zhengruifeng commented on code in PR #40695: URL: https://github.com/apache/spark/pull/40695#discussion_r1162185540
########## python/pyspark/ml/torch/distributor.py: ########## @@ -548,12 +560,23 @@ def set_torch_config(context: "BarrierTaskContext") -> None: os.environ["NODE_RANK"] = str(context.partitionId()) os.environ["RANK"] = str(context.partitionId()) - def set_gpus(context: "BarrierTaskContext") -> None: - if CUDA_VISIBLE_DEVICES in os.environ: - return + if local_master: + # distributed training on a local mode spark cluster + def set_gpus(context: "BarrierTaskContext") -> None: + if CUDA_VISIBLE_DEVICES in os.environ: + return Review Comment: I am not sure about this, here follows the distributed mode which respects the `CUDA_VISIBLE_DEVICES` env -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org