xinyu-intel commented on issue #11937: Fix quantized graphpass bug
URL: https://github.com/apache/incubator-mxnet/pull/11937#issuecomment-412310145
 
 
   @reminisce yes, both gpu and mkldnn will come across below error when 
quantizing residual unit like resnet-50-v1:
   ```
   RuntimeError: simple_bind error. Arguments:
   data: (4, 4, 10, 10)
   softmax_label: (4, 10)
   Error in operator _plus0: [08:21:17] 
/home/chenxiny/mxnet-gpu/3rdparty/mshadow/../../src/operator/tensor/../elemwise_op_common.h:133:
 Check failed: assign(&dattr, (*vec)[i]) Incompatible attr in node _plus0 at 
1-th input: expected float32, got int8
   ```

----------------------------------------------------------------
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:
us...@infra.apache.org


With regards,
Apache Git Services

Reply via email to