Uing cuda with pytorch

I’ve a new machine with a low-gpu and I want to use pytorch with CUDA in practicing deep learning but it gives me false when i run torch.cuda.is_available() even after installing with this cmd :conda install pytorch torchvision torchaudio cudatoolkit=11.3 -c pytorch". Need help in using the CUDA.

Have already installed cuda with “conda install cuda”

nvidia-smi:
±----------------------------------------------------------------------------+
| NVIDIA-SMI 512.59 Driver Version: 512.59 CUDA Version: 11.6 |
|-------------------------------±---------------------±---------------------+
| GPU Name TCC/WDDM | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|===============================+======================+======================|
| 0 NVIDIA GeForce … WDDM | 00000000:01:00.0 Off | N/A |
| N/A 41C P0 N/A / N/A | 0MiB / 2048MiB | 3% Default |
| | | N/A |
±------------------------------±---------------------±---------------------+

±----------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=============================================================================|
| No running processes found |
±----------------------------------------------------------------------------+

Which GPU are you using and what does torch.version.cuda return?

nvidia geforce MX250
returns ‘none’ in python env

nvcc -V:

nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2021 NVIDIA Corporation
Built on Sun_Mar_21_19:24:09_Pacific_Daylight_Time_2021
Cuda compilation tools, release 11.3, V11.3.58
Build cuda_11.3.r11.3/compiler.29745058_0

Still waiting to hear from you @ptrblck

This would mean that you’ve installed a CPU-only binary or built from source without a properly installed CUDA toolkit. Uninstall the current PyTorch installation or create a new virtual environment and reinstall PyTorch with the desired CUDA runtime.