New CodeProject.AI Object Detection (YOLO11 .NET) Module
- Thread starter MikeLud1
- Start date
I just released Object Detection (YOLO11 .NET) 1.4.0 module for windows. This version supports both YOLO11 and YOLOv5 models. All of my YOLOv5 custom models are included until I can train the YOLO11 versions of the models. The YOLOv5 models have a suffix of -v5, to use them you would for example ipcam-general is now ipcam-general-v5.
BORIStheBLADE
Getting comfortable
A quick observation here...
I've been using YOLOv5 6.2 with the GPU (cuda) for a while. I just tried using yolo 11 and even when I chose the correct device #for my nividia GPU and "enable GPU", it will only use my Intel IGPU. Anyone have the same issue?
I've been using YOLOv5 6.2 with the GPU (cuda) for a while. I just tried using yolo 11 and even when I chose the correct device #for my nividia GPU and "enable GPU", it will only use my Intel IGPU. Anyone have the same issue?
Doing a fresh install today on a new PC and cant get passed this. clicking through some of the links from the appsettings it looks like some of the backend code project links were nuked? Any one else or have any ideas.? clicking check of updates only results in a "unable to check for updates" message
Check the below folder and see if appsettings.json is missing. Also try refreshing the webpage by doing CTRL + F5.
How are you determining that it is only using your Intel iGPU? What number does is show next to GPU like the below screenshot. Also if you open up a command prompt and run nvidia-smi what number is to the left of NVIDIA
BORIStheBLADE
Getting comfortable
You mentioned in my prior question to go off device ID from the task manager. After checking using command prompt, mine shows a different ID. I set it the the correct one based of command prompt and is working now. Thanks!
Attachments
Hey Mike!
Just updated to object detector 1.4.0 and I switched my non-driveway cameras to using ipcam-combined-v5 properly, but I'm getting a lot of (custom) nothing found 0ms inferences now:
Also, ever since I switched to your custom appsettings, the AI Online/Offline Grey/Green icon always starts green then goes grey.
Did I pass something to CPAI incorrectly here?
EDIT: Model folder looks like it didn't download the -v5 model? I probably screwed up something here:
SECOND EDIT: I copied the ipcam-combined.onnx from the YOLO5NET folder and renamed it to -v5 and now the detector is going nuts:
The Yolo11 models detect in about 50ms or less, the combined-v5 are in the 250-750 range.
Is this working properly or did I cludge it together improperly?
Thanks!
Just updated to object detector 1.4.0 and I switched my non-driveway cameras to using ipcam-combined-v5 properly, but I'm getting a lot of (custom) nothing found 0ms inferences now:
Also, ever since I switched to your custom appsettings, the AI Online/Offline Grey/Green icon always starts green then goes grey.
Did I pass something to CPAI incorrectly here?
EDIT: Model folder looks like it didn't download the -v5 model? I probably screwed up something here:
SECOND EDIT: I copied the ipcam-combined.onnx from the YOLO5NET folder and renamed it to -v5 and now the detector is going nuts:
The Yolo11 models detect in about 50ms or less, the combined-v5 are in the 250-750 range.
Is this working properly or did I cludge it together improperly?
Thanks!
Last edited:
I'm not so sure about that. I setup my folders the way you have here and moved the file over and I'm back to 0ms nothing found. I think the models are supposed to be all in the custom-models folder.
That being said, my AI is apparently watching cats cows and pigs ride bicycles around my yard.