tmp/video-velocity-model/data_processing/ucf101/scripts/run_data_process_siglip2.sh

#!/bin/bash
# Extract SigLIP2 latent features for the UCF101 train split.
# REPO_ROOT is derived from this script's own location, so it works no matter
# which working directory it is invoked from (e.g. via slurm srun).
set -euo pipefail

SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
REPO_ROOT="$(cd "${SCRIPT_DIR}/../../.." && pwd)"

# forward.py imports `stage1` / `utils` from RAE/src.
export PYTHONPATH="${REPO_ROOT}/RAE/src:${PYTHONPATH:-}"

# Prefer the project venv (torch, decord, ...); fall back to any python3.
PYTHON_BIN="${REPO_ROOT}/.venv/bin/python"
if [[ ! -x "${PYTHON_BIN}" ]]; then
    PYTHON_BIN="$(command -v python3)"
fi

exec "${PYTHON_BIN}" "${REPO_ROOT}/data_processing/tools/forward.py" \
  --rae-backbone siglip2 \
  --device cuda:0 \
  --encoder-path /home/sz128/scratch_sz128/models/siglip2-base-patch16-256 \
  --rae-root /home/sz128/scratch_sz128/models/RAE-collections \
  --dataset-name ucf101 \
  --video-data-dir-path /home/sz128/scratch_sz128/datasets/UCF101/data \
  --data-split train