## 20150421

### Linux Mint 17.1, Nvidia, CUDA, and cuDNN

I recently replaced a Titan X, which was on loan, with a GTX 980. After messing with drivers for nearly a day I was able to get my dual monitor setup running again. Unfortunately whatever i did freaked out Theano yielding the error:
export LD_LIBRARY_PATH=/usr/local/cuda-7.0/lib64:$LD_LIBRARY_PATH export CUDA_ROOT=/usr/local/cuda-7.0 #### cuDNN Installing cuDNN for use with Theano can be found on the cuDNN page of deeplearning.net. I used the first method currently suggested on that page which is to copy *.h to$CUDA_ROOT/includes and *.so* to \$CUDA_ROOT/lib64.

1. Thanks for these installation instructions, very helpful!

I found that by installing CUDA, the 346 video drivers are automatically installed, so the first steps involving the ppa repository can be skipped.

Also I did not separately install the build-essential packages, but it was installed after installing CUDA. So either it's installed with Linux Mint 17.1 or the CUDA installation takes care of that.

(I used the Ubuntu 14.04 Local Package Installer DEB instead of the Network version, so perhaps there's a difference there.)

2. This comment has been removed by a blog administrator.

3. Thanks! very straight forward approach. Much easier than other methods.

4. If network installer did not work properly with "apt-get install", use "aptitude install" instead.
Thanks for the guide.