Atlas300I Duo 卡,启动./mindieservice_daemon的时候报错
收藏回复举报
Atlas300I Duo 卡,启动./mindieservice_daemon的时候报错
t('forum.solved') 已解决
发表于2025-04-17 11:09:04
0 查看

Atlas300I Duo 卡 ,拉取的镜像是:

swr.cn-south-1.myhuaweicloud.com/ascendhub/mindie:1.0.0-300I-Duo-py311-openeuler24.03-lts

启动./mindieservice_daemon的时候报错:

[2025-04-17 11:02:52,736] [258] [281466347241824] [llm] [INFO] [logging.py-331] : model_runner.quantize: None, model_runner.kv_quant_type: None, model_runner.fa_quant_type: None, model_runner.dtype: torch.float16 

[2025-04-17 11:02:57,788] [260] [281467202945376] [llm] [INFO] [dist.py-77] : initialize_distributed has been Set 

[2025-04-17 11:02:57,869] [260] [281467202945376] [llm] [INFO] [flash_causal_qwen.py-101] : >>>> qwen_QwenDecoderModel is called. 

2025-04-17 11:02:57,872 [ERROR] model.py:39 - [Model]    >>> Exception:weight transformer.wte.weight does not exist 

Traceback (most recent call last): 

  File "/usr/local/lib/python3.11/site-packages/model_wrapper/model.py", line 37, in initialize 

    return self.python_model.initialize(config) 

           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/usr/local/lib/python3.11/site-packages/model_wrapper/standard_model.py", line 146, in initialize 

    self.generator = Generator( 

                     ^^^^^^^^^^ 

  File "/usr/local/lib/python3.11/site-packages/mindie_llm/text_generator/generator.py", line 79, in __init__ 

    self.generator_backend = get_generator_backend(model_config) 

                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/usr/local/lib/python3.11/site-packages/mindie_llm/text_generator/adapter/__init__.py", line 26, in get_generator_backend 

    return generator_cls(model_config) 

           ^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/usr/local/lib/python3.11/site-packages/mindie_llm/text_generator/adapter/generator_torch.py", line 96, in __init__ 

    super().__init__(model_config) 

  File "/usr/local/lib/python3.11/site-packages/mindie_llm/text_generator/adapter/generator_backend.py", line 107, in __init__ 

    self.model_wrapper = get_model_wrapper(model_config, backend_type) 

                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/usr/local/lib/python3.11/site-packages/mindie_llm/modeling/model_wrapper/__init__.py", line 15, in get_model_wrapper 

    return wrapper_cls(**model_config) 

           ^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/usr/local/lib/python3.11/site-packages/mindie_llm/modeling/model_wrapper/atb/atb_model_wrapper.py", line 46, in __init__ 

    self.model_runner.load_weights() 

  File "/usr/local/Ascend/atb-models/atb_llm/runner/model_runner.py", line 149, in load_weights 

    self.model = self.model_cls(self.config, weights, inference_mode=self.inference_mode, 

                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/usr/local/Ascend/atb-models/atb_llm/models/qwen/flash_causal_qwen.py", line 33, in __init__ 

    self.transformer = FlashQwenModel(config, weights) 

                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/usr/local/Ascend/atb-models/atb_llm/models/qwen/modeling_qwen.py", line 256, in __init__ 

    self.wte = TensorParallelEmbedding( 

               ^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/usr/local/Ascend/atb-models/atb_llm/utils/layers/embedding/tensor_embedding.py", line 44, in __init__ 

    weight = weights.get_partial_sharded(f"{prefix}.weight", dim=1) 

             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/usr/local/Ascend/atb-models/atb_llm/utils/weights.py", line 155, in get_partial_sharded 

    slice_ = self._get_slice(tensor_name) 

             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/usr/local/Ascend/atb-models/atb_llm/utils/weights.py", line 733, in _get_slice 

    filename, tensor_name = self.get_filename(tensor_name) 

                            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/usr/local/Ascend/atb-models/atb_llm/utils/weights.py", line 103, in get_filename 

    raise AssertionError(f"weight {tensor_name} does not exist") 

AssertionError: weight transformer.wte.weight does not exist 

2025-04-17 11:02:57,878 [ERROR] model.py:42 - [Model]    >>> return initialize error result: {'status': 'error', 'npuBlockNum': '0', 'cpuBlockNum': '0'} 

[2025-04-17 11:02:58,729] [258] [281466347241824] [llm] [INFO] [dist.py-77] : initialize_distributed has been Set 

本帖最后由 匿名用户 于 2025/04/17 16:23:46 编辑

我要发帖子