Oncoming Street facing House Issues with Headlights causing alerts

Yes sorry, by killed I meant it pegged my CPU at 100% so uninstalled it all.

Since then I’ve got DeepquestAI running in a docker container and am using AI Tools as the front end GUI and that runs much better. According to the IOS mobile app, using substreams my CPU used to sit at around 7-10%, with the AI running in docker it now maxes out to around 30% so much, much better.

Still playing with it though so am not fully moved over to using it and am currently testing with 3 of my cams.
 
Yes sorry, by killed I meant it pegged my CPU at 100% so uninstalled it all.

Since then I’ve got DeepquestAI running in a docker container and am using AI Tools as the front end GUI and that runs much better. According to the IOS mobile app, using substreams my CPU used to sit at around 7-10%, with the AI running in docker it now maxes out to around 30% so much, much better.

Still playing with it though so am not fully moved over to using it and am currently testing with 3 of my cams.
Was the high CPU usage due to a big flood of images needing to be processed by deepstack all at once? Making sure your motion trigger settings aren't sending too many pictures to be processed at once may help.
 
Not that I know off as I had the cooldown time set to 3 minutes. The CPU loaf was simply due to me running the actual Windows version of DeepQuest. Since moving over to a docked container CPU usage now idles are around 20% and peaks at around 50% so am much happier.

But moving over to object detection causes its own problems which I’m now looking into it.
 
Not that I know off as I had the cooldown time set to 3 minutes. The CPU loaf was simply due to me running the actual Windows version of DeepQuest. Since moving over to a docked container CPU usage now idles are around 20% and peaks at around 50% so am much happier.

But moving over to object detection causes its own problems which I’m now looking into it.
Okay good to know about the docker container. I still haven't given it a try yet.