Skip to content

Add optional SyncBatchNorm support to ImageNet example #1406

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
snehasonkusare-tech wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Add optional SyncBatchNorm support to ImageNet example #1406

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 12 additions & 1 deletion imagenet/main.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -79,6 +79,12 @@
'fastest way to use PyTorch for either single node or '
'multi node data parallel training')
parser.add_argument('--dummy', action='store_true', help="use fake data to benchmark")
parser.add_argument(
'--sync-bn',
action='store_true',
help='Convert BatchNorm layers to SyncBatchNorm for multi-GPU training'
)


best_acc1 = 0

Expand Down Expand Up @@ -160,10 +166,15 @@ def main_worker(gpu, ngpus_per_node, args):
if args.pretrained:
print("=> using pre-trained model '{}'".format(args.arch))
model = models.__dict__[args.arch](pretrained=True)

else:
print("=> creating model '{}'".format(args.arch))
model = models.__dict__[args.arch]()


# Convert BN → SyncBatchNorm if requested AND distributed training is enabled
if args.distributed and args.sync_bn:
print("=> Converting BatchNorm layers to SyncBatchNorm")
model = nn.SyncBatchNorm.convert_sync_batchnorm(model)
if not use_accel:
print('using CPU, this will be slow')
elif args.distributed:
Expand Down