Skip to content

Instantly share code, notes, and snippets.

@adityaiitb
Last active August 5, 2020 23:15
Show Gist options
  • Save adityaiitb/1f4e694141ec43510b2066bfe929eb6e to your computer and use it in GitHub Desktop.
Save adityaiitb/1f4e694141ec43510b2066bfe929eb6e to your computer and use it in GitHub Desktop.
Install CUDA (10.1 or newer) and cuDNN
  1. Download the CUDA toolkit i.e. runfile(local) from Nvidia.
  2. Download the cuDNN .tgz file from Nvidia. You can use this hack from Nvidia Developer Forum to wget a cuDNN file.
  3. Install CUDA toolkit (includes cuBLAS).
export CUDA_ROOT=/where/to/install/cuda
bash cuda_<version>.run --silent \
  --toolkit --toolkitpath=$CUDA_ROOT \
  --samples --samplespath=$CUDA_ROOT/samples 
  --tmpdir=$HOME/tmp
  1. Unpack the cuDNN .tgz into the CUDA root directory.
tar --strip-components=1 -C $CUDA_ROOT -zxvf cudnn-<version>.tgz
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment