Your library is really interesting, congrats and thanks for sharing!
I am considering to use it to train over an existing environment that runs on a regular (desktop) JVM, then use that model for predictions on an Android app. I wonder if there is any previous experience or obstacle for of running the library on an Android app (going to try it, but decided to ask beforehand just in case).
An alternative would be to convert the saved trained models to a TensorFlow-compatible format (so it could be used with TensorFlow Lite). Is there any way to do that conversion?
Any suggestions are appreciated! 🙇
Your library is really interesting, congrats and thanks for sharing!
I am considering to use it to train over an existing environment that runs on a regular (desktop) JVM, then use that model for predictions on an Android app. I wonder if there is any previous experience or obstacle for of running the library on an Android app (going to try it, but decided to ask beforehand just in case).
An alternative would be to convert the saved trained models to a TensorFlow-compatible format (so it could be used with TensorFlow Lite). Is there any way to do that conversion?
Any suggestions are appreciated! 🙇