#!/bin/bash #SBATCH -p wzhdexclu03 #SBATCH -N 1 #SBATCH --cpus-per-task=1 #SBATCH --ntasks-per-node=32 #SBATCH --gres=dcu:4 #SBATCH -J onenode_4dcu #SBATCH -o logs/pt-%j.out #SBATCH -e logs/pt-%j.err source ~/miniconda3/etc/profile.d/conda.sh conda activate torch1.10-dtk22.10-py38 #conda activate base module purge module load compiler/devtoolset/7.3.1 mpi/hpcx/gcc-7.3.1 compiler/dtk/23.04 module list export HIP_VISIBLE_DEVICES=0,1,2,3 python3 -m torch.distributed.run --nproc_per_node 4 driver.py #需要运行的程序