VictorMylle/Kohya-ss-sd-scripts
1
0
Fork 1
mirror of https://github.com/kohya-ss/sd-scripts.git synced 2026-04-09 06:45:09 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix: distributed training in windows

Browse Source
This commit is contained in:
Akegarasu
2024-10-10 14:03:59 +08:00
parent 83e3048cb0
commit 3de42b6edb
1 changed files with 12 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

21
library/train_util.py
View File

@@ -5045,17 +5045,18 @@ def prepare_accelerator(args: argparse.Namespace):
if args.torch_compile:
dynamo_backend = args.dynamo_backend
kwargs_handlers = (
InitProcessGroupKwargs(timeout=datetime.timedelta(minutes=args.ddp_timeout)) if args.ddp_timeout else None,
(
kwargs_handlers = [
InitProcessGroupKwargs(
backend = "gloo" if os.name == "nt" or not torch.cuda.is_available() else "nccl",
init_method="env://?use_libuv=False" if os.name == "nt" else None,
timeout=datetime.timedelta(minutes=args.ddp_timeout) if args.ddp_timeout else None
) if torch.cuda.device_count() > 1 else None,
DistributedDataParallelKwargs(
gradient_as_bucket_view=args.ddp_gradient_as_bucket_view, static_graph=args.ddp_static_graph
)
if args.ddp_gradient_as_bucket_view or args.ddp_static_graph
else None
),
)
kwargs_handlers = list(filter(lambda x: x is not None, kwargs_handlers))
gradient_as_bucket_view=args.ddp_gradient_as_bucket_view,
static_graph=args.ddp_static_graph
) if args.ddp_gradient_as_bucket_view or args.ddp_static_graph else None
]
kwargs_handlers = [i for i in kwargs_handlers if i is not None]
deepspeed_plugin = deepspeed_utils.prepare_deepspeed_plugin(args)
accelerator = Accelerator(