A problem when deepmd was executed in background

[Tubakabura]

I was tto run deepmd in background by adding '&' and wait until the task is accomplished. The job was submitted to slurm system and the script was like

#!/bin/bash
#SBATCH -p gpu
#SBATCH -N 1
#SBATCH --gpus-per-node=1
#SBATCH --cpus-per-gpu=18
export KMP_BLOCKTIME=0
export KMP_AFFINITY=granularity=fine,verbose,compact,1,0
export OMP_NUM_THREADS=9
export TF_INTRA_OP_PARALLELISM_THREADS=9
export TF_INTER_OP_PARALLELISM_THREADS=2
cd **
dp train input.json -l train.log --init-frz-model **.pb &
wait

The job kept running after the training process was over until I kill it manually.

It appears that the job was hindered by a process monitoring the performance of gpu😂. I apologize for my recklessness.