added trainig script for testing run in kisski

code/submit_train_dlc_kisski.sh

+#!/bin/bash
+#SBATCH --job-name=train-nn-gpu-dlc
+#SBATCH -t 00:20:00                  # estimated time # TODO: adapt to your needs
+#SBATCH -p kisski              # the partition you are training on (i.e., which nodes), for nodes see sinfo -o "%25N  %5c  %10m  %32f  %10G %18P " | grep gpu
+#SBATCH -G A100:1                   # take 1 GPU, see https://www.hlrn.de/doc/display/PUB/GPU+Usage for more options
+#SBATCH --mem-per-gpu=5G             # setting the right constraints for the splitted gpu partitions
+#SBATCH --nodes=1                    # total number of nodes
+#SBATCH --ntasks=1                   # total number of tasks
+#SBATCH --cpus-per-task=4            # number cores per task
+#SBATCH --output=./slurm_files/slurm-%x-%j.out     # where to write output, %x give job name, %j names job id
+#SBATCH --error=./slurm_files/slurm-%x-%j.err      # where to write slurm error
+
+module load apptainer
+
+# Printing out some info.
+echo "Submitting job with sbatch from directory: ${SLURM_SUBMIT_DIR}"
+echo "Home directory: ${HOME}"
+echo "Working directory: $PWD"
+echo "Current node: ${SLURM_NODELIST}"
+
+pwd
+ls
+
+# For debugging purposes.
+echo ""
+echo "test environment"
+echo "-----------------------------------------------------------------------"
+apptainer exec --nv --bind $PWD /scratch/projects/workshops/gpu-workshop/dlc-dlwgpu.sif python ${PWD}/test_env.py
+echo "-----------------------------------------------------------------------"
+
+# Run the script:
+
+echo ""
+echo "model training"
+echo "-----------------------------------------------------------------------"
+apptainer exec --nv --bind /scratch,$PWD /scratch/projects/workshops/gpu-workshop/dlc-dlwgpu.sif python -u ${PWD}/train_with_logger.py -l ~/${SLURM_JOB_NAME}_${SLURM_JOB_ID}  -t True -p True -d True -s False -f True #TODO adapt path
+echo "-----------------------------------------------------------------------"