research-rainfallradar/aimodel/slurm-encoderonly-rainfall.job

#!/usr/bin/env bash
#SBATCH -J EncORain
#SBATCH -N 1
#SBATCH -n 14
#SBATCH --gres=gpu:1
#SBATCH -o %j.%N.%a.encoderonly-rainfall.out.log
#SBATCH -e %j.%N.%a.encoderonly-rainfall.err.log
#SBATCH -p gpu
#SBATCH --no-requeue
#SBATCH --time=5-00:00:00
#SBATCH --mem=30000
# ---> in MiB
# no-requeue: ref https://support.hull.ac.uk/tas/public/ssp/content/detail/incident?unid=652db7ac6e73485c9f7658db78b2b628

module load utilities/multi
module load readline/7.0
module load gcc/10.2.0
module load cuda/11.5.0
export XLA_FLAGS="--xla_gpu_cuda_data_dir=/home/ViperAppsFiles/cuda/11.5.0"; # weird... this wasn't needed before?

module load python/anaconda/4.6/miniconda/3.7


show_help() {
	echo -e "Trains a TEST encoderonly model using rainfall radar and water depth data." >&2;
	echo -e "" >&2;
	echo -e "Usage:" >&2;
	echo -e "    sbatch slurm-encoderonly-rainfall.job" >&2;
	echo -e "" >&2;
	echo -e "....where:" >&2;
	echo -e "    BATCH_SIZE=64		Optional. Sets the batch size to train the model with." >&2;
	echo -e "    DIRPATH_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 "    CHANNELS=8         The number of channels the input data has." >&2;
	echo -e "    WINDOW_SIZE=33     The window size to use when convolving the input dataset for single pixel prediction." >&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 "    VAL_STEPS_PER_EPOCH The number of validation steps to consider an epoch. Defaults to None, which means use the entire dataset." >&2;
	echo -e "    STEPS_PER_EXECUTION	The number of steps to do before returning to do callbacks. High numbers boost performance. Defaults to 1. If set then STEPS_PER_EPOCH and VAL_STEPS_PER_EPOCH must also be set." >&2;
	echo -e "    EPOCHS=25			The number of epochs to train for." >&2;
	echo -e "    LEARNING_RATE      The learning rate to use. Default: 0.001." >&2;
# 	echo -e "    NO_REMOVE_ISOLATED_PIXELS	Set to any value to avoid the engine from removing isolated pixels - that is, water pixels with no other surrounding pixels, either side to side to diagonally." >&2;
# 	echo -e "    PATH_CHECKPOINT 	The path to a checkcpoint to load. If specified, a model will be loaded instead of being trained." >&2;
# 	echo -e "    PREDICT_COUNT		The number of items from the (SCRAMBLED) dataset to make a prediction for." >&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;
	exit;
}

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

CODE="encoderonly_rainfall";

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

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

mkdir -p "${DIRPATH_OUTPUT}";

echo -e ">>> NOW: $(date)";
echo -e ">>> DIR_OUTPUT: ${DIR_OUTPUT}";
echo -e ">>> Additional args: ${ARGS}";

export PATH=$HOME/software/bin:$PATH;
export BATCH_SIZE DIRPATH_RAINFALLWATER PATH_HEIGHTMAP STEPS_PER_EPOCH VAL_STEPS_PER_EPOCH DIRPATH_OUTPUT PATH_CHECKPOINT CHANNELS WINDOW_SIZE EPOCHS LEARNING_RATE STEPS_PER_EXECUTION;
#LOSS ;

echo ">>> Installing requirements";
conda run -n py38 pip install -q -r requirements.txt;
echo ">>> Training model";
#shellcheck disable=SC2016
/usr/bin/env time -v conda run -n py38 bash -c 'src/encoderonly_test_rainfall.py >>"${DIRPATH_OUTPUT}/experiment.${SLURM_JOB_ID}.out.log" 2>>"${DIRPATH_OUTPUT}/experiment.${SLURM_JOB_ID}.err.log"; echo "[slurm_runner] EXIT_CODE: $?" >>"${DIRPATH_OUTPUT}/experiment.${SLURM_JOB_ID}.out.log";';
echo ">>> exited with code $?";