[Speedster] TensorRt OSError: [WinError 127] The specified procedure could not be found
Author: nemeziz69Created Aug 22, 2023Updated Aug 22, 2023
I've installed the TensorRt manually follow this https://docs.nvidia.com/deeplearning/tensorrt/install-guide/index.html#installing-zip. However, when run script use from spedster import optimize_model, got following error:
Traceback (most recent call last):
File "optimize_yolo7_model.py", line 4, in <module>
from speedster import optimize_model
File "D:\NAVIS_ODKI3_Nebullvm\env\lib\site-packages\speedster\__init__.py", line 1, in <module>
from speedster.api.functions import optimize_model # noqa: F401
File "D:\NAVIS_ODKI3_Nebullvm\env\lib\site-packages\speedster\api\functions.py", line 17, in <module>
from speedster.root_op import SpeedsterRootOp
File "D:\NAVIS_ODKI3_Nebullvm\env\lib\site-packages\speedster\root_op.py", line 18, in <module>
from nebullvm.operations.base import Operation
File "D:\NAVIS_ODKI3_Nebullvm\env\lib\site-packages\nebullvm\operations\base.py", line 8, in <module>
from nebullvm.tools.utils import check_device
File "D:\NAVIS_ODKI3_Nebullvm\env\lib\site-packages\nebullvm\tools\\\utils.py", line 36, in <module>
from nebullvm.tools.pytorch import (
File "D:\NAVIS_ODKI3_Nebullvm\env\lib\site-packages\nebullvm\tools\pytorch.py", line 9, in <module>
from nebullvm.tools.diffusers import get_default_dynamic_info
File "D:\NAVIS_ODKI3_Nebullvm\env\lib\site-packages\nebullvm\tools\diffusers.py", line 33, in <module>
from nebullvm.optional_modules.tensor_rt import fold_constants
File "D:\NAVIS_ODKI3_Nebullvm\env\lib\site-packages\nebullvm\optional_modules\tensor_rt.py", line 4, in <module>
import tensorrt
File "C:\Users\User\AppData\Local\Programs\Python\Python38\lib\site-packages\tensorrt\__init__.py", line 129, in <module>
ctypes.CDLL(find_lib(lib))
File "C:\Users\User\AppData\Local\Programs\Python\Python38\lib\ctypes\__init__.py", line 373, in __init__
self._handle = _dlopen(self._name, mode)
OSError: [WinError 127] The specified procedure could not be foundFYI, I run on CUDA version 11.8, cudNN 8.9.0, tensorrt 8.6.1
Source: nebuly-ai/optimate