DCU vllm封装包运行异常
Traceback (most recent call last):
File "test_vllm.py", line 22, in <module> 
llm = LLM(model="/mnt/pengfeng/kubernetes-dcu/models/opt-1b3")
File "/root/miniconda3/envs/pytorch/lib/python3.8/site-packages/vllm/entrypoints/llm.py", line 109, in init
self.llm_engine = LLMEngine.from_engine_args(engine_args)
File "/root/miniconda3/envs/pytorch/lib/python3.8/site-packages/vllm/engine/llm_engine.py", line 391, in from_engine_args
engine = cls(*engine_configs,
File "/root/miniconda3/envs/pytorch/lib/python3.8/site-packages/vllm/engine/llm_engine.py", line 128, in init
self._init_workers()
File "/root/miniconda3/envs/pytorch/lib/python3.8/site-packages/vllm/engine/llm_engine.py", line 181, in _init_workers
self._run_workers("load_model")
File "/root/miniconda3/envs/pytorch/lib/python3.8/site-packages/vllm/engine/llm_engine.py", line 1041, in _run_workers
driver_worker_output = getattr(self.driver_worker,
File "/root/miniconda3/envs/pytorch/lib/python3.8/site-packages/vllm/worker/worker.py", line 100, in load_model
self.model_runner.load_model()
File "/root/miniconda3/envs/pytorch/lib/python3.8/site-packages/vllm/worker/model_runner.py", line 88, in load_model
self.model = get_model(self.model_config,
File "/root/miniconda3/envs/pytorch/lib/python3.8/site-packages/vllm/model_executor/utils.py", line 49, in get_model
imported_model_loader = importlib.import_module(
File "/root/miniconda3/envs/pytorch/lib/python3.8/importlib/init .py", line 127, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
File "  <frozen importlib._bootstrap>  ", line 1014, in _gcd_import
File "  <frozen importlib._bootstrap>  ", line 991, in _find_and_load
File "  <frozen importlib._bootstrap>  ", line 975, in _find_and_load_unlocked
File "  <frozen importlib._bootstrap>  ", line 671, in _load_unlocked
File "  <frozen importlib._bootstrap_external>  ", line 843, in exec_module
File "  <frozen importlib._bootstrap>  ", line 219, in _call_with_frames_removed
File "/root/miniconda3/envs/pytorch/lib/python3.8/site-packages/vllm/model_executor/model_loader.py", line 10, in <module> 
from vllm.model_executor.weight_utils import (get_quant_config,
File "/root/miniconda3/envs/pytorch/lib/python3.8/site-packages/vllm/model_executor/weight_utils.py", line 18, in <module> 
from vllm.model_executor.layers.quantization import (get_quantization_config,
File "/root/miniconda3/envs/pytorch/lib/python3.8/site-packages/vllm/model_executor/layers/quantization/init .py", line 4, in <module> 
from vllm.model_executor.layers.quantization.awq import AWQConfig
File "/root/miniconda3/envs/pytorch/lib/python3.8/site-packages/vllm/model_executor/layers/quantization/awq.py", line 6, in <module> 
from vllm._C import ops
ImportError: /root/miniconda3/envs/pytorch/lib/python3.8/site-packages/vllm/_C.cpython-38-x86_64-linux-gnu.so: undefined symbol: _ZN3c106detail23torchInternalAssertFailEPKcS2_jS2_RKSs

用的是社区里面封装的vllm的whl