From 18e2fb1e778275dff37498da07e6e2e126a74cfc Mon Sep 17 00:00:00 2001
From: mgaughan <mgaughan@proton.me>
Date: Mon, 23 Jun 2025 21:52:47 -0500
Subject: [PATCH] updating the quest version with new ocr stuff

---
 containers/ocr_job_script.sh         | 23 +++++++++++++++++++++++
 containers/olmocr-pipeline-debug.log |  0
 containers/olmocr_docker.sh          |  2 +-
 3 files changed, 24 insertions(+), 1 deletion(-)
 create mode 100644 containers/ocr_job_script.sh
 create mode 100644 containers/olmocr-pipeline-debug.log

diff --git a/containers/ocr_job_script.sh b/containers/ocr_job_script.sh
new file mode 100644
index 0000000..a31454d
--- /dev/null
+++ b/containers/ocr_job_script.sh
@@ -0,0 +1,23 @@
+#!/bin/bash
+#SBATCH -A p32852
+#SBATCH -p gengpu
+#SBATCH --gres=gpu:a100:1
+#SBATCH --nodes=2
+#SBATCH --ntasks-per-node=1
+#SBATCH --time=24:00:00
+#SBATCH --mem=64G
+#SBATCH --cpus-per-task=4
+#SBATCH --job-name=SLR_OCR 
+#SBATCH --output=slr_ocr_job.log
+#SBATCH --mail-type=BEGIN,END,FAIL
+#SBATCH --mail-user=gaughan@u.northwestern.edu
+
+echo "setting up the environment by loading singularity at $(date)"
+
+module load singularity 
+
+echo "singularity loaded, running the job command at $(date)"
+
+singularity run --nv /home/nws8519/git/adaptation-slr/containers/new_olmocr_container.sif python -m olmocr.pipeline /home/nws8519/git/adaptation-slr/ocr_studies_text/ --markdown --pdfs /home/nws8519/git/adaptation-slr/studies_pdfs/*.pdf
+
+echo "job pau at: $(date)"
diff --git a/containers/olmocr-pipeline-debug.log b/containers/olmocr-pipeline-debug.log
new file mode 100644
index 0000000..e69de29
diff --git a/containers/olmocr_docker.sh b/containers/olmocr_docker.sh
index cdc2024..89f582e 100644
--- a/containers/olmocr_docker.sh
+++ b/containers/olmocr_docker.sh
@@ -17,5 +17,5 @@ module load singularity
 
 export SINGULARITY_CACHEDIR=$TMPDIR
 
-singularity pull olmocr_container.sif docker://alleninstituteforai/olmocr:latest
+singularity pull new_olmocr_container.sif docker://alleninstituteforai/olmocr:latest