Skip to content

Latest commit

 

History

History
9 lines (6 loc) · 728 Bytes

File metadata and controls

9 lines (6 loc) · 728 Bytes

Caution

Deprecated example. Please note, this example is no longer actively maintained and is provided as-is.

RNN unrolling for TFLite

This repository contains a Jupyter notebook that will demonstrate how to train a Recurrent Neural Network (RNN) in TensorFlow and then prepare it for exporting to TensorFlow Lite format by unrolling it.

It also shows how to quantize your unrolled model so that it is ready to be optimized by Vela in order for the model to be deployed onto Arm Ethos-U55 or Arm Ethos-U65 NPU.