Mirrored_Repos/InvokeAI
1
0
Fork 0
mirror of https://github.com/invoke-ai/InvokeAI synced 2024-08-30 20:32:17 +00:00
Code Issues Packages Projects Releases Wiki Activity

Remove explicit gc.collect() after transferring models from device to CPU. I'm not sure why this was there in the first place, but it was taking a significant amount of time (up to ~1sec in my tests).

Browse Source
This commit is contained in:
Ryan Dick 2023-11-03 11:08:09 -04:00 committed by Kent Keirsey
parent 5b420653f9
commit 43b300498f
1 changed files with 0 additions and 1 deletions

1
invokeai/backend/model_management/model_cache.py

@ -507,7 +507,6 @@ class ModelCache(object):
vram_in_use = torch.cuda.memory_allocated()
self.logger.debug(f"{(vram_in_use/GIG):.2f}GB VRAM used for models; max allowed={(reserved/GIG):.2f}GB")
gc.collect()
torch.cuda.empty_cache()
if choose_torch_device() == torch.device("mps"):
mps.empty_cache()