For those of you who want to try this out on macOS, I would suggest skipping the GPU installation. It doesn't just install and run. Also, there aren't very many video cards that come with Macs that have enough video memory to do even the notMNIST exercises from the Udacity TensorFlow course.<p>If you still feel like you want to install the GPU version on a recent version of macOS (I'm on 10.12.2), you'll see in the installation instructions to run the command<p>$ sudo ln -s /Developer/NVIDIA/CUDA-8.0/lib/libcudnn* /usr/local/cuda/lib/<p>Yes, you need to do that. What you won't find is that you ALSO need to do a<p>export LD_LIBRARY_PATH=/usr/local/cuda/lib<p>to get it to work. Ignore the github issues that talk about
disabling SIP (I didn't need to do that) so that you can set a DYLD_LIBRARY_PATH or other various calls to 'ln'. I just spent hours on figuring this out this week, so I want to pass this on.
Very nice library, but I hope they let us use GPUs in Google Cloud instead of CPUs soon. This is a big non-sense to me, they provide a good library but they don't provide a proper infrastructure to use it yet.