Skip to content

Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.

find_unused_parameters

游雁0bf8edca37d3
diff --git a/funasr/bin/train_ds.py b/funasr/bin/train_ds.py
index 69c2712fe..5b1eeaa9c 100644
--- a/funasr/bin/train_ds.py
+++ b/funasr/bin/train_ds.py
@@ -84,9 +84,6 @@ def main(**kwargs):
dist.init_process_group(
backend=kwargs.get("backend", "nccl"),
init_method="env://",
- find_unused_parameters=kwargs.get("train_conf", {}).get(
- "find_unused_parameters", False
- ),
)
torch.cuda.set_device(local_rank)