#Install Tensorflow With GPU Support On Debian

1. As root,

   ```shell
   echo "deb http://ftp.us.debian.org/debian unstable main contrib non-free" > /etc/apt/sources.list.d/unstable.list
   apt-get update
   apt-get install gcc-5
   sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-5 1 #to set first priority to gcc-5 (credit: https://gist.github.com/beci/2a2091f282042ed20cda)
   ```

2. Install Cuda:

   ```shell
   wget https://developer.nvidia.com/compute/cuda/8.0/Prod2/local_installers/cuda_8.0.61_375.26_linux-run
   chmod +x cuda_8.0.61_375.26_linux.run # (credit: https://unix.stackexchange.com/a/359670)
   sh ./cuda_8.0.61_375.26_linux.run --tar mxvf 
   sudo cp InstallUtils.pm /usr/share/perl5/
   sudo ./cuda_8.0.61_375.26_linux.run --override # `--override` prevents the installer from caring too much about the gcc versions)
   # Clean up:
   rm cuda_8.0.61_375.26_linux.run InstallUtils.pm uninstall_cuda.pl cuda-installer.pl
   rm -r run_files
   ```

3. Add in your `~/.zshrc`:

   ```shell
   export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda/lib64/
   ```

4. Re-login. Install cuDNN v5.1 for Cuda 8.0:

   ```shell
   libcudnn5-dev_5.1.10-1+cuda8.0_amd64.deb
   libcudnn5_5.1.10-1+cuda8.0_amd64.deb
   ```

5. Create Conda Env:

   ```shell
   conda create -n tf                            
   source activate tf
   pip install --ignore-installed https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.2.1-cp36-cp36m-linux_x86_64.whl
   ```