lihaofd commented on issue #14476: Change RNN OP to stateful
URL: https://github.com/apache/incubator-mxnet/pull/14476#issuecomment-475602423
 
 
   In unix-gpu, the case in Python3: MKLDNN-GPU-NOCUDNN failed,
   May I know the compile option of Python3: MKLDNN-GPU-NOCUDNN?
   Is it make  USE_BLAS=mkl USE_CUDA=1 USE_CUDNN=0 USE_MKLDNN=1 
USE_CUDA_PATH=<cuda path>?
   I build mxnet master 
   commit 29e13b4cbffcf47d3006b9478e242683136fff82
   Author: Nathalie Rauschmayr <[email protected]>
   Date:   Thu Mar 21 19:33:24 2019 -0700
   
   Add then run
   DEVICES=/CUDA_VISIBLE_DEVICES=0 DEVICES=/gpu:0 nosetests -v 
tests/python/gpu/test_gluon_gpu.py:test_lstm_cpu_inference
   it shows error like
   
   Traceback (most recent call last):
     File "/usr/lib/python2.7/site-packages/nose/case.py", line 197, in runTest
       self.test(*self.arg)
     File "/usr/lib/python2.7/site-packages/nose/util.py", line 620, in newfunc
       return func(*arg, **kw)
     File 
"/home/hli32/mxnet/stateful/mxnet_gpu_upstream/tests/python/gpu/../unittest/common.py",
 line 110, in test_new
       orig_test(*args, **kwargs)
     File 
"/home/hli32/mxnet/stateful/mxnet_gpu_upstream/tests/python/gpu/../unittest/test_gluon_rnn.py",
 line 81, in test_lstm_cpu_inference
       x = mx.nd.ones(shape=(2, 2, 2))
     File "/home/hli32/mxnet/mxnet/python/mxnet/ndarray/ndarray.py", line 2421, 
in ones
       return _internal._ones(shape=shape, ctx=ctx, dtype=dtype, **kwargs)
     File "<string>", line 34, in _ones
     File "/home/hli32/mxnet/mxnet/python/mxnet/_ctypes/ndarray.py", line 92, 
in _imperative_invoke
       ctypes.byref(out_stypes)))
     File "/home/hli32/mxnet/mxnet/python/mxnet/base.py", line 252, in 
check_call
       raise MXNetError(py_str(_LIB.MXGetLastError()))
   MXNetError: [20:25:05] src/imperative/imperative.cc:79: Operator _ones is 
not implemented for GPU.
   
   

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

Reply via email to