coral-pi-rest-server
                                
                                
                                
                                    coral-pi-rest-server copied to clipboard
                            
                            
                            
                        Perform inferencing of tensorflow-lite models on an RPi with acceleration from Coral USB stick
I'm stuck with the legacy api due to inability to switch to 2.0 / pycoral. Probably has something to do with me having the m2 TPU rather than the USB...
https://github.com/ricardodeazambuja/ALPR-EdgeTPU-RPI0
https://github.com/ricardodeazambuja/Maple-Syrup-Pi-Camera
I have a possible use case for this, but the rest server would be running on windows rather than a Pi. Would that be possible? I don't have a coral...
https://github.com/google-coral/aiy-maker-kit
I'm getting a lot of bounding boxes, a LOT. Many of them are often very low confidence. Is there any way to correct this?    
Promise to be more performant https://coral.withgoogle.com/models
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
Flask debug mode causes the app to initialise twice, opening another "handle" to the edge TPU. Turning debug mode off fixes this. Ref: https://stackoverflow.com/questions/9449101/how-to-stop-flask-from-initialising-twice-in-debug-mode#:~:text=When%20building%20a%20Flask%20service,Flask%20service%20only%20initialises%20once.