modelscope / ms-swift

Use PEFT or Full-parameter to finetune 350+ LLMs or 100+ MLLMs. (LLM: Qwen2.5, Llama3.2, GLM4, Internlm2.5, Yi1.5, Mistral, Baichuan2, DeepSeek, Gemma2, ...; MLLM: Qwen2-VL, Qwen2-Audio, Llama3.2-Vision, Llava, InternVL2, MiniCPM-V-2.6, GLM4v, Xcomposer2.5, Yi-VL, DeepSeek-VL, Phi3.5-Vision, ...)
https://swift.readthedocs.io/zh-cn/latest/Instruction/index.html
Apache License 2.0
3.93k stars 347 forks source link

Finetuning Qwen2.5 with provided script, it comes out AttributeError: 'FieldInfo' object has no attribute 'required'. Did you mean: 'is_required'? #2291

Open Labmem009 opened 4 days ago

Labmem009 commented 4 days ago

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)

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)

Traceback (most recent call last): File "/ssd4/zhuyongchi/facemoji_rephrase/SimejiUM/ms-swift/swift/cli/sft.py", line 5, in sft_main() File "/ssd4/zhuyongchi/facemoji_rephrase/SimejiUM/ms-swift/swift/utils/run_utils.py", line 32, in x_main result = llm_x(args, kwargs) File "/ssd4/zhuyongchi/facemoji_rephrase/SimejiUM/ms-swift/swift/llm/sft.py", line 541, in llm_sft return trainer_train(args, model, template, train_dataset, val_dataset, callbacks=callbacks, msg=msg) File "/ssd4/zhuyongchi/facemoji_rephrase/SimejiUM/ms-swift/swift/llm/sft.py", line 491, in trainer_train trainer.train(training_args.resume_from_checkpoint) File "/ssd4/zhuyongchi/facemoji_rephrase/SimejiUM/ms-swift/swift/trainers/mixin.py", line 479, in train self._save_initial_model(self.args.output_dir) File "/ssd4/zhuyongchi/facemoji_rephrase/SimejiUM/ms-swift/swift/trainers/mixin.py", line 179, in _save_initial_model model = unwrap_model(self.model) File "/ssd4/zhuyongchi/anaconda3/envs/swift/lib/python3.10/site-packages/transformers/modeling_utils.py", line 5237, in unwrap_model return extract_model_from_parallel(model, kwargs) File "/ssd4/zhuyongchi/anaconda3/envs/swift/lib/python3.10/site-packages/accelerate/utils/other.py", line 77, in extract_model_from_parallel from deepspeed import DeepSpeedEngine File "/ssd4/zhuyongchi/anaconda3/envs/swift/lib/python3.10/site-packages/deepspeed/init.py", line 22, in from . import module_inject File "/ssd4/zhuyongchi/anaconda3/envs/swift/lib/python3.10/site-packages/deepspeed/module_inject/init.py", line 6, in from .replace_module import replace_transformer_layer, revert_transformer_layer, ReplaceWithTensorSlicing, GroupQuantizer, generic_injection File "/ssd4/zhuyongchi/anaconda3/envs/swift/lib/python3.10/site-packages/deepspeed/module_inject/replace_module.py", line 567, in from ..pipe import PipelineModule File "/ssd4/zhuyongchi/anaconda3/envs/swift/lib/python3.10/site-packages/deepspeed/pipe/init.py", line 6, in from ..runtime.pipe import PipelineModule, LayerSpec, TiedLayerSpec File "/ssd4/zhuyongchi/anaconda3/envs/swift/lib/python3.10/site-packages/deepspeed/runtime/pipe/init.py", line 6, in from .module import PipelineModule, LayerSpec, TiedLayerSpec File "/ssd4/zhuyongchi/anaconda3/envs/swift/lib/python3.10/site-packages/deepspeed/runtime/pipe/module.py", line 19, in from ..activation_checkpointing import checkpointing File "/ssd4/zhuyongchi/anaconda3/envs/swift/lib/python3.10/site-packages/deepspeed/runtime/activation_checkpointing/checkpointing.py", line 26, in from deepspeed.runtime.config import DeepSpeedConfig File "/ssd4/zhuyongchi/anaconda3/envs/swift/lib/python3.10/site-packages/deepspeed/runtime/config.py", line 29, in from .zero.config import get_zero_config, ZeroStageEnum File "/ssd4/zhuyongchi/anaconda3/envs/swift/lib/python3.10/site-packages/deepspeed/runtime/zero/init.py", line 6, in from .partition_parameters import ZeroParamType File "/ssd4/zhuyongchi/anaconda3/envs/swift/lib/python3.10/site-packages/deepspeed/runtime/zero/partition_parameters.py", line 709, in class Init(InsertPostInitMethodToModuleSubClasses): File "/ssd4/zhuyongchi/anaconda3/envs/swift/lib/python3.10/site-packages/deepspeed/runtime/zero/partition_parameters.py", line 711, in Init param_persistence_threshold = get_config_default(DeepSpeedZeroConfig, "param_persistence_threshold") File "/ssd4/zhuyongchi/anaconda3/envs/swift/lib/python3.10/site-packages/deepspeed/runtime/config_utils.py", line 116, in get_config_default field_name).required, f"'{field_name}' is a required field and does not have a default value" AttributeError: 'FieldInfo' object has no attribute 'required'. Did you mean: 'is_required'?

Jintao-Huang commented 1 day ago

pip install deepspeed==0.14.5