| # Model arguments | |
| model_name_or_path: Qwen/Qwen2-VL-2B-Instruct | |
| model_revision: main | |
| torch_dtype: bfloat16 | |
| # Data training arguments | |
| dataset_name: /home/test/test08/fkt/R1-V-main/GEOQA_R1V_Train_8K | |
| dataset_configs: | |
| - all | |
| preprocessing_num_workers: 4 | |
| # SFT trainer config | |
| bf16: true | |
| do_eval: true | |
| eval_strategy: "no" | |
| gradient_accumulation_steps: 4 | |
| gradient_checkpointing: true | |
| gradient_checkpointing_kwargs: | |
| use_reentrant: false | |
| learning_rate: 2.0e-05 | |
| log_level: info | |
| logging_steps: 5 | |
| logging_strategy: steps | |
| lr_scheduler_type: cosine | |
| packing: true | |
| max_seq_length: 4096 | |
| max_steps: -1 | |
| num_train_epochs: 1 | |
| output_dir: ./log/Qwen2-VL-2B-Instruct-SFT | |
| overwrite_output_dir: true | |
| per_device_eval_batch_size: 1 | |
| per_device_train_batch_size: 1 | |
| report_to: | |
| - wandb | |
| save_strategy: "no" | |
| seed: 42 | |
| warmup_ratio: 0.1 |