Kevin Turner 8a0d45ac5a new OffloadingDevice loads one model at a time, on demand (#2596) ... * new OffloadingDevice loads one model at a time, on demand * fixup! new OffloadingDevice loads one model at a time, on demand * fix(prompt_to_embeddings): call the text encoder directly instead of its forward method allowing any associated hooks to run with it. * more attempts to get things on the right device from the offloader * more attempts to get things on the right device from the offloader * make offloading methods an explicit part of the pipeline interface * inlining some calls where device is only used once * ensure model group is ready after pipeline.to is called * fixup! Strategize slicing based on free [V]RAM (#2572) * doc(offloading): docstrings for offloading.ModelGroup * doc(offloading): docstrings for offloading-related pipeline methods * refactor(offloading): s/SimpleModelGroup/FullyLoadedModelGroup * refactor(offloading): s/HotSeatModelGroup/LazilyLoadedModelGroup to frame it is the same terms as "FullyLoadedModelGroup" --------- Co-authored-by: Damian Stewart <null@damianstewart.com>		2023-02-16 23:48:27 +00:00
..
ckpt_generator	further improvements to ability to find location of data files	2023-01-31 12:24:55 -05:00
config	adding support for ESRGAN denoising strength, which allows for improved detail retention when upscaling photorelistic faces	2023-02-08 22:36:35 -06:00
generator	new OffloadingDevice loads one model at a time, on demand (#2596)	2023-02-16 23:48:27 +00:00
restoration	Change denoise_str to an arg instead of a class variable	2023-02-09 20:16:23 +13:00
training	ask user to make window larger if not enough space for textual inversion/merge gui	2023-02-12 17:38:46 -05:00
__init__.py	use 🧨diffusers model (#1583)	2023-01-15 09:22:46 -05:00
_version.py	bump version number; update contributors	2023-02-09 13:01:08 -05:00
args.py	Change denoise_str default to 0.75	2023-02-09 20:21:09 +13:00
ckpt_to_diffuser.py	remove debugging statement	2023-02-06 09:45:21 -05:00
CLI.py	pulling esrgan denoise strength through to the generate API.	2023-02-12 02:47:37 +13:00
concepts_lib.py	improve embed trigger token not found error	2023-01-19 15:46:58 -05:00
conditioning.py	Beautify Token Log Outputs	2023-02-05 22:55:29 +13:00
devices.py	Various fixes	2023-01-30 18:42:17 -05:00
globals.py	a few adjustments	2023-02-12 17:20:13 -05:00
image_util.py	Global replace [ \t]+$, add "GB" (#1751)	2022-12-19 16:36:39 +00:00
log.py	rename all modules from ldm.dream to ldm.invoke	2022-10-08 11:37:23 -04:00
merge_diffusers.py	ask user to make window larger if not enough space for textual inversion/merge gui	2023-02-12 17:38:46 -05:00
model_manager.py	new OffloadingDevice loads one model at a time, on demand (#2596)	2023-02-16 23:48:27 +00:00
offloading.py	new OffloadingDevice loads one model at a time, on demand (#2596)	2023-02-16 23:48:27 +00:00
patchmatch.py	defer patchmatch loading (#2039)	2022-12-20 15:32:35 -08:00
pngwriter.py	correct bug when trying to enhance JPG images (#1928)	2022-12-11 13:48:47 -05:00
prompt_parser.py	Merge branch 'main' into dev/installer	2023-02-01 17:50:22 -05:00
readline.py	use 🧨diffusers model (#1583)	2023-01-15 09:22:46 -05:00
seamless.py	use 🧨diffusers model (#1583)	2023-01-15 09:22:46 -05:00
server_legacy.py	Global replace [ \t]+$, add "GB" (#1751)	2022-12-19 16:36:39 +00:00
server.py	Global replace [ \t]+$, add "GB" (#1751)	2022-12-19 16:36:39 +00:00
txt2mask.py	remove dead code	2023-02-03 17:35:35 -05:00