HCAI-Lab/w2-consensus-deepdive-unlearning-artifacts / social-data-attribution-w2 /scripts /slurm /sampling /stratified_sample.sh
| #SBATCH --job-name=stratified_sample | |
| #SBATCH --account=cse | |
| #SBATCH --qos=coc-ice | |
| #SBATCH --cpus-per-task=8 | |
| #SBATCH --mem=64G | |
| #SBATCH --time=04:00:00 | |
| #SBATCH --output=logs/sample/%x_%j.out | |
| #SBATCH --error=logs/sample/%x_%j.err | |
| set -euo pipefail | |
| mkdir -p logs/sample | |
| cd "${SLURM_SUBMIT_DIR:-$(pwd)}" | |
| INPUT_MANIFEST="${INPUT_MANIFEST:-}" | |
| if [[ -n "$INPUT_MANIFEST" ]]; then | |
| TARGET_DOCS_PER_BIN="${TARGET_DOCS_PER_BIN:-100000}" | |
| uv run data-attribution-manifest-sample \ | |
| --input "$INPUT_MANIFEST" \ | |
| --target-docs-per-bin "$TARGET_DOCS_PER_BIN" | |
| else | |
| DUMMY_DOCS="${DUMMY_DOCS:-10000000}" | |
| TARGET_DOCS_PER_BIN="${TARGET_DOCS_PER_BIN:-1000}" | |
| uv run data-attribution-manifest-sample \ | |
| --dummy \ | |
| --n-dummy-docs "$DUMMY_DOCS" \ | |
| --target-docs-per-bin "$TARGET_DOCS_PER_BIN" | |
| fi | |
| echo "Done: $(date)" | |
Xet Storage Details
- Size:
- 837 Bytes
- Xet hash:
- ce1baad1612e1bd3b356d472921d5947b06505d711c323f01031d7e742e2b18c
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.