05-simple-test-container-torch2.0.1-rocm5.5.1.sh

#!/bin/bash -e

# Source: Tools in action by Samuel Antao (AMD)
# https://docs.lumi-supercomputer.eu/software/packages/pytorch/
# Contribution: M. Jaros (IT4Innovations)

wd=$(pwd)

#
# Example assume allocation was created, e.g.:
# N=1 ; salloc -p standard-g  --threads-per-core 1 --exclusive -N $N --gpus $((N*8)) -t 4:00:00 --mem 0
#

set -x

# download from https://hub.docker.com/r/rocm/pytorch or use from Samuel Antao:
#SIF=/pfs/lustrep2/projappl/project_462000125/samantao-public/containers/lumi-pytorch-rocm-5.5.1-python-3.10-pytorch-v2.0.1-dockerhash-4305da4654f4.sif
SIF=/pfs/lustrep2/projappl/project_462000125/samantao-public/containers/lumi-pytorch-rocm-5.5.1-python-3.10-pytorch-v2.0.1-dockerhash-d55f9163ed80.sif

# check script
if [ ! -f $wd/run-me.sh ] ; then
    echo "use 05-install-container-torch2.0.1-rocm5.5.1.sh"
    exit 1
fi

srun -n 1 \
  singularity exec \
    -B /var/spool/slurmd:/var/spool/slurmd \
    -B /opt/cray:/opt/cray \
    -B /usr/lib64/libcxi.so.1:/usr/lib64/libcxi.so.1 \
    -B $wd:/workdir \
    $SIF /workdir/run-me.sh