anijain2305 commented on a change in pull request #4797: [AutoTVM] Minor bug
fixes in AutoTVM for QNN graphs
URL: https://github.com/apache/incubator-tvm/pull/4797#discussion_r374311898
##########
File path: python/tvm/autotvm/graph_tuner/utils/utils.py
##########
@@ -73,7 +73,7 @@ def is_boundary_node(node_entry, input_names):
# Operators dependent on original layouts.
_LAYOUT_FIXED_OP = ["batch_flatten", "transpose", "reshape",
"multibox_prior", "multibox_transform_loc", "where",
- "non_max_suppression", "strided_slice"]
Review comment:
This was causing problem in quantized mobilenet conv2d. QNN inserts strided
slice for quantized conv2d. This caused many strided slice, and somehow graph
tuning dependency map generation was incorrect. We already have a
stridedSliceInferLayout attritbute, so I don't think this is required. But, let
me know, I dont fully understand the graph tuner dependency collection code
yet. If you know, why this is here, I can also test the changes.
----------------------------------------------------------------
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
