I have a freshly installed (licensed) BlueIris 5 setup and it suggested to use CodeProjectAI. Since this is on a beefier machine, i decided to give it a try.
I installed CodeProject.AI-Server_2.9.5_win_x64 with the default options which installed 3 modules:
System Details:
(System is VM runing in Proxmox with the Intel UHD 630 passed through.
However, in the status, I see that some modules randomly appear to fail to start with the following errors:
and
So it fails to start, it exists, but it detects objects?
In the module summary it says:
Do I need both YOLOv5 object detection modules?
Should I just keep the .NET one despite the huge error stack since it eventually appears to work and use the GPU?
Any ideas on how to fix the errors:
12:50:21:detect_adapter.py: ModuleNotFoundError: No module named 'yolov5'
I installed CodeProject.AI-Server_2.9.5_win_x64 with the default options which installed 3 modules:
- Face Processing 1.12.3
- Object Detection (YOLOv5 .NET) 1.14.0
- Object Detection (YOLOv5 6.2) 1.10.0
System Details:
(System is VM runing in Proxmox with the Intel UHD 630 passed through.
Code:
12:49:27:System: Windows
12:49:27:Operating System: Windows (Windows 10 Redstone)
12:49:27:CPUs: Intel(R) Core(TM) i5-8500 CPU @ 3.00GHz (Intel)
12:49:27: 1 CPU x 4 cores. 4 logical processors (x64)
12:49:27:GPU (Primary): Intel(R) UHD Graphics 630 (1,024 MiB) (Intel Corporation)
12:49:27: Driver: 31.0.101.2111
12:49:27:System RAM: 10 GiB
12:49:27:Platform: Windows
12:49:27:BuildConfig: Release
12:49:27:Execution Env: Native
12:49:27:Runtime Env: Production
12:49:27:Runtimes installed:
12:49:27: .NET runtime: 9.0.0
12:49:27: .NET SDK: Not found
12:49:27: Default Python: Not found
12:49:27: Go: Not found
12:49:27: NodeJS: Not found
12:49:27: Rust: Not found
However, in the status, I see that some modules randomly appear to fail to start with the following errors:
Code:
12:49:42:Object Detection (YOLOv5 .NET): Object Detection (YOLOv5 .NET) module started.
12:49:44:detect_adapter.py: Traceback (most recent call last):
12:49:44:detect_adapter.py: File "C:\Program Files\CodeProject\AI\modules\ObjectDetectionYOLOv5-6.2\detect_adapter.py", line 25, in
12:49:44:detect_adapter.py: from detect import do_detection
12:49:44:detect_adapter.py: File "C:\Program Files\CodeProject\AI\modules\ObjectDetectionYOLOv5-6.2\detect.py", line 8, in
12:49:44:detect_adapter.py: from yolov5.models.common import DetectMultiBackend, AutoShape
12:49:44:detect_adapter.py: ModuleNotFoundError: No module named 'yolov5'
12:49:44:Module ObjectDetectionYOLOv5-6.2 has shutdown
12:49:44:detect_adapter.py: has exited
and
Code:
True
12:50:54:
12:50:54:Started Object Detection (YOLOv5 .NET) module
12:50:59:ObjectDetectionYOLOv5Net.exe: Application started. Press Ctrl+C to shut down.
12:50:59:ObjectDetectionYOLOv5Net.exe: Hosting environment: Production
12:50:59:ObjectDetectionYOLOv5Net.exe: Content root path: C:\Program Files\CodeProject\AI\modules\ObjectDetectionYOLOv5Net
12:51:00:Object Detection (YOLOv5 .NET): Object Detection (YOLOv5 .NET) module started.
12:51:05:ObjectDetectionYOLOv5Net.exe: Fatal error. 0xC0000005
12:51:05:ObjectDetectionYOLOv5Net.exe: at Microsoft.ML.OnnxRuntime.InferenceSession.Init(Byte[], Microsoft.ML.OnnxRuntime.SessionOptions, Microsoft.ML.OnnxRuntime.PrePackedWeightsContainer)
12:51:05:ObjectDetectionYOLOv5Net.exe: at Microsoft.ML.OnnxRuntime.InferenceSession..ctor(Byte[], Microsoft.ML.OnnxRuntime.SessionOptions)
12:51:05:System.__Canon, System.Private.CoreLib, Version=9.0.0.0, Culture=neutral, PublicKeyToken=7cec85d7bea7798e: ..ctor(System.String, Microsoft.ML.OnnxRuntime.SessionOptions)
12:51:05:ObjectDetectionYOLOv5Net.exe: at CodeProject.AI.Modules.ObjectDetection.YOLOv5.ObjectDetector..ctor(System.String, Microsoft.Extensions.Logging.ILogger`1)
...
12:51:05:ObjectDetectionYOLOv5Net.exe: at System.Threading.PortableThreadPool+IOCompletionPoller+Callback.Invoke(Event)
12:51:05:System.Threading.PortableThreadPool+IOCompletionPoller+Event, System.Private.CoreLib, Version=9.0.0.0, Culture=neutral, PublicKeyToken=7cec85d7bea7798e],[System.Threading.PortableThreadPool+IOCompletionPoller+Callback, System.Private.CoreLib, Version=9.0.0.0, Culture=neutral, PublicKeyToken=7cec85d7bea7798e: .System.Threading.IThreadPoolWorkItem.Execute()
12:51:05:ObjectDetectionYOLOv5Net.exe: at System.Threading.ThreadPoolWorkQueue.Dispatch()
12:51:05:ObjectDetectionYOLOv5Net.exe: at System.Threading.PortableThreadPool+WorkerThread.WorkerThreadStart()
12:51:08:Module ObjectDetectionYOLOv5Net has shutdown
12:51:08:ObjectDetectionYOLOv5Net.exe: has exited
12:55:55:Response rec'd from Object Detection (YOLOv5 .NET) command 'detect' (...c58e1b) ['Found car'] took 4988ms
12:56:09:Response rec'd from Object Detection (YOLOv5 .NET) command 'detect' (...b47ee4) ['Found car'] took 279ms
So it fails to start, it exists, but it detects objects?
In the module summary it says:
- Face Processing 1.12.3 - Stopped
- Object Detection (YOLOv5 .NET) 1.14.0 Started GPU (DirectML)
- Object Detection (YOLOv5 6.2) 1.10.0 CPU
Do I need both YOLOv5 object detection modules?
Should I just keep the .NET one despite the huge error stack since it eventually appears to work and use the GPU?
Any ideas on how to fix the errors:
12:50:21:detect_adapter.py: ModuleNotFoundError: No module named 'yolov5'