在运行时有报错:
rank0: Traceback (most recent call last):
rank0: File "/mnt/code/yzy/swift/swift/cli/sft.py", line 5, in
rank0: File "/mnt/code/yzy/swift/swift/utils/run_utils.py", line 27, in x_main
rank0: result = llm_x(args, **kwargs)
rank0: File "/mnt/code/yzy/swift/swift/llm/sft.py", line 251, in llm_sft
rank0: td0, tkwargs0 = template.encode(train_dataset[0])
rank0: File "/mnt/code/yzy/swift/swift/llm/utils/template.py", line 1049, in encode
rank0: inputs['images'] = inputs2'images': File "/root/miniforge-pypy3/envs/swift/lib/python3.10/site-packages/transformers/tokenization_utils_base.py", line 253, in getitemrank0: return self.data[item]
Describe the bug 我依照文档的介绍构建数据集和运行脚本,运行脚本如下: NPROC_PER_NODE=4 \ CUDA_VISIBLE_DEVICES=0,1,2,3 swift sft \ --model_type glm4v-9b-chat \ --model_id_or_path /mnt/model/glm/glm-4-9b-chat\ --dataset /mnt/code/yzy/data/train_data.jsonl \ --ddp_find_unused_parameters true \ --output_dir /mnt/model/glm/glm4v_ft
我的数据集构建是 {"query": "xx", "response": "1.存在", "images": ["/mnt/code/yzy/data/train/40_1.jpg"]} {"query": "yy", "response": "1.存在", "images": ["/mnt/code/yzy/data/train/44_1.jpg"]}
在运行时有报错: rank0: Traceback (most recent call last): rank0: File "/mnt/code/yzy/swift/swift/cli/sft.py", line 5, in
rank0: File "/mnt/code/yzy/swift/swift/utils/run_utils.py", line 27, in x_main rank0: result = llm_x(args, **kwargs) rank0: File "/mnt/code/yzy/swift/swift/llm/sft.py", line 251, in llm_sft rank0: td0, tkwargs0 = template.encode(train_dataset[0]) rank0: File "/mnt/code/yzy/swift/swift/llm/utils/template.py", line 1049, in encode rank0: inputs['images'] = inputs2'images': File "/root/miniforge-pypy3/envs/swift/lib/python3.10/site-packages/transformers/tokenization_utils_base.py", line 253, in getitem rank0: return self.data[item]
我使用json格式进行输入也是一样的报错
请问如何解决