Skip to content

Distributed mode (#55) #56

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

Distributed mode (#55) #56

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
16 changes: 15 additions & 1 deletion resemble_enhance/enhancer/__main__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,11 +4,12 @@
from pathlib import Path

import torch
import torch.distributed as dist
import torchaudio
from tqdm import tqdm

from .inference import denoise, enhance

from ..utils.distributed import local_rank, fix_unset_envs

@torch.inference_mode()
def main():
Expand Down Expand Up @@ -68,6 +69,11 @@ def main():
action="store_true",
help="Shuffle the audio paths and skip the existing ones, enabling multiple jobs to run in parallel",
)
parser.add_argument(
"--distributed_mode",
action="store_true",
help="Enable distributed training across multiple GPUs",
)

args = parser.parse_args()

Expand All @@ -86,6 +92,14 @@ def main():
if args.parallel_mode:
random.shuffle(paths)

if args.distributed_mode:
fix_unset_envs()
dist.init_process_group(backend='nccl' if device == "cuda" else "gloo")
torch.cuda.set_device(local_rank())
num_processed = dist.get_world_size()
rank = dist.get_rank()
paths = paths[rank::num_processed]

if len(paths) == 0:
print(f"No {args.suffix} files found in the following path: {args.in_dir}")
return
Expand Down