Hi Tensorflow Lite Users,
A couple of weeks ago we announced Android ML Platform at Android Developers Blog in order to help with the common challenges development teams face while deploying on-device ML on Android.
You can read the blog post here, but as a summary: we will provide tools for you to reduce your apk sizes and get optimal performance on all devices whilst improving consistency of NNAPI.
As we are rolling out the Android ML platform throughout the year, we also kicked off an Early Access Program for Tensorflow Lite for Android -- Android’s new Tensorflow Lite runtime pre-installed to all devices with Google Play Services.
By migrating to Tensorflow Lite for Android, you can immediately save on your apk sizes and keep your Tensorflow Lite runtime updated automatically.
Android ML team is looking for development teams to partner with for this EAP - if you are interested in this opportunity, you can sign-up here.
Thanks,
Ali on behalf of Android ML and Tensorflow Lite teams