compute_reference.sh

uv run python packages/ltx-trainer/scripts/compute_reference.py \
    /home/xinyuy/flexvideo/LTX-2/temp_test_suite/multi_spatial/condition/000_bg.mp4 \
    --type depth \
    --depth-repo-path packages/ltx-trainer/scripts/Video-Depth-Anything \
    --encoder vitl \
    --num-frames 81

uv run python packages/ltx-trainer/scripts/compute_reference.py \
    /home/xinyuy/flexvideo/LTX-2/temp_test_suite/multi_spatial/condition/000_bg.mp4 \
    --type canny \
    --num-frames 81

uv run python packages/ltx-trainer/scripts/compute_reference.py \
    /home/xinyuy/flexvideo/LTX-2/temp_test_suite/multi_spatial/condition/000.mp4 \
    --type pose \
    --dwpose-repo-path packages/ltx-trainer/scripts/DWPose

uv run python packages/ltx-trainer/scripts/compute_reference.py \
    /mnt/data/xinyuy/datasets/LTX/Pose-Control-Dataset-No-Face \
    --type pose \
    --dwpose-repo-path packages/ltx-trainer/scripts/DWPose \
    --no-face \
    --output /mnt/data/xinyuy/datasets/LTX/Pose-Control-Dataset-No-Face/dataset.json

uv run python packages/ltx-trainer/scripts/process_dataset_multi_gpu.py --num-gpus 8 \
    /mnt/dataset/xinyuy/datasets/LTX-2/Pose-Dataset/dataset.json \
    --resolution-buckets "960x544x81" \
    --model-path /home/xinyuy/flexvideo/LTX-2/ckpts/LTX-2.3/ltx-2.3-22b-dev.safetensors \
    --text-encoder-path /mnt/data/xinyuy/models/Gemma3 \
    --reference-column "reference_path"