szha
commented
7 years ago The gist of the error is:
nvcc fatal : The version ('80100') of the host compiler ('Apple clang') is not supportedThe problem is nvcc doesn't support the new clang complier that comes with the newer Xcode versions. You will need to revert your command line tool to 7.3.3 or 8.0. You may refer to my stackoverflow answer on how to revert the command line tool (note that the answer reverts to 7.2, and you can pick the appropriate version when downloading).
SumNeuron
Please complete the python GPU enabled install for MacOS. This guide is complete / following it to complete results in errors. https://mxnet.incubator.apache.org/get_started/osx_setup.html
Also, please update mx.gpu() to return -1 if no gpu detected. Right now I run mx.gpu() and it returns 0, which means it should be connected to my first gpu (a functioning external NVIDIA GeForce TITAN X), but trying to run anything with this command (e.g. in the demo
mx.nd.ones((100,100),mx.gpu())
results in errors.
Environment info
Operating System: MacOS 10.12.6
Package used (Python/R/Scala/Julia): Python MXNet version: 0.11
MXNet commit hash (
git rev-parse HEAD): a5edbf94094581ee27157eae4f2113115a3994e7Python version and distribution: 3.6
Error Message:
Please paste the full error message, including stack trace.
`>>> b = mx.nd.ones((100,100), gpu_device) [14:16:34] src/c_api/c_api_ndarray.cc:147: GPU support is disabled. Compile MXNet with USE_CUDA=1 to enable GPU support. [14:16:34] /Users/travis/build/dmlc/mxnet-distro/mxnet-build/dmlc-core/include/dmlc/logging.h:308: [14:16:34] src/c_api/c_api_ndarray.cc:416: Operator _ones is not implemented for GPU.
Stack trace returned 5 entries: [bt] (0) 0 libmxnet.so 0x0000000104be1ad8 _ZN4dmlc15LogMessageFatalD2Ev + 40 [bt] (1) 1 libmxnet.so 0x00000001052e0bbf _Z20ImperativeInvokeImplRKN5mxnet7ContextERKN4nnvm9NodeAttrsEPNSt3__16vectorINS_7NDArrayENS7_9allocatorIS9EEEESD + 2079 [bt] (2) 2 libmxnet.so 0x00000001052e17f1 MXImperativeInvoke + 433 [bt] (3) 3 _ctypes.cpython-36m-darwin.so 0x0000000103d3742f ffi_call_unix64 + 79 [bt] (4) 4 ??? 0x00007fff5c4ba950 0x0 + 140734741850448
Traceback (most recent call last): File "", line 1, in
File "/usr/local/lib/python3.6/site-packages/mxnet/ndarray.py", line 1216, in ones
return _internal._ones(shape=shape, ctx=ctx, dtype=dtype, **kwargs)
File "", line 15, in _ones
File "/usr/local/lib/python3.6/site-packages/mxnet/_ctypes/ndarray.py", line 89, in _imperative_invoke
c_array(ctypes.c_char_p, [c_str(str(val)) for val in vals])))
File "/usr/local/lib/python3.6/site-packages/mxnet/base.py", line 129, in check_call
raise MXNetError(py_str(_LIB.MXGetLastError()))
mxnet.base.MXNetError: [14:16:34] src/c_api/c_api_ndarray.cc:416: Operator _ones is not implemented for GPU.
Stack trace returned 5 entries: [bt] (0) 0 libmxnet.so 0x0000000104be1ad8 _ZN4dmlc15LogMessageFatalD2Ev + 40 [bt] (1) 1 libmxnet.so 0x00000001052e0bbf _Z20ImperativeInvokeImplRKN5mxnet7ContextERKN4nnvm9NodeAttrsEPNSt3__16vectorINS_7NDArrayENS7_9allocatorIS9EEEESD + 2079 [bt] (2) 2 libmxnet.so 0x00000001052e17f1 MXImperativeInvoke + 433 [bt] (3) 3 _ctypes.cpython-36m-darwin.so 0x0000000103d3742f ffi_call_unix64 + 79 [bt] (4) 4 ??? `
Minimum reproducible example
if you are using your own code, please provide a short script that reproduces the error.
gpu_device = mx.gpu() b = mx.nd.ones((100,100), gpu_device)
Steps to reproduce
or if you are running standard examples, please provide the commands you have run that lead to the error.
1. 2. 3.
What have you tried to solve it?