[Bug] Conv3Dtranspose default kernel layout should be IODHW #14326
Labels
needs-triage
PRs or issues that need to be investigated by maintainers to find the right assignees to address it
status: help wanted
type: bug
Hi, I found some bugs in TVM relay type relation of conv3dtranspose.
tvm/src/relay/op/nn/convolution.cc
Line 597 in fe3fa9d
I believe the above line should be set to the IODHW layout, since the below conv3dtranspose kernel is made with the IODHW layout.
tvm/src/relay/op/nn/convolution.cc
Lines 629 to 630 in fe3fa9d
Additionally, this line should use 'indexdiv(param->channels, param->groups)' like Conv2dTransposeRel instead of dividing the input channels by the number of groups.
tvm/src/relay/op/nn/convolution.cc
Line 663 in fe3fa9d
The weight of this test is treated as 'OIDHW' layout(the default kernel layout for conv3dtranspose is "OIDHW") despite having 10 input channels and 12 output channels.
tvm/tests/python/relay/test_op_level2.py
Lines 785 to 787 in fe3fa9d
Correct me if I'm wrong. Once again, thanks for your wonderful framework!
The text was updated successfully, but these errors were encountered: