Setup

Setting up Coral TPU

Using Coral TPU - Classification: How to identify a parrot? A Step by Step Guide

Solving Coral TPU code fails

Facemask Detection on Reachy using Rasp pi Camera and Coral TPU

Sanity Test

*Based on Using Coral TPU - Classification: How to identify a parrot? A Step by Step Guide

Run the following commands:

cd ~/coral/tflite/python/examples/classification
python3 classify_image.py --model models/mobilenet_v2_1.0_224_inat_bird_quant.tflite --labels models/inat_bird_labels.txt --input images/parrot.jpg

If it's working, you should see the following output:

----INFERENCE TIME----
Note: The first inference on Edge TPU is slow because it includes loading the model into Edge TPU memory.
166.3ms
116.0ms
115.8ms
115.8ms
116.1ms
-------RESULTS--------
Ara macao (Scarlet Macaw): 0.77734

Overview

Edge TPU inferencing overview | Coral

Retrain an image classification model

Retrain an image classification model | Coral

FULL TUTORIAL how to run Object Detection / Classification using Raspi4, and Coral TPU for Reachy with Webcam or RaspiCam

1) Install Raspbian OS + Reachy packages on Raspi4 board + Coral TPU USB ** 1.1) OS on Raspberry 4 :** - I have done a fresh Raspbian OS installation on SD card: GNU/Linux version 10 (“buster”) I chose the SD Card : “SanDisk Ultra 64GB MicroSD” https://www.amazon.de/gp/product/B073JYVKNX/ref=ppx_yo_dt_b_asin_title_o06_s01?ie=UTF8&psc=1  ****