jinhuang415 commented on a change in pull request #10433: [MXNET-290] MKLDNN
support for model quantization
URL: https://github.com/apache/incubator-mxnet/pull/10433#discussion_r190447983
##########
File path: src/operator/quantization/quantize_graph_pass.cc
##########
@@ -159,7 +159,7 @@ Graph QuantizeGraph(Graph &&src) {
uint32_t min_index = 1;
uint32_t max_index = 2;
if (quantized_op_map.count(e.node->op())) {
- size_t num_outputs = e.node->num_outputs();
+ size_t num_outputs = mirror_node->num_outputs() - 2;
Review comment:
Yes, currently all quantized op only have 1 min and 1 max output so I think
it's ok to make the assumption here? Or is there any better suggested way to
handle this?
----------------------------------------------------------------
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:
[email protected]
With regards,
Apache Git Services
