You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@mxnet.apache.org by GitBox <gi...@apache.org> on 2020/03/23 01:59:13 UTC
[GitHub] [incubator-mxnet] yajiedesign opened a new issue #17888: cuda error
with SequenceLast.
yajiedesign opened a new issue #17888: cuda error with SequenceLast.
URL: https://github.com/apache/incubator-mxnet/issues/17888
## Description
CUDA: an illegal memory access was encountered error with SequenceLast.
### Error Message
CUDA: an illegal memory access was encountered
## To Reproduce
simple lstm net work.
`class LSTMNetwork(gluon.HybridBlock):
def __init__(self, ctx=mx.gpu(0), **kwargs):
super(LSTMNetwork, self).__init__(**kwargs)
self.ctx = ctx
with self.name_scope():
self.main = gluon.nn.HybridSequential()
self.main.add(nn.Dropout(0.5))
self.main.add(rnn.LSTM (128, num_layers=1, dropout=0, bidirectional=True))
self.main.add(nn.BatchNorm())
self.output = gluon.nn.HybridSequential()
self.output.add(nn.Dropout(0.5))
self.output.add(nn.Dense(2, flatten=True))
def hybrid_forward(self, F, x, **kwargs):
x = x.transpose((2, 0, 1))
output = self.main(x)
# this is work done
# output1 = output[-1]
output = F.SequenceLast(output)
output = self.output(output)
return output
`
### Steps to reproduce
## What have you tried to solve it?
## Environment
----------Python Info----------
Version : 3.7.3
Compiler : MSC v.1915 64 bit (AMD64)
Build : ('default', 'Mar 27 2019 17:13:21')
Arch : ('64bit', 'WindowsPE')
------------Pip Info-----------
Version : 20.0.2
Directory : D:\Anaconda3\lib\site-packages\pip
----------MXNet Info-----------
Version : 1.6.0
Directory : D:\Anaconda3\lib\site-packages\mxnet
Num GPUs : 1
Hashtag not found. Not installed from pre-built package.
----------System Info----------
Platform : Windows-10-10.0.18362-SP0
system : Windows
node : DESKTOP-T6GH9E6
release : 10
version : 10.0.18362
----------Hardware Info----------
machine : AMD64
processor : Intel64 Family 6 Model 158 Stepping 12, GenuineIntel
Name
Intel(R) Core(TM) i7-9700KF CPU @ 3.60GHz
----------Network Test----------
Setting timeout: 10
Timing for MXNet: https://github.com/apache/incubator-mxnet, DNS: 0.0040 sec, LOAD: 1.4730 sec.
Timing for GluonNLP GitHub: https://github.com/dmlc/gluon-nlp, DNS: 0.0000 sec, LOAD: 1.4230 sec.
Timing for GluonNLP: http://gluon-nlp.mxnet.io, DNS: 0.3570 sec, LOAD: 2.3780 sec.
Timing for D2L: http://d2l.ai, DNS: 0.5670 sec, LOAD: 0.8950 sec.
Timing for D2L (zh-cn): http://zh.d2l.ai, DNS: 0.0040 sec, LOAD: 0.3690 sec.
Timing for FashionMNIST: https://repo.mxnet.io/gluon/dataset/fashion-mnist/train-labels-idx1-ubyte.gz, DNS: 1.0500 sec, LOAD: 2.0910 sec.
Timing for PYPI: https://pypi.python.org/pypi/pip, DNS: 0.0030 sec, LOAD: 14.5190 sec.
Timing for Conda: https://repo.continuum.io/pkgs/free/, DNS: 0.0040 sec, LOAD: 1.6770 sec.
----------------------------------------------------------------
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:
users@infra.apache.org
With regards,
Apache Git Services