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 2018/06/19 00:32:46 UTC

[GitHub] jonomon opened a new issue #11329: Cannot call data() on parameters initialized on multiple GPUs

jonomon opened a new issue #11329: Cannot call data() on parameters initialized on multiple GPUs
URL: https://github.com/apache/incubator-mxnet/issues/11329
 
 
   When attempting to access data (using the .data() function) in layers that were initialized on multiple GPUs, the following error message is shown: 
   
   ```
   Parameter 'dense0_weight' was not initialized on context cpu(0). It was only initialized on [gpu(0), gpu(1)].
   ```
   
   ## Minimum reproducible example
   
   Example of code that has layers initialized on multiple GPUs. This code crashes.
   
   ```python code
   import mxnet as mx
   from mxnet import nd, autograd, gluon
   
   net = mx.gluon.nn.Dense(2, in_units=10)
   net.initialize(ctx=[mx.gpu(0), mx.gpu(1)])
   net.weight.data()
   ```
   
   Example of code that has layers initialized on a single GPU. This doesn't crash.
   ```python code
   net = mx.gluon.nn.Dense(2, in_units=10)
   net.initialize(ctx=mx.gpu())
   net.weight.data()
   ```
   
   ## Environment info
   
   * mxnet version = 1.2.0
   * python version 3.6
   * on ubuntu 16.04.4

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on 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