Clear model from memory before switch, fixes out of memory
This commit is contained in:
parent
ea422ca68a
commit
0c4de4776e
@ -1,4 +1,5 @@
|
|||||||
import torch
|
import torch
|
||||||
|
import gc
|
||||||
|
|
||||||
from lama_cleaner.model.fcf import FcF
|
from lama_cleaner.model.fcf import FcF
|
||||||
from lama_cleaner.model.lama import LaMa
|
from lama_cleaner.model.lama import LaMa
|
||||||
@ -42,6 +43,12 @@ class ModelManager:
|
|||||||
if new_name == self.name:
|
if new_name == self.name:
|
||||||
return
|
return
|
||||||
try:
|
try:
|
||||||
|
if (torch.cuda.memory_allocated() > 0):
|
||||||
|
# Clear current loaded model from memory
|
||||||
|
torch.cuda.empty_cache()
|
||||||
|
gc.collect()
|
||||||
|
del self.model
|
||||||
|
|
||||||
self.model = self.init_model(new_name, self.device, **self.kwargs)
|
self.model = self.init_model(new_name, self.device, **self.kwargs)
|
||||||
self.name = new_name
|
self.name = new_name
|
||||||
except NotImplementedError as e:
|
except NotImplementedError as e:
|
||||||
|
Loading…
Reference in New Issue
Block a user