按照ChatGLM2-6B部署文档安装完所有库依赖后，使用以下推理命令报错

 

from transformers import AutoTokenizer, AutoModel 

# 修�~T�CHECKPOINT路�~D 

CHECKPOINT="/home/HwHiAiUser/chatglm2-6b/model" 

tokenizer = AutoTokenizer.from_pretrained(CHECKPOINT, trust_remote_code=True) 

model = AutoModel.from_pretrained(CHECKPOINT, trust_remote_code=True, device='npu') 

model = model.eval() 

print("请�~S�~E�对�~]") 

_input=input(">>") 

while _input: 

    response, history = model.chat(tokenizer, _input, history=[]) 

    print(response) 

    _input=input(">>") 

报错信息：

 

Traceback (most recent call last): 

  File "infer.py", line 6, in <module> 

    model = AutoModel.from_pretrained(CHECKPOINT, trust_remote_code=True, device='npu') 

  File "/home/HwHiAiUser/.local/lib/python3.7/site-packages/transformers/models/auto/auto_factory.py", line 480, in from_pretrained 

    pretrained_model_name_or_path, *model_args, config=config, **hub_kwargs, **kwargs 

  File "/home/HwHiAiUser/.local/lib/python3.7/site-packages/transformers/modeling_utils.py", line 2675, in from_pretrained 

    model = cls(config, *model_args, **model_kwargs) 

  File "/home/HwHiAiUser/.cache/huggingface/modules/transformers_modules/model/modeling_chatglm.py", line 881, in __init__ 

    self.transformer = ChatGLMModel(config, empty_init=empty_init, device=device) 

  File "/home/HwHiAiUser/.cache/huggingface/modules/transformers_modules/model/modeling_chatglm.py", line 766, in __init__ 

    self.embedding = init_method(Embedding, config, **init_kwargs) 

  File "/usr/local/python3.7.5/lib/python3.7/site-packages/torch/nn/utils/init.py", line 51, in skip_init 

    return module_cls(*args, **kwargs).to_empty(device=final_device) 

  File "/usr/local/python3.7.5/lib/python3.7/site-packages/torch/nn/modules/module.py", line 788, in to_empty 

    return self._apply(lambda t: torch.empty_like(t, device=device)) 

  File "/usr/local/python3.7.5/lib/python3.7/site-packages/torch/nn/modules/module.py", line 578, in _apply 

    module._apply(fn) 

  File "/usr/local/python3.7.5/lib/python3.7/site-packages/torch/nn/modules/module.py", line 601, in _apply 

    param_applied = fn(param) 

  File "/usr/local/python3.7.5/lib/python3.7/site-packages/torch/nn/modules/module.py", line 788, in <lambda> 

    return self._apply(lambda t: torch.empty_like(t, device=device)) 

  File "/usr/local/python3.7.5/lib/python3.7/site-packages/torch_npu/utils/device_guard.py", line 38, in wrapper 

    return func(*args, **kwargs) 

  File "/usr/local/python3.7.5/lib/python3.7/site-packages/torch_npu/utils/torch_funcs.py", line 108, in _empty_like 

    return torch_npu.empty_like(*args, **kwargs) 

RuntimeError: Initialize:/usr1/workspace/FPTA_Daily_open_pytorchv1.11.0-5.0.rc1/CODE/torch_npu/csrc/core/npu/sys_ctrl/npu_sys_ctrl.cpp:177 NPU error, error code is 4294967295 

E40020: Failed to import Python module [ModuleNotFoundError: No module named 'tbe.common'.]. 

        Solution: Check that all required components are properly installed and the specified Python path matches the Python installation directory. (If the path does not match the directory, run set_env.sh in the installation package.) 

        TraceBack (most recent call last): 

        [GraphOpt][InitializeInner][InitTbeFunc] Failed to init tbe.[FUNC:InitializeInner][FILE:tbe_op_store_adapter.cc][LINE:1533] 

        [SubGraphOpt][PreCompileOp][InitAdapter] InitializeAdapter adapter [tbe_op_adapter] failed! Ret [4294967295][FUNC:InitializeAdapter][FILE:op_store_adapter_manager.cc][LINE:67] 

        [SubGraphOpt][PreCompileOp][Init] Initialize op store adapter failed, OpsStoreName[tbe-custom].[FUNC:Initialize][FILE:op_store_adapter_manager.cc][LINE:115] 

        [FusionMngr][Init] Op store adapter manager init failed.[FUNC:Initialize][FILE:fusion_manager.cc][LINE:357] 

        PluginManager InvokeAll failed.[FUNC:Initialize][FILE:ops_kernel_manager.cc][LINE:100] 

        OpsManager initialize failed.[FUNC:InnerInitialize][FILE:gelib.cc][LINE:188] 

        GELib::InnerInitialize failed.[FUNC:Initialize][FILE:gelib.cc][LINE:121] 

 

THPModule_npu_shutdown success.