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 2018/06/08 06:40:58 UTC

[GitHub] eric-haibin-lin commented on a change in pull request #11194: Enable CUDNN for conv1D

eric-haibin-lin commented on a change in pull request #11194: Enable CUDNN for conv1D
URL: https://github.com/apache/incubator-mxnet/pull/11194#discussion_r193964898
 
 

 ##########
 File path: src/operator/nn/convolution.cc
 ##########
 @@ -384,12 +384,9 @@ struct ConvolutionGrad {
   const char *op_name;
   std::vector<nnvm::NodeEntry> operator()(const nnvm::NodePtr& n,
                                           const std::vector<nnvm::NodeEntry>& ograds) const {
-    const ConvolutionParam& param = nnvm::get<ConvolutionParam>(n->attrs.parsed);
     std::vector<nnvm::NodeEntry> heads(ograds.begin(), ograds.end());
     heads.push_back(n->inputs[conv::kData]);
     heads.push_back(n->inputs[conv::kWeight]);
-    if (!param.no_bias)
-      heads.push_back(n->inputs[conv::kBias]);
 
 Review comment:
   I missed some of the mkldnn code. reverted now.

----------------------------------------------------------------
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