jwfromm commented on a change in pull request #5132: [Torch] Fix conv2d
conversion for group conv (group > 1 but != in channels)
URL: https://github.com/apache/incubator-tvm/pull/5132#discussion_r396793785
##########
File path: python/tvm/relay/frontend/pytorch.py
##########
@@ -250,7 +250,9 @@ def _impl(inputs, input_types):
channels = weight_shape[0]
groups = int(inputs[8])
- if groups > 1:
+ # weight_shape[1] is always in_channels // groups
+ # For depthwise, in_channels == groups
Review comment:
This comment is a little confusing, Make it clear that we're checking if
this is a depthwise conv and that if weight_shape[1] != 1, then its a group
conv.
----------------------------------------------------------------
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
