Add run order to example script file names

example-job-scripts/split-logs.sh → example-job-scripts/00-split-logs.sh

@@ -13,7 +13,7 @@
 module load Anaconda3
 conda activate gpfs
 
-logs=($(find /data/rc/gpfs-policy/data -path "*/list-policy_data-project_list-path-external_slurm-*/raw/*.gz"))
+logs=($(find /data/rc/gpfs-policy/data -path "*/list-policy_data-user_list-path-external_slurm-31[35]*/raw/*.gz"))
 log=${logs[${SLURM_ARRAY_TASK_ID}]}
 
 split-log --no-clobber ${log}
\ No newline at end of file

example-job-scripts/calculate-churn.sh → example-job-scripts/30-calculate-churn.sh

@@ -10,10 +10,10 @@
 #SBATCH --gres=gpu:1
 #SBATCH --output=out/churn-%A-%a.out
 #SBATCH --error=out/churn-%A-%a.err
-#SBATCH --array=1-162
+#SBATCH --array=0-166
 
 module load Anaconda3
-conda activate gpfs-dev
+conda activate gpfs
 
 hive="/data/rc/gpfs-policy/data/gpfs-hive/data-project/"
 tlds=($(find ${hive} -name "tld=*" -type d | sed -n "s/.*tld=//p"))
@@ -23,13 +23,20 @@ echo "TLD: ${tld}"
 
 python << END
 from pathlib import Path
+from datetime import datetime
+from dateutil.rrule import rrule, DAILY
 from rc_gpfs.process.process import calculate_churn
 
 hive_path = Path("${hive}")
 tld = "${tld}"
 
-acqs = hive_path.joinpath(f"tld={tld}").glob("acq=*")
-acq_dates = [p.name.removeprefix("acq=") for p in acqs]
+#acqs = hive_path.joinpath(f"tld={tld}").glob("acq=*")
+#acq_dates = [p.name.removeprefix("acq=") for p in acqs]
+
+start_date = datetime(2025, 1, 20)
+end_date = datetime(2025, 2, 4)
+
+acq_dates = [datetime.strftime(d,'%Y-%m-%d') for d in list(rrule(DAILY, dtstart=start_date, until=end_date))]
 
 with_cuda=True