You can find the compute capability of your device at. Please specify a list of comma-separated Cuda compute capabilities you want to build with. Please specify the location where cuDNN library is installed. Please specify the cuDNN version you want to use. Please specify which gcc should be used by nvcc as the host compiler. Please specify the location where CUDA toolkit is installed. Please specify the CUDA SDK version you want to use, e.g. No OpenCL support will be enabled for TensorFlowĭo you wish to build TensorFlow with CUDA support? yĬUDA support will be enabled for TensorFlowĭo you want to use clang as CUDA compiler? No VERBS support will be enabled for TensorFlowĭo you wish to build TensorFlow with OpenCL support? No XLA support will be enabled for TensorFlowĭo you wish to build TensorFlow with VERBS support? No Hadoop File System support will be enabled for TensorFlowĭo you wish to build TensorFlow with the XLA just-in-time compiler (experimental)? No Google Cloud Platform support will be enabled for TensorFlowĭo you wish to build TensorFlow with Hadoop File System support? Please specify optimization flags to use during compilation when bazel option "-config=opt" is specified :ĭo you wish to build TensorFlow with Google Cloud Platform support? No MKL support will be enabled for TensorFlow
Using python library path: /Users/m/code/3rd/conda/envs/p3gpu/lib/python3.6/site-packagesĭo you wish to build TensorFlow with MKL support? N On my 3.2 GHz iMac this took about 37 minutes.
#Graphics card for mac os sierra install#
Now build with bazel and finish the installation as instructed by the official install guide.But some discussion in a github issue mentioned bazel 0.5.2 also didn’t have the problem. I had some problems building with the latest bazel 0.5.3, so I reverted to using 0.4.5 that I already had installed.Some people had issues with zmuldefs, but I assume that was with earlier versions thanks to udnaan for pointing out that it’s OK to comment out these lines. In my case I commented out line 98 of tensorflow/third_party/gpus/cuda/BUILD.tpl, which contained linkopts = (but the location of the line might obviously change). We could try to build TensorFlow with gcc 4 (which I didn’t manage), or simply remove the line that includes OpenMP from the build file. It should speed up multithreaded TensorFlow on multi-CPU machines, but it will also compile without it. TensorFlow 1.2 expects a C library called OpenMP, which is not available in the current Apple Clang.Using /usr/bin/gcc as your compiler will actually use Clang that comes with macOS / XCode.
Don’t use Clang as the CUDA compiler: this will lead you to an error “Inconsistent crosstool configuration no toolchain corresponding to 'local_darwin' found for cpu 'darwin'.”. I chose the default options in most cases, except for: Python library path, CUDA support and compute capacity. configure, pay attention to the Python library path: it sometimes suggests an incorrect one. Follow the official tutorial “ Installing TensorFlow from Sources”, but obviously substitute git checkout r1.0 with git checkout r1.2.If you have an external GPU, YellowPillow's answer (or mine) might help you get things set up. Once you get that working, the CUDA set-up would also work if you’re compiling TensorFlow. If you haven’t used a TensorFlow-GPU set-up before, I suggest first setting everything up with TensorFlow 1.0 or 1.1, where you can still do pip install tensorflow-gpu.I think it's customary to copy relevant parts to SO, so here it goes: I wrote a little tutorial on compiling TensorFlow 1.2 with GPU support on macOS.
0 kommentar(er)
