File size: 876 Bytes
cb2428f |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 |
# If `num_labels` is provided, it will be considered a classification task.
# You can also specify `--model Qwen/Qwen2.5-VL-2B-Instruct --use_chat_template true`.
CUDA_VISIBLE_DEVICES=0 \
MAX_PIXELS=1003520 \
swift sft \
--model Qwen/Qwen2-VL-2B \
--train_type lora \
--dataset 'tany0699/garbage265#20000' \
--torch_dtype bfloat16 \
--num_train_epochs 1 \
--per_device_train_batch_size 1 \
--per_device_eval_batch_size 1 \
--learning_rate 1e-4 \
--lora_rank 8 \
--lora_alpha 32 \
--target_modules all-linear \
--gradient_accumulation_steps 16 \
--eval_steps 50 \
--save_steps 50 \
--save_total_limit 2 \
--logging_steps 5 \
--max_length 2048 \
--output_dir output \
--warmup_ratio 0.05 \
--dataloader_num_workers 4 \
--num_labels 265 \
--task_type seq_cls \
--use_chat_template false
|