For those who don't have Keras or TensorFlow on GPU on macOS Sierra
Introduction
It's not a solution, this problem is possible
The problem that occurred
――The following error appears even though both CUDA and CuDNN are entered correctly.
- Library not loaded: @rpath/libcudnn.5.dylib
--The GPU is not used even though it is entered correctly (the one below is not displayed)
- I tensorflow/stream_executor/dso_loader.cc:128] successfully opened CUDA library libcublas.8.0.dylib locally
- I tensorflow/stream_executor/dso_loader.cc:128] successfully opened CUDA library libcudnn.5.dylib locally
- I tensorflow/stream_executor/dso_loader.cc:128] successfully opened CUDA library libcufft.8.0.dylib locally
- I tensorflow/stream_executor/dso_loader.cc:128] successfully opened CUDA library libcuda.1.dylib locally
- I tensorflow/stream_executor/dso_loader.cc:128] successfully opened CUDA library libcurand.8.0.dylib locally
The cause
Let's check the GPU, isn't it made by Interl instead of NVIDIA?
It took me a while to notice because the GPU mode was available on the previous machine, but it seems that the GPU is no longer made by NVIDIA from around the 2015 model, and the GPU mode can no longer be used. It can't be helped, so let's run it in CPU mode. .. ..
If anyone knows any good way, please comment. m (_ _) m