#!/bin/bash
#SBATCH -A p32852
#SBATCH -p gengpu
#SBATCH --gres=gpu:a100:2
#SBATCH --constraint=sxm
#SBATCH --nodes=2
#SBATCH --ntasks-per-node=2
#SBATCH --time=48:00:00
#SBATCH --mem=64G
#SBATCH --cpus-per-task=4
#SBATCH --job-name=MW-info-typology
#SBATCH --output=parallel-mw-olmo-info-cat.log
#SBATCH --mail-type=BEGIN,END,FAIL
#SBATCH --mail-user=gaughan@u.northwestern.edu

module purge

eval "$(conda shell.bash hook)"

echo "setting up the environment by loading in conda environment at $(date)"

conda activate olmo

echo "running the olmo labeling job at $(date)"

# Get master node address for rendezvous
MASTER_ADDR=$(scontrol show hostnames $SLURM_NODELIST | head -n 1)
MASTER_PORT=29502

export MASTER_ADDR
export MASTER_PORT

srun torchrun \
	--nnodes 2 \
	--nproc-per-node 2 \
	--rdzv_id $SLURM_JOB_ID \
    	--rdzv_backend c10d \
    	--rdzv_endpoint "$MASTER_ADDR:$MASTER_PORT" \
	/home/nws8519/git/mw-lifecycle-analysis/p2/quest/python_scripts/olmo_parallel_cat.py 

echo "unsupervised olmo categorization pau at $(date)"