train.slurm

#! /bin/bash
#SBATCH --job-name=coop_training
#SBATCH --nodes=1
#SBATCH --partition=gpu32,gpu40,gpu80
#SBATCH --gres=gpu:1
#SBATCH --cpus-per-gpu=2
#SBATCH --mem=10G
#SBATCH --output=/cgi/home/lindsaymb/workspace/documents/code/rcdo/slurm_logs/%x-%j.out
#SBATCH --array=0-100%34
distances=($(seq 0.01 0.0009 0.1))

enroot start --rw \
    -m /cgi/data/erdc_xai:/develop/data \
    -m /cgi/data/erdc_xai/resolution_constrained_deep_optics/results:/develop/results \
    -m /cgi/home/lindsaymb/workspace/documents/code:/develop/code \
    -m /cgi/data/erdc_xai/pretrained_models:/develop/pretrained_models \
    -m /cgi/data/erdc_xai/models:/develop/models \
    rcdo \
    python3 /develop/code/rcdo/experiment.py -config /develop/code/rcdo/config.yaml -job_id $SLURM_JOB_ID -distance ${distances[${SLURM_ARRAY_TASK_ID}]} -model_id coop_lrnxfer_randomInit_${distances[${SLURM_ARRAY_TASK_ID}]}


    #python3 /develop/code/rcdo/experiment.py -config /develop/code/rcdo/config.yaml -job_id $SLURM_JOB_ID -distance ${distances[0]} -model_id baseline_${distances[0]}