Files
CS5446-Project/train.slurm
2025-11-02 15:06:42 +08:00

29 lines
876 B
Bash

#!/bin/bash
#SBATCH --job-name=gen_algo_test # Name of your job
#SBATCH --output=logs/train_%j.out # Output log (%j = job ID)
#SBATCH --error=logs/train_%j.err # Error log
#SBATCH --time=48:00:00 # Max time (hh:mm:ss)
#SBATCH --mail-type=END,FAIL # Email notifications
#SBATCH --mail-user=e1373809@u.nus.edu
#SBATCH --gres=gpu:a100-40:1
#SBATCH -C cuda80
echo "Job started on $(hostname) at $(date)"
# Load environment
source ~/.bashrc
conda activate jlab || { echo "Failed to activate conda"; exit 1; }
# Print GPU info
echo "========== GPU Info =========="
nvidia-smi
echo "=============================="
export LD_LIBRARY_PATH=$CONDA_PREFIX/lib:$LD_LIBRARY_PATH
echo "LD_LIBRARY_PATH set to: $LD_LIBRARY_PATH"
# Run training
papermill proposed_prompt.ipynb outs_prompt.ipynb
echo "Job finished at $(date)"