【求助帖】林希老师 25.2.11，LLM->多模态理论+案例讲解 MiniGPT-v2 #571

New Issue

wangjo · 2025-02-20T22:20:28+08:00

wangjo commented

2025-02-20 22:20:28 +08:00

启动报错
(minigptv) root@autodl-container-26a34fbbce-1d3f4720:~/autodl-tmp/MiniGPT-4# python demo_v2.py --cfg-path eval_configs/minigptv2_eval.yaml --gpu-id 0
Initializing Chat
Loading checkpoint shards: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:04<00:00, 2.46s/it]
Traceback (most recent call last):
File "/root/autodl-tmp/MiniGPT-4/demo_v2.py", line 63, in
model = model_cls.from_config(model_config).to(device)
File "/root/autodl-tmp/MiniGPT-4/minigpt4/models/minigpt_v2.py", line 114, in from_config
model = cls(
File "/root/autodl-tmp/MiniGPT-4/minigpt4/models/minigpt_v2.py", line 46, in init
super().init(
File "/root/autodl-tmp/MiniGPT-4/minigpt4/models/minigpt_base.py", line 41, in init
self.llama_model, self.llama_tokenizer = self.init_llm(
File "/root/autodl-tmp/MiniGPT-4/minigpt4/models/base_model.py", line 199, in init_llm
llama_model = get_peft_model(llama_model, loraconfig)
File "/root/miniconda3/envs/minigptv/lib/python3.9/site-packages/peft/mapping.py", line 145, in get_peft_model
return MODEL_TYPE_TO_PEFT_MODEL_MAPPING[peft_config.task_type](model, peft_config)
File "/root/miniconda3/envs/minigptv/lib/python3.9/site-packages/peft/peft_model.py", line 514, in init
super().init(model, peft_config)
File "/root/miniconda3/envs/minigptv/lib/python3.9/site-packages/peft/peft_model.py", line 79, in init
self.base_model = LoraModel(peft_config, model)
File "/root/miniconda3/envs/minigptv/lib/python3.9/site-packages/peft/tuners/lora.py", line 118, in init
self._find_and_replace()
File "/root/miniconda3/envs/minigptv/lib/python3.9/site-packages/peft/tuners/lora.py", line 152, in _find_and_replace
"memory_efficient_backward": target.state.memory_efficient_backward,
AttributeError: 'MatmulLtState' object has no attribute 'memory_efficient_backward'

是否是包依赖的版本问题，最好老师可以给到MiniGPT-v2完整的requirment.txt信息。

启动报错 (minigptv) root@autodl-container-26a34fbbce-1d3f4720:~/autodl-tmp/MiniGPT-4# python demo_v2.py --cfg-path eval_configs/minigptv2_eval.yaml --gpu-id 0 Initializing Chat Loading checkpoint shards: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:04<00:00, 2.46s/it] Traceback (most recent call last): File "/root/autodl-tmp/MiniGPT-4/demo_v2.py", line 63, in <module> model = model_cls.from_config(model_config).to(device) File "/root/autodl-tmp/MiniGPT-4/minigpt4/models/minigpt_v2.py", line 114, in from_config model = cls( File "/root/autodl-tmp/MiniGPT-4/minigpt4/models/minigpt_v2.py", line 46, in __init__ super().__init__( File "/root/autodl-tmp/MiniGPT-4/minigpt4/models/minigpt_base.py", line 41, in __init__ self.llama_model, self.llama_tokenizer = self.init_llm( File "/root/autodl-tmp/MiniGPT-4/minigpt4/models/base_model.py", line 199, in init_llm llama_model = get_peft_model(llama_model, loraconfig) File "/root/miniconda3/envs/minigptv/lib/python3.9/site-packages/peft/mapping.py", line 145, in get_peft_model return MODEL_TYPE_TO_PEFT_MODEL_MAPPING[peft_config.task_type](model, peft_config) File "/root/miniconda3/envs/minigptv/lib/python3.9/site-packages/peft/peft_model.py", line 514, in __init__ super().__init__(model, peft_config) File "/root/miniconda3/envs/minigptv/lib/python3.9/site-packages/peft/peft_model.py", line 79, in __init__ self.base_model = LoraModel(peft_config, model) File "/root/miniconda3/envs/minigptv/lib/python3.9/site-packages/peft/tuners/lora.py", line 118, in __init__ self._find_and_replace() File "/root/miniconda3/envs/minigptv/lib/python3.9/site-packages/peft/tuners/lora.py", line 152, in _find_and_replace "memory_efficient_backward": target.state.memory_efficient_backward, AttributeError: 'MatmulLtState' object has no attribute 'memory_efficient_backward' 是否是包依赖的版本问题，最好老师可以给到MiniGPT-v2完整的requirment.txt信息。

MiniGPTV2.png

109 KiB