Caenorst commented on a change in pull request #16408: Add MXNet Ops for fast
multihead attention
URL: https://github.com/apache/incubator-mxnet/pull/16408#discussion_r341646687
##########
File path: src/operator/contrib/transformer-inl.h
##########
@@ -34,6 +34,19 @@
namespace mxnet {
namespace op {
+struct InterleavedMatMulParam : public dmlc::Parameter<InterleavedMatMulParam>
{
+ int heads;
+ bool bwd_ignore_zero_init;
+ DMLC_DECLARE_PARAMETER(InterleavedMatMulParam) {
+ DMLC_DECLARE_FIELD(heads)
+ .describe("Set number of heads");
+ DMLC_DECLARE_FIELD(bwd_ignore_zero_init)
+ .describe("Make backward pass ignore AddTo and not init to 0. "
+ " /!\\ Only enable with MXNET_ENABLE_EXEC_ADDTO fonctionality")
Review comment:
I found a way to raise an exception while cover my edge case. By checking
with a custom ParamParser, as it's called during bindind. Hope the solution
suit 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:
us...@infra.apache.org
With regards,
Apache Git Services
