research-rainfallradar/aimodel/slurm-TEST-deeplabv3p-rainfall.job

#!/usr/bin/env bash
#SBATCH -J DeepRain
#SBATCH -N 1
#SBATCH -n 14
#SBATCH --gres=gpu:1
#SBATCH -o %j.%N.%a.segment-mono.out.log
#SBATCH -e %j.%N.%a.segment-mono.err.log
#SBATCH -p gpu05,gpu
#SBATCH --time=5-00:00:00
#SBATCH --mem=30000
# ---> in MiB

module load utilities/multi
module load readline/7.0
module load gcc/10.2.0
module load cuda/11.5.0

module load python/anaconda/4.6/miniconda/3.7


show_help() {
	echo -e "Trains a TEST DeepLabv3+ model using rainfall radar and water depth data." >&2;
	echo -e "" >&2;
	echo -e "Usage:" >&2;
	echo -e "    sbatch slurm-TEST-deeplabv3p-rainfall.job" >&2;
	echo -e "" >&2;
	echo -e "....where:" >&2;
	echo -e "    IMAGE_SIZE=128		Optional. Sets the size of the 'images' that the DeepLabV3+ model will work with." >&2;
	echo -e "    BATCH_SIZE=64		Optional. Sets the batch size to train the model with." >&2;
	echo -e "    DIR_RAINFALLWATER	The path to the directory the .tfrecord files containing the rainfall radar / water depth data." >&2;
	echo -e "    PATH_HEIGHTMAP 	The path to the heightmap jsonl file to read in." >&2;
	echo -e "    PATH_COLOURMAP 	The path to the colourmap for predictive purposes." >&2;
	echo -e "    STEPS_PER_EPOCH	The number of steps to consider an epoch. Defaults to None, which means use the entire dataset." >&2;
	echo -e "    POSTFIX			Postfix to append to the output dir (auto calculated)." >&2;
	echo -e "    ARGS				Optional. Any additional arguments to pass to the python program." >&2;
	echo -e "" >&2;
	echo -e "It is strongly advised that all filepaths do NOT contain spaces." >&2;
	echo -e "" >&2;
	echo -e "The code used to identify the run is taken automatically from the filename of the config file." >&2;
	exit;
}

DIR_RAINFALLWATER="${DIR_RAINFALLWATER:-$HOME/rainfallwater_records_tfrecord}";
PATH_HEIGHTMAP="${PATH_HEIGHTMAP:-$HOME/data/terrain50-nimrodsized.json.gz}";
PATH_COLOURMAP="${PATH_COLOURMAP:-$HOME/instance-level-human-parsing/instance-level_human_parsing/human_colormap.mat}";

CODE="train_mono";

if [[ -n "${POSTFIX}" ]]; then
	echo -e ">>> Applying postfix of ${POSTFIX}" >&2;
	CODE="${CODE}_${POSTFIX}";
fi

DIR_OUTPUT="output/$(date -u --rfc-3339=date)_${CODE}";

echo -e ">>> Additional args: ${ARGS}";

export PATH=$HOME/software/bin:$PATH;
export IMAGE_SIZE BATCH_SIZE DIR_RAINFALLWATER PATH_HEIGHTMAP PATH_COLOURMAP STEPS_PER_EPOCH DIR_OUTPUT;

echo ">>> Installing requirements";
conda run -n py38 pip install -r requirements.txt;
echo ">>> Training model";
#shellcheck disable=SC2086
/usr/bin/env time -v conda run -n py38 src/deeplabv3_plus_test_rainfall.py
echo ">>> exited with code $?";