hzfan commented on a change in pull request #15851: [Numpy] Numpy copysign
URL: https://github.com/apache/incubator-mxnet/pull/15851#discussion_r324453173
##########
File path: src/operator/numpy/np_elemwise_broadcast_op.cc
##########
@@ -182,5 +202,25 @@
MXNET_OPERATOR_REGISTER_NP_BINARY_SCALAR(_npi_rpower_scalar)
.set_attr<FCompute>("FCompute<cpu>", BinaryScalarOp::Compute<cpu,
mshadow_op::rpower>)
.set_attr<nnvm::FGradient>("FGradient",
ElemwiseGradUseOut{"_backward_rpower_scalar"});
+MXNET_OPERATOR_REGISTER_NP_BINARY_SCALAR(_npi_copysign_scalar)
+.set_attr<FCompute>("FCompute<cpu>", BinaryScalarOp::Compute<cpu,
mshadow_op::copysign>)
+.set_attr<nnvm::FGradient>("FGradient",
ElemwiseGradUseIn{"_backward_npi_copysign_scalar"});
+
+MXNET_OPERATOR_REGISTER_NP_BINARY_SCALAR(_npi_rcopysign_scalar)
+.set_attr<FCompute>("FCompute<cpu>", BinaryScalarOp::Compute<cpu,
mshadow_op::rcopysign>)
+.set_attr<nnvm::FGradient>("FGradient",
ElemwiseGradUseIn{"_backward_npi_rcopysign_scalar"});
+
+MXNET_OPERATOR_REGISTER_BINARY(_backward_npi_copysign_scalar)
+.add_argument("scalar", "float", "scalar value")
Review comment:
Yes. Have fixed it. And Have added additional tests for scalar cases. Thank
you!
----------------------------------------------------------------
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:
[email protected]
With regards,
Apache Git Services
