HongChuang/ultralytics-ascend
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 3

BaseTrainer with find_unused_parameters=True when using DistributedDataParallel() (#14323)

Browse source
This commit is contained in:
Chia-Hsiang Tsai 2024-07-11 01:59:16 +08:00 • committed by GitHub
parent 997f2c92cd
commit 470b120a1b
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
ultralytics/engine/trainer.py
View file

@ -266,7 +266,7 @@ class BaseTrainer:
self.amp = bool(self.amp) # as boolean self.amp = bool(self.amp) # as boolean
self.scaler = torch.cuda.amp.GradScaler(enabled=self.amp) self.scaler = torch.cuda.amp.GradScaler(enabled=self.amp)
if world_size > 1: if world_size > 1:
self.model = nn.parallel.DistributedDataParallel(self.model, device_ids=[RANK]) self.model = nn.parallel.DistributedDataParallel(self.model, device_ids=[RANK], find_unused_parameters=True)
# Check imgsz # Check imgsz
gs = max(int(self.model.stride.max() if hasattr(self.model, "stride") else 32), 32) # grid size (max stride) gs = max(int(self.model.stride.max() if hasattr(self.model, "stride") else 32), 32) # grid size (max stride)