export OMP_NUM_THREADS=8
export NCCL_IB_DISABLE=0
export NCCL_IB_GID_INDEX=3
export NCCL_SOCKET_IFNAME=eth0
export NCCL_DEBUG=INFO
############### Finetune ################
MIXTURE="qwen"
PROMPT_VERSION="qwen_vla"
RUN_NAME="openvla-${MIXTURE}" 
PREV_STAGE_CHECKPOINT="models" 
echo "PREV_STAGE_CHECKPOINT: ${PREV_STAGE_CHECKPOINT}"
echo "MID_RUN_NAME: ${RUN_NAME}"

NUM_GPUS=1
NNODES=1

ACCELERATE_CPU_AFFINITY=1 torchrun --nproc_per_node="${NUM_GPUS}" --nnodes="${NNODES}" \
    llava/train/train_mem.py \
    --deepspeed scripts/zero2.json \
    --model_name_or_path $PREV_STAGE_CHECKPOINT \
    --version $PROMPT_VERSION \
    --data_path scripts/train/vla.yaml \
    --image_folder /images \
    --mm_tunable_parts="mm_language_model" \
    --mm_vision_tower_lr=2e-6 \
    --vision_tower google/siglip-so400m-patch14-384 \
    --mm_projector_type mlp2x_gelu \
    --mm_vision_select_layer -2 \
    --mm_use_im_start_end False \
    --mm_use_im_patch_token False \
    --group_by_modality_length True \
    --image_aspect_ratio anyres_max_9 \
    --image_grid_pinpoints  "(1x1),...,(6x6)" \
    --mm_patch_merge_type spatial_unpad \
    --bf16 True \
    --run_name $RUN_NAME \
    --output_dir /$RUN_NAME \
    --num_train_epochs 5 \
    --per_device_train_batch_size 8 \
    --per_device_eval_batch_size 100 \
    --gradient_accumulation_steps 2 \
    --evaluation_strategy "no" \
    --save_strategy "steps" \
    --save_steps 900 \
    --save_total_limit 2 \
    --learning_rate 1e-4 \
    --weight_decay 0. \
    --warmup_ratio 0.03 \
    --lr_scheduler_type "cosine" \
    --logging_steps 1 \
    --tf32 True \
    --model_max_length 4096 \
    --gradient_checkpointing True \
    --dataloader_num_workers 1 \
    --lazy_preprocess True \
    --report_to wandb \
    --torch_compile True \
    --torch_compile_backend "inductor" \
    --dataloader_drop_last True \
    --frames_upbound 1
exit 0;