anijain2305 commented on a change in pull request #4292: Retain qnn input
kernel scales
URL: https://github.com/apache/incubator-tvm/pull/4292#discussion_r344840883
##########
File path: include/tvm/relay/qnn/attrs.h
##########
@@ -177,6 +181,10 @@ struct QnnConv2DAttrs : public
tvm::AttrsNode<QnnConv2DAttrs> {
.describe("The zero point of the input tensor.");
TVM_ATTR_FIELD(kernel_zero_point)
.describe("The zero point of the kernel tensor.");
+ TVM_ATTR_FIELD(input_scale)
+ .describe("The quantization scale for the input tensor.");
Review comment:
Minor - Can we have same description as Dense here :)
"The input tensor scale."
----------------------------------------------------------------
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
