model_name_or_path: Qwen/Qwen2.5-VL-7B-Instruct
image_max_pixels: 262144
video_max_pixels: 16384
trust_remote_code: true

### method
stage: sft
do_train: true
finetuning_type: lora
lora_rank: 32
lora_target: all

### dataset
dataset: EditScore-Reward-Data  
template: qwen2_vl
cutoff_len: 8192
max_samples: 100000
overwrite_cache: true
preprocessing_num_workers: 16
dataloader_num_workers: 4

### output
output_dir: LLaMA-Factory/output/editscore_7B/
logging_steps: 1
save_steps: 250
plot_loss: true
overwrite_output_dir: true
save_only_model: false
report_to: none  # choices: [none, wandb, tensorboard, swanlab, mlflow]

### train
per_device_train_batch_size: 1
gradient_accumulation_steps: 16
learning_rate: 1.0e-4
num_train_epochs: 3.0
lr_scheduler_type: cosine
warmup_ratio: 0.1
bf16: true
ddp_timeout: 180000000
resume_from_checkpoint: null