[SOLVED] How to run a TFLITE model trained using GCP on raspberry Pi

Issue

I used the Google Cloud platform, AutoML, to train a model that classifies an image of trash as either "wet" or "dry". Using the GCP I downloaded the trained model as a TFLite model and got:

1. .txt file with the 2 possible labels
2. .json file
3. .tflite file for the actual model.

Given these files, how do I run an inference on the raspberry Pi. I have researched a fair bit and I have not been able to find out how to do the same. Could someone possibly share a helpful resource or any relevant experience that with something similar?

Solution

Have you tried the Raspberry PI example in TFLite: https://github.com/tensorflow/examples/tree/master/lite/examples/image_classification/raspberry_pi

It seems what you want to do is quite similar to that.

Answered By - Tian Lin

Answer Checked By - Dawn Plyler (WPSolving Volunteer)