-
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] Add ACL deconvolution operation #18655
[ARM CPU] Add ACL deconvolution operation #18655
Conversation
add api update new update update tests update comments update
…into an/exp_deconv_1
…into an/exp_deconv_1
…into an/exp_deconv_1
static bool customIsSupported(const DeconvAttrs& deconvAttrs, | ||
const std::vector<MemoryDescPtr>& srcDescs, | ||
const std::vector<MemoryDescPtr>& dstDescs) { | ||
if (srcDescs[0]->getShape().getDims().size() != 4 || |
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.
ACL says input could be 3D or 4D and output precision should be exact as input one.
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.
@alvoron done
} | ||
|
||
private: | ||
static bool validate_deconvolution_output_dimensions(unsigned int in_width, unsigned int in_height, |
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.
looks like this method is copied from ACL.
Shouldn't we rely on NEDeconvolutionLayer::validate? I assume it makes the same checks as well.
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.
@alvoron done
…into an/exp_deconv_1
…into an/exp_deconv_1
No description provided.