Description
Describe the bug
What the bug is, and how to reproduce, better with screenshots(描述bug以及复现过程，最好有截图)
File "/data/aobozhang/miniconda3/envs/llama-factory/lib/python3.10/site-packages/swift/trainers/mixin.py", line 419, in train
res = super().train(*args, **kwargs)
File "/data/aobozhang/miniconda3/envs/llama-factory/lib/python3.10/site-packages/transformers/trainer.py", line 2245, in train
return inner_training_loop(
File "/data/aobozhang/miniconda3/envs/llama-factory/lib/python3.10/site-packages/transformers/trainer.py", line 2560, in _inner_training_loop
tr_loss_step = self.training_step(model, inputs, num_items_in_batch)
File "/data/aobozhang/miniconda3/envs/llama-factory/lib/python3.10/site-packages/swift/trainers/trainers.py", line 346, in training_step
return super().training_step(model, inputs, *args, **kwargs)
File "/data/aobozhang/miniconda3/envs/llama-factory/lib/python3.10/site-packages/transformers/trainer.py", line 3736, in training_step
loss = self.compute_loss(model, inputs, num_items_in_batch=num_items_in_batch)
File "/data/aobozhang/miniconda3/envs/llama-factory/lib/python3.10/site-packages/swift/trainers/trainers.py", line 326, in compute_loss
loss = compute_loss_func(outputs, labels, num_items_in_batch=num_items_in_batch, **loss_kwargs)
File "/data/aobozhang/miniconda3/envs/llama-factory/lib/python3.10/site-packages/swift/plugin/loss.py", line 520, in channel_loss_func
channels = trainer.args.channels
AttributeError: 'NoneType' object has no attribute 'args'
报错信息
CUDA_VISIBLE_DEVICES=1 swift sft --model_type qwen3 --train_type lora --lora_rank 96 --model /data/changjianhou/Qwen3-8B --dataset /data/changjianhou/DATA/got/ppt_online/all_sft_0623_non_cot_all.json --split_dataset_ratio 0.1 --output_dir /data/changjianhou/DATA/model_output/ppt_model_3——8B/lora_96_v7 --per_device_train_batch_size 1 --num_train_epochs 4 --max_length 23000 --per_device_eval_batch_size 1 --eval_step 1000 --gradient_accumulation_steps 6 --loss_type channel_loss --channels 'think' 'structure7' 'structure6' 'structure5' 'think' 'flow' 'ppt_q1' 'ppt_q2' 'comments'
训练命令
数据集也根据指定的方式进行重置
Your hardware and system info
Write your system info like CUDA version/system/GPU/torch version here(在这里给出硬件信息和系统信息，如CUDA版本，系统，GPU型号和torch版本等)
Additional context
Add any other context about the problem here(在这里补充其他信息)