Hey everyone,
I am currently working on a project, where I need to optimize conv1D-Layers on ARM CPUs. I followed this guide for autotuning a generic model on ARM CPUs, but when I am using my own model, I don’t get any tasks to optimize.
Since there is no Conv1D template for ARM CPUs (Except M-DSP) I know, that I need to write my own template, but I can’t figure out how and where I need to register my template.
How do I need to use the decorator above the template function and is there a name I need to match?
Thank you very much in advance!
Bene