Model Conversion

How to accelerate your model on ARTPEC-8

0. Before you begin, read the Optimization Tips for information that is useful for the following steps.

1. Train your model using the framework of your choice. Common options are PyTorch or TensorFlow.

2. Quantize (int8/uint8 per-tensor quantization) and export your trained model to the .tflite format. How you do this depends on the framework you use. We provide guides for PyTorch and TensorFlow:
   PyTorch-to-tflite
   TensorFlow-to-tflite

3. Run the quantized .tflite model on your device. Example test run:

   larod-client -g <my-model>.tflite -c axis-a8-dlpu-tflite -i ""

   More info here

Flowchart