I was following the tutorial for deploying classification using custom vision model on iot edge for Raspberry PI. The link is :
https://docs.microsoft.com/en-us/samples/azure-samples/custom-vision-service-iot-edge-raspberry-pi/custom-vision-azure-iot/
I am using Raspberry PI4 with Raspbian Buster OS. I followed all the steps but my camera-capture module is stopped. My image-classsifier module is also stopped.
I am also not able to view logs as to why the module has stopped.
Hello @AnuragShelar-8594, Just wanted to check on this, did you get a chance to respond to the below comment by Sander? Please let us know how can we help you get unblocked in this matter.
Hi @SatishBoddu-MSFT I did go through it.Did not find anything in edgeAgent logs. The module just stops and no specific error is being shown.
Hello @AnuragShelar-8594,Just checking in if you have had a chance to see the previous response.
Please enable your preferred alert types on your Profile's Settings to get instant notifications on this thread's discussions. Here is how to subscribe to a notification.
It seems the edgeAgent is not able to actually load the modules into the runtime.
Can you check the log of the edgeAgent module? (this could be quite extensive, please search for an actual exception message if applicable).
Hi ,@SandervandeVelde42 .Sorry for late reply. I restarted the edge Daemon and the camera capture module started working. Also I noticed that whenever we make any code changes and push it to ACR, the edge device takes a lot of time to pull the updated code in ACR.
Thanks for the help.
5 people are following this question.
