-
Notifications
You must be signed in to change notification settings - Fork 2.4k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[ARM CPU] Enable deconvolution SLT tests (New) #21263
[ARM CPU] Enable deconvolution SLT tests (New) #21263
Conversation
34b4c42
to
03d87ac
Compare
src/plugins/intel_cpu/tests/functional/custom/subgraph_tests/src/arm/reorder_deconv_nhwc.cpp
Show resolved
Hide resolved
beaaed5
to
bc49ae4
Compare
std::swap(shape[1], shape[2]); | ||
std::swap(shape[2], shape[3]); | ||
|
||
inline void changeLayoutToNH_C(const std::vector<arm_compute::TensorShape*> &_listDims) { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
My major concern is that we have quite untrivial logic to convert CPU internal memory representation to ACL one. We need to revert shape and adjust it in accordance with layout. The problem is that this is done with separate functions which need to be called explicitly. Easy to make a mistake.
Hopefull we can refactor this story during ACL Matmul implementation within new Executors arch.
@allnes @EgorDuplensky
cee22d3
to
4bcf8a8
Compare
CVS-118716 CVS-114087 CVS-127974 CVS-131706 CVS-132490
CVS-118716 CVS-114087 CVS-127974 CVS-131706 CVS-132490
CVS-118716
CVS-114087
CVS-127974
CVS-131706
CVS-132490