train.sh

CKPT=llava-uhd-144-13b
OUTPUT_DIR=./checkpoints_new/$CKPT
LLM_CKPT_DIR=./pretrained_models/vicuna-13b-v1.5
CLIP_CKPT_DIR=./pretrained_models/clip-vit-large-patch14-336
echo $OUTPUT_DIR


# pretraining script
PRBATCH=32
ACCU_STEPS=1
deepspeed \
    --master_port=12322 \
    llava/train/train_mem.py \
    --deepspeed ./scripts/zero2.json \
    --model_name_or_path $LLM_CKPT_DIR \
    --version plain \
    --data_path ./playground/data/LLaVA-Pretrain/blip_laion_cc_sbu_558k.json \
    --image_folder ./playground/data/LLaVA-Pretrain/images \
    --vision_tower $CLIP_CKPT_DIR \
    --mm_projector_type adapt_spatial_resampler \
    --tune_mm_mlp_adapter True \
    --mm_vision_select_layer -2 \
    --mm_use_im_start_end False \
    --mm_use_im_patch_token False \
    --bf16 True \
    --output_dir $OUTPUT_DIR \
    --num_train_epochs 1 \
    --per_device_train_batch_size $PRBATCH \
    --per_device_eval_batch_size 4 \
    --gradient_accumulation_steps $ACCU_STEPS \
    --evaluation_strategy "no" \
    --save_strategy "steps" \
    --save_steps 24000 \
    --save_total_limit 1 \
    --learning_rate 1e-3 \
    --weight_decay 0. \
    --warmup_ratio 0.03 \
    --lr_scheduler_type "cosine" \
    --logging_steps 1 \
    --tf32 True \
    --model_max_length 2048 \
    --gradient_checkpointing True \
    --dataloader_num_workers 4 \
    --lazy_preprocess True \
    --report_to wandb \
    --single True

# full ft script
FTRBATCH=4
ACCU_STEPS=4
deepspeed \
    --master_port=12302 \
    llava/train/train_mem.py \
    --deepspeed ./scripts/zero2.json \
    --model_name_or_path $LLM_CKPT_DIR \
    --version v1 \
    --data_path ./playground/data/llava_v1_5_mix665k.json \
    --image_folder ./playground/data \
    --vision_tower $CLIP_CKPT_DIR \
    --pretrain_mm_mlp_adapter $OUTPUT_DIR/mm_projector.bin \
    --mm_projector_type adapt_spatial_resampler \
    --mm_vision_select_layer -2 \
    --mm_use_im_start_end False \
    --mm_use_im_patch_token False \
    --image_aspect_ratio pad \
    --group_by_modality_length True \
    --bf16 True \
    --output_dir $OUTPUT_DIR \
    --num_train_epochs 1 \
    --per_device_train_batch_size $FTRBATCH \
    --per_device_eval_batch_size 4 \
    --gradient_accumulation_steps $ACCU_STEPS \
    --evaluation_strategy "no" \
    --save_strategy "steps" \
    --save_steps 50000 \
    --save_total_limit 1 \
    --learning_rate 2e-5 \
    --weight_decay 0. \
    --warmup_ratio 0.03 \
    --lr_scheduler_type "cosine" \
    --logging_steps 1 \
    --tf32 True \
    --model_max_length 2048 \
    --gradient_checkpointing True \
    --dataloader_num_workers 4 \
    --lazy_preprocess True \
    --report_to wandb

# evaluation

sh eval.sh $CKPT