samskalicky commented on a change in pull request #17569: Adding sparse support
to MXTensor for custom operators
URL: https://github.com/apache/incubator-mxnet/pull/17569#discussion_r392022241
##########
File path: src/c_api/c_api.cc
##########
@@ -341,6 +412,8 @@ int MXLoadLib(const char *path) {
<< "' custom op, Forward or CreateOpState function
was not set.";
CHECK(type_fp != nullptr) << "Error loading '" << name
<< "' custom op, InferType function was not set.";
+ CHECK(stype_fp != nullptr) << "Error loading '" << name
+ << "' custom op, InferSType function was not set.";
Review comment:
Does it need to be set? or should we have a default to make all outputs
dense and then let users specify it only when they are doing sparse?
----------------------------------------------------------------
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
