Robin Cole

Try out efficient net

2

Promise to be more performant https://coral.withgoogle.com/models

Add docker container

3

https://blogs.sap.com/2020/02/11/containerizing-a-tensorflow-lite-edge-tpu-ml-application-with-hardware-access-on-raspbian/

https://github.com/bogdannedelcu/yolov5-export-to-coraldevmini

Investigate AutoML Vision

1

What can we do with this, streamline model training? https://cloud.google.com/vision/automl/docs/edge-quickstart

no module edgetpu

1

Hit issue https://github.com/f0cal/google-coral/issues/42 but did not resolve by removing refs to edgetpu. Working with this hardware is frustrating

Coral stick becomes unresponsive

16

Using `sudo systemctl start coral.service` the app appears to die after 12 hours, no errors in logs, just doesn't respond to requests

Can install from https://www.coral.ai/software/

https://github.com/Kao1126/EdgeTPU-FaceNet

As title

As title