Benchmarking custom model

[aqibsaeed]

Hi,

Is it possible to run benchmarking on a model (e.g., a custom model built/trained using larq) as described here (https://docs.larq.dev/compute-engine/benchmark/) under Android phone section?

Thanks in advance.

[aqibsaeed]

I was able to run it on android phone, but I have another question about inference times.

"Inference timings in us" what does "us" mean? is it micro seconds?

[aqibsaeed]

Got it from here. https://docs.larq.dev/compute-engine/end_to_end/

[Tombana]

Good to hear that you got it working!

"Inference timings in us" what does "us" mean? is it micro seconds?

That is correct.