Recent content by hapstabu

FWIW, I've seen this happen myself. If you are using one of the YOLO modules with Custom Models checked on the AI tab, and change the CPAI module to Coral, you will continue getting responses from CPAI until either Blue Iris restarts, or the machine reboots. Then, you will get "Nothing Found"...

As far as I know, the Coral module will not support the "Use Custom" in the AI tab. I have only had luck with the Coral module by unchecking the Use Custom checkbox on the AI tab and checking the Default Object Detection checkbox. However, I have had some issues with the latest Coral version...

@mailseth I had tried to change the Coral model size from small to medium several days ago. It took a lot of fiddling to get the status to stop displaying CPU Tflite to TPU Tflite. I am now looking at the log after several days being occupied with other things, and I am seeing lots and lots of...

Label on my monitor helps me remember. Sent from my iPlay_50 using Tapatalk

I can try uninstalling and reinstalling the Coral module and see if that improves it. I don't suppose there is a somewhat easy way to roll back to ver 2.2.0? I could swear that I was seeing zero fails with 2.2.0 I believe.

I don't really know what to look for. In searching, I don't see any responses to "fail" or "error". I do see a bunch of these entries: 2024-04-21 16:26:52: objectdetection_coral_adapter.py: Ï?™äÜ4¬äÕ?®AŽ½Ó>Ó?‘ƒ•™qÃ?eÔ:Á -?=XlÔ€²©¿ Info: no TPU interpreters: Falling back to CPU...

I am sorry to say, I just updated the Coral Module from ver 2.2.0 to 2.2.2. I am now seeing the same behavior that I was seeing regarding the failed inferences. AFAIK, the settings for the Coral module are the same. Environment Variables CPAI_CORAL_MODEL_NAME = YOLOv8 CPAI_CORAL_MULTI_TPU...

FYI, you are replying to a post that I made which was based on data from late February. Since then, @mailseth has been working on the Coral object detection module. Until last week, I was using the 2.2.0 version of the Coral module, with excellent results. Having seen zero failed inferences on...

For me also, the results with the three models for the Coral tpu are poor. My mailbox is a toilet, my garage is a truck, the street is a train, a tree next to a distant house is a traffic light. I also am using Frigate NVR with a Coral USB on a Linux machine, with Home Assistant to handle...

So does that mean that Ken has added some functionality to modify the models listbox contents, as opposed to the listbox being populated automatically when Blue Iris starts? (And you could do nothing to change it?) Interesting. I'll have to look at that. Sent from my iPlay_50 using Tapatalk

CodeProject.AI Server service needs to be up and running before Blue Iris starts. Blue Iris pulls the models when it starts, so if CodeProject is not there yet, you will get no model list. Set the Blue Iris service to Automatic (Delayed Start). Sent from my iPlay_50 using Tapatalk

I am also back to yolo5vNET (Medium) also with custom models. I see the same thing with the Coral M.2. Many "nothing found" in Blue Iris, also, in the week or so that I used the Coral, I saw about 18% failed inferences. Meaning 18,000 failed out of ~100,000 inferences. With the Yolo5vNet, before...

All: Blue Iris sends a call to CodeProject.AI.Server to list the models when Blue Iris starts (or starts the service if you are running it as a service.) CodeProject.AI.Server service needs to be running BEFORE Blue Iris starts. That's why sometimes you will restart the machine and that field...

I changed to CPAI from Deep stack about six months ago. I am currently on ver 2.5.1. Sent from my iPlay_50 using Tapatalk

My advice: Check "Download Only". Then if you want, run the update manually. If it doesn't work for you, you can run an earlier update. Monitor this thread for indications that people are having trouble with an update. As far as Critical or Stable is concerned, I have occasionally had issues...