 +10:00
restoration add migration script and update convert and face restoration paths 2023-06-13 01:27:51 -04:00
stable_diffusion Mac MPS FP16 fixes 2023-07-04 18:10:53 -04:00
training documentation tweaks; fixed initialization in a couple more places 2023-05-25 21:10:00 -04:00
util Mac MPS FP16 fixes 2023-07-04 18:10:53 -04:00
web Add dpmpp_sde and dpmpp_2m_sde schedulers(with karras) 2023-06-18 23:38:15 +03:00
__init__.py Remove more old logic 2023-06-19 15:57:28 +10:00
safety_checker.py add migration script and update convert and face restoration paths 2023-06-13 01:27:51 -04:00