You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by GitBox <gi...@apache.org> on 2022/09/08 07:20:31 UTC

[GitHub] [tvm] zenyuhao commented on issue #12672: [Bug] its fuzziness that the num of outputs per node > 1

zenyuhao commented on issue #12672:
URL: https://github.com/apache/tvm/issues/12672#issuecomment-1240326084

   thanks for reply
   In tensorflow graph, output of graph node is naming by `op_name:output_index`, so at least  there alwasy be 0 after outout.
   im not sure that the same design in onnx, and i think the stable way to get output by output index is offical in tvm doc,
   even though get_output_by_name is not stable in some situation, here is simple fix  which works well
   
   
   ```cpp
     for (size_t i = 0; i < outputs_.size(); i++) {
       const uint32_t nid = outputs_[i].node_id;
       std::string& name = nodes_[nid].name;
       std::stringstream ss;
       ss << name << ":" << i;
       output_map_[ss.str()] = i;
     }
   ```
   
   
   


-- 
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.

To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org