I need help please!
I need to use my GPU to perform CNN, but torch does not seem to detect my CUDA.

When I do

nvcc --version 

on my terminal:

nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2024 NVIDIA Corporation
Built on Fri_Jun_14_16:44:19_Pacific_Daylight_Time_2024
Cuda compilation tools, release 12.6, V12.6.20
Build cuda_12.6.r12.6/compiler.34431801_0

I have set the PATH in my variable environment.

PATH of my CUDA

finally when I do:

import torch

print("CUDA available:", torch.cuda.is_available())
print("CUDA version:", torch.version.cuda)

print(torch.__version__)

I obtain:

CUDA available: False
CUDA version: None
2.3.1+cpu

I’m on windows 11.

I tried to change the version of torch, by the way torch 2.3.0 doesn’t work for some reason…

I expected to see:

CUDA available: True
CUDA version: 12.6
2.3.1+cpu

Sorry for the long post! Any help appreciated