OSError: [WinError 1114] A dynamic link library (DLL) initialization routine failed. Error loading "C:\Users\Toby\Desktop\RVC\RVC1006Nvidia\runtime\lib\site-packages\torch\lib\cublas64_11.dll" or one of its dependencies. · RVC-Project/Retrieval-based-Voice-Conversion-WebUI#2019

(2 留言) (0 反應) (0 負責人)Python (18,427 star) (2,849 fork)batch import

help wanted

描述

Whenever I run go-realtime-gui I get this error:

C:\Users\Toby\Desktop\RVC\RVC1006Nvidia>runtime\python.exe guiv1.py Traceback (most recent call last): File "C:\Users\Toby\Desktop\RVC\RVC1006Nvidia\gui_v1.py", line 59, in import torch File "C:\Users\Toby\Desktop\RVC\RVC1006Nvidia\runtime\lib\site-packages\torch_init.py", line 122, in raise err OSError: [WinError 1114] A dynamic link library (DLL) initialization routine failed. Error loading "C:\Users\Toby\Desktop\RVC\RVC1006Nvidia\runtime\lib\site-packages\torch\lib\cublas64_11.dll" or one of its dependencies.

I have CUDA 11.8 installed with Pytorch 2.3.0. I've added environmental variables for CUDA bin and libnvvp. I've also tried reinstalling my drivers, Python and Cuda.

貢獻者指南

技術棧: pythonpytorch
領域: backend
議題類型: bug
難度: 4
預計時間: 1-3 hours
活動狀態: needs maintainer response
清晰度: clear
前置要求: 無
新手友善度: 25
研究方向: The error indicates that the DLL 'cublas64 11.dll' is failing to initialize. This is likely due to a CUDA version mismatch or missing dependencies. The user has CUDA 11.8 and PyTorch 2.3.0, but the DLL version might be incompatible. To resolve, first verify that the CUDA toolkit version matches PyTorch's requirement (e.g., PyTorch 2.3.0 requires CUDA 11.8). Check that the CUDA bin directory is in the PATH and that the Visual C++ redistributables are installed. Use tools like Dependency Walker to identify missing DLLs. Suggest reinstalling PyTorch from the official website with the exact CUDA version.