compute-engine icon indicating copy to clipboard operation
compute-engine copied to clipboard

Creating LCE compatible TFLite iOS apps

Open arashb opened this issue 5 years ago • 2 comments

TFLite provides a collection of example mobile apps for specific DL applications. We already have a setup in place to build LCE Lite AAR to support Android (see #238) . No process (and documentation) is currently in place to support iOS. what needs to be done:

  • [ ] setting up the LCE build process for iOS
  • [ ] building the LCE Lite package with integrated LCE custom operators for iOS
  • [ ] integrating the LCE Lite for iOS in one of TF Lite example apps and test on an iOS device

arashb avatar Mar 15 '20 13:03 arashb

to create a LCE Lite package for iOS, first we need to create a LCE C API library similar to TF Lite C API. Then the LCE C API will be used in TensorFlow Lite for iOS and its swift interface.

arashb avatar Mar 21 '20 23:03 arashb

I have this working locally, and will make a PR to add support.

AdamHillier avatar Apr 14 '20 12:04 AdamHillier