Change CUDA Version in Colab or Kaggle

deployment
1

Hi, so i go to Start Locally | PyTorch

and choose my system requirements i.e the following:

Pytorch Build: Stable (1.10)

Linux

Pip

Python

CUDA Version: 11.3

The CUDA part can cause problems, as i run the following cmd on both kaggle and colab and get different CUDA version.

Kaggle

!nvcc --version

nvcc: NVIDIA (R) Cuda compiler driver

Copyright (c) 2005-2020 NVIDIA Corporation

Built on Wed_Jul_22_19:09:09_PDT_2020

Cuda compilation tools, release 11.0, V11.0.221

Build cuda_11.0_bu.TC445_37.28845127_0

Google Colab :

nvcc: NVIDIA (R) Cuda compiler driver

Copyright (c) 2005-2020 NVIDIA Corporation

Built on Mon_Oct_12_20:09:46_PDT_2020

Cuda compilation tools, release 11.1, V11.1.105

Build cuda_11.1.TC455_06.29190527_0

So how can i change the CUDA Version to match 11.3 ? Or will it mess up the Notebooks environment. Thanks :slight_smile:

1 Like
2

You won’t be able to change the local CUDA toolkit easily. The PyTorch binaries ship with their own CUDA runtime so unless you are building PyTorch from source or a custom CUDA extension the local CUDA toolkit won’t be used.