runtime icon indicating copy to clipboard operation
runtime copied to clipboard

Mobile PoC level support ETA

Open chunseoklee opened this issue 5 years ago • 4 comments

When is the ETA for supporting mobile devices ?

I wonder the expected performance and memory footprint on android devices.

Thank you.

chunseoklee avatar Apr 30 '20 13:04 chunseoklee

Footprint is a interesting topic I don't know if this project will cover also TF Micro and Edge TPU.

bhack avatar Apr 30 '20 14:04 bhack

Hi. TFRT is a core component where many other efforts align to provide more seamless deployment of TF on mobile deployments. This is work in progress. We will update you once we have a solid ETA.

raziel avatar Apr 30 '20 20:04 raziel

Take a look at https://github.com/tensorflow/community/pull/266

bhack avatar Jul 15 '20 01:07 bhack

Are there any updates to share regarding this question?

Regarding accelerators, are NPU vendors themselves expected to implement drivers to execute BEFs? Or there are existing efforts to support them from TensorFlow side? What about edge TPUs or NNAPI?

anddero avatar Oct 13 '23 14:10 anddero