|
Thank you kindly. One final thing, could you please share your System Info tab from your CodeProject.AI Server dashboard?
Thanks,
Sean Ewington
CodeProject
|
|
|
|
|
Server version: 2.6.2
System: Windows
Operating System: Windows (Microsoft Windows 10.0.19045)
CPUs: Intel(R) Core(TM) i7-6700T CPU @ 2.80GHz (Intel)
1 CPU x 4 cores. 8 logical processors (x64)
GPU (Primary): NVIDIA GeForce GTX 960 (4 GiB) (NVIDIA)
Driver: 552.12, CUDA: 12.4 (up to: 12.4), Compute: 5.2, cuDNN:
System RAM: 16 GiB
Platform: Windows
BuildConfig: Release
Execution Env: Native
Runtime Env: Production
Runtimes installed:
.NET runtime: 7.0.5
.NET SDK: Not found
Default Python: Not found
Go: Not found
NodeJS: Not found
Rust: Not found
Video adapter info:
NVIDIA GeForce GTX 960:
Driver Version 31.0.15.5212
Video Processor NVIDIA GeForce GTX 960
Microsoft Remote Display Adapter:
Driver Version 10.0.19041.3636
Video Processor
System GPU info:
GPU 3D Usage 0%
GPU RAM Usage 2.8 GiB
Global Environment variables:
CPAI_APPROOTPATH = <root>
CPAI_PORT = 32168
|
|
|
|
|
Sorry, one more thing. In C:/Program Files/CodeProject/AI/modules/ObjectDetectionYOLOv8 you will find a install.log file. Can you please share that with us?
Thanks,
Sean Ewington
CodeProject
|
|
|
|
|
2024-04-17 19:38:50: Installing CodeProject.AI Analysis Module
2024-04-17 19:38:50: ======================================================================
2024-04-17 19:38:50: CodeProject.AI Installer
2024-04-17 19:38:50: ======================================================================
2024-04-17 19:38:50: 91.9Gb of 953Gb available on
2024-04-17 19:38:50: General CodeProject.AI setup
2024-04-17 19:38:50: Creating Directories...Done
2024-04-17 19:38:50: GPU support
2024-04-17 19:38:50: CUDA Present...Yes (CUDA 11.3, No cuDNN found)
2024-04-17 19:38:51: ROCm Present...No
2024-04-17 19:38:53: Reading ObjectDetectionYOLOv8 settings.......Done
2024-04-17 19:38:53: Installing module Object Detection (YOLOv8) 1.4.3
2024-04-17 19:38:53: Installing Python 3.9
2024-04-17 19:38:53: Python 3.9 is already installed
2024-04-17 19:39:03: Creating Virtual Environment (Local)...Done
2024-04-17 19:39:04: Confirming we have Python 3.9 in our virtual environment...present
2024-04-17 19:39:04: Installing Python packages for Object Detection (YOLOv8)
2024-04-17 19:39:04: [0;Installing GPU-enabled libraries: If available
2024-04-17 19:39:07: Ensuring Python package manager (pip) is installed...Done
2024-04-17 19:39:19: Ensuring Python package manager (pip) is up to date...Done
2024-04-17 19:39:19: Python packages specified by requirements.windows.cuda.txt
2024-04-17 19:40:34: - Installing PyTorch, an open source machine learning framework...(checked) Done
2024-04-17 19:42:03: - Installing TorchVision, for working with computer vision models...(checked) Done
2024-04-17 19:44:16: - Installing Ultralytics package for object detection in images...(checked) Done
2024-04-17 19:44:16: Installing Python packages for the CodeProject.AI Server SDK
2024-04-17 19:44:18: Ensuring Python package manager (pip) is installed...Done
2024-04-17 19:44:21: Ensuring Python package manager (pip) is up to date...Done
2024-04-17 19:44:21: Python packages specified by requirements.txt
2024-04-17 19:44:23: - Installing Pillow, a Python Image Library...Already installed
2024-04-17 19:44:24: - Installing Charset normalizer...Already installed
2024-04-17 19:44:30: - Installing aiohttp, the Async IO HTTP library...(checked) Done
2024-04-17 19:44:34: - Installing aiofiles, the Async IO Files library...(checked) Done
2024-04-17 19:44:35: - Installing py-cpuinfo to allow us to query CPU info...Already installed
2024-04-17 19:44:36: - Installing Requests, the HTTP library...Already installed
2024-04-17 19:44:46: Self test: Self-test passed
2024-04-17 19:44:46: Module setup time 00:05:55.23
2024-04-17 19:44:46: Setup complete
2024-04-17 19:44:46: Total setup time 00:05:56.17
Installer exited with code 0
|
|
|
|
|
That's odd. Self-test passed but you're getting a "model not found" error.
What's in the folder C:\Program Files\CodeProject\AI\modules\ObjectDetectionYOLOv8\assets ?
cheers
Chris Maunder
|
|
|
|
|
Hi Chris,
Do you know if ipcam-combined.pt will be added to yolov8 soon? Only has general.pt at the moment. Always found combined faster than general in older models like yolov3 and 5 (less objects?)
Thanks
|
|
|
|
|
I have been busy and had no time to make all the YOLOv8 Custom Models. Hopefully in a month or two I will get to making the models.
|
|
|
|
|
nothing, there is no assets folder in C:\Program Files\CodeProject\AI\modules\ObjectDetectionYOLOv8\
|
|
|
|
|
Hi,
Sorry, if I am sounding like a broken record, as this is my second time asking for help with the Coral module not starting. But since v2.2 and the one before and even with the newest v2.2.2 of the Coral object detection. I have been unable to actually get the module to work.
The last time I asked, it was an issue with selecting Multi-TPU support, which fixed it and allow the module to start.
However, ever since that release, it has been unable to start, falling back to CPU. Everything I've tried , changing models, changing size of models, re-installing, enabling Multi-TPU results in the same. The module fails to start or falls back to CPU.
The service also restarts most of the time (on windows) as well?
Below fails to start
Any suggests would be appreciated?
Thanks
12:17:58:Server version: 2.6.2
12:18:01:
12:18:01:Module 'Object Detection (Coral)' 2.2.2 (ID: ObjectDetectionCoral)
12:18:01:Valid: True
12:18:01:Module Path: <root>\modules\ObjectDetectionCoral
12:18:01:AutoStart: True
12:18:01:Queue: objectdetection_queue
12:18:01:Runtime: python3.9
12:18:01:Runtime Loc: Local
12:18:01:FilePath: objectdetection_coral_adapter.py
12:18:01:Start pause: 1 sec
12:18:01:Parallelism: 16
12:18:01:LogVerbosity:
12:18:01:Platforms: all
12:18:01:GPU Libraries: installed if available
12:18:01:GPU Enabled: enabled
12:18:01:Accelerator:
12:18:01:Half Precis.: enable
12:18:01:Environment Variables
12:18:01:CPAI_CORAL_MODEL_NAME = EfficientDet-Lite
12:18:01:CPAI_CORAL_MULTI_TPU = True
12:18:01:MODELS_DIR = <root>\modules\ObjectDetectionCoral\assets
12:18:01:MODEL_SIZE = small
12:18:01:
12:18:01:Started Object Detection (Coral) module
12:18:04:Server: This is the latest version
modified 23-Apr-24 12:51pm.
|
|
|
|
|
No need to apologise. Let's get this sorted.
Can you remind me what system you're on (you mentioned Windows). Are you using Blue Iris? If so, disable the start/stop CodeProject.AI within Blue Iris and just let CodeProject.AI run without being stopped.
Are you using a Coral USB or PCIe unit? USB on Windows is really unstable. Seth has done a ton of digging and feels there are memory issues with the Coral library. My personal experience is it works till it doesn't then it's a matter of shutting down the module, switching to a different USB port, restarting, then it works. No issues on Linux, though, which I assume was a focus of the Coral dev team.
cheers
Chris Maunder
|
|
|
|
|
Thanks Chris!
Yes, I'm running on Windows with BI, I felt that it would be speedier rather than having CPAI run on a different host.
I've made the adjustment in BI, ie unchecked the start/stop
I'm running the USB version of Coral on Windows, yes, I've been doing that , shutting down and switching to a different USB port. It use to work great, until v2 of the module. After, the last question about the Coral TPU on here!
|
|
|
|
|
As far as I know there are no memory problems with the Coral module in the latest code. When I ran some longer tests, the memory consumption grew, but leveled off at around 1.1 gb of usage. This points me to Python memory fragmentation, which isn’t really a problem or something I can fix. (AFAIK)
Unfortunately, I don’t have a USB module so I can’t debug anything there. It may be worthwhile for me to get one just for that reason…
|
|
|
|
|
I'm having the exact same problem with the USB Corel. What I have observed, and this might be how it works, but when I installed Corel USB, the device shows under Device Manager as "Corel USB Accelerator". When you start CPAI, it then changes to "UsbDk" and shows no driver. Then when it's running after a reboot, it picks up the adapter in Multi-TPU mode, and after a while will start displaying shutdown messages because of no work to do. Don't know the relevance to this problem, but just to shed some light on what I have observed. To note I'm running on a Beelink SER5 Max mini-PC with a Ryzen 7 5800H APU, so no discreet Graphics Card.
09:21:51:System: Windows
09:21:51:Operating System: Windows (Microsoft Windows 11 version 10.0.22631)
09:21:51:CPUs: AMD Ryzen 7 5800H with Radeon Graphics (AMD)
09:21:51: 1 CPU x 8 cores. 16 logical processors (x64)
09:21:51:GPU (Primary): AMD Radeon(TM) Graphics (3 GiB) (Advanced Micro Devices, Inc.)
09:21:51: Driver: 31.0.21910.5
09:21:51:System RAM: 29 GiB
09:21:51:Platform: Windows
09:21:51:BuildConfig: Release
09:21:51:Execution Env: Native
09:21:51:Runtime Env: Production
09:21:51:Runtimes installed:
09:21:51: .NET runtime: 7.0.10
09:21:51: .NET SDK: Not found
09:21:51: Default Python: Not found
09:21:51: Go: Not found
09:21:51: NodeJS: Not found
09:21:51: Rust: Not found
09:21:51:App DataDir: C:\ProgramData\CodeProject\AI
09:21:51:Video adapter info:
09:21:51: AMD Radeon(TM) Graphics:
09:21:51: Driver Version 31.0.21910.5
09:21:51: Video Processor AMD Radeon Graphics Processor (0x1638)
09:21:51: Microsoft Remote Display Adapter:
09:21:51: Driver Version 10.0.22621.3235
09:21:51: Video Processor
09:21:51:STARTING CODEPROJECT.AI SERVER
09:21:51:RUNTIMES_PATH = C:\Program Files\CodeProject\AI\runtimes
09:21:51:PREINSTALLED_MODULES_PATH = C:\Program Files\CodeProject\AI\preinstalled-modules
09:21:51:DEMO_MODULES_PATH = C:\Program Files\CodeProject\AI\demos\modules
09:21:51:MODULES_PATH = C:\Program Files\CodeProject\AI\modules
09:21:51:PYTHON_PATH = \bin\windows\%PYTHON_NAME%\venv\Scripts\python
09:21:51:Data Dir = C:\ProgramData\CodeProject\AI
09:21:52:Server version: 2.6.2
09:21:55:
09:21:55:Module 'Object Detection (Coral)' 2.2.2 (ID: ObjectDetectionCoral)
09:21:55:Valid: True
09:21:55:Module Path: <root>\modules\ObjectDetectionCoral
09:21:55:AutoStart: True
09:21:55:Queue: objectdetection_queue
09:21:55:Runtime: python3.9
09:21:55:Runtime Loc: Local
09:21:55:FilePath: objectdetection_coral_adapter.py
09:21:55:Start pause: 1 sec
09:21:55:Parallelism: 16
09:21:55:LogVerbosity:
09:21:55:Platforms: all
09:21:55:GPU Libraries: installed if available
09:21:55:GPU Enabled: enabled
09:21:55:Accelerator:
09:21:55:Half Precis.: enable
09:21:55:Environment Variables
09:21:55:CPAI_CORAL_MODEL_NAME = YOLOv8
09:21:55:CPAI_CORAL_MULTI_TPU = True
09:21:55:MODELS_DIR = <root>\modules\ObjectDetectionCoral\assets
09:21:55:MODEL_SIZE = Small
09:21:55:
09:21:55:Started Object Detection (Coral) module
09:21:57:Server: This is the latest version
09:22:02:objectdetection_coral_adapter.py: TPU detected
09:22:02:objectdetection_coral_adapter.py: Attempting multi-TPU initialisation
09:22:02:objectdetection_coral_adapter.py: Supporting multiple Edge TPUs
09:22:58:objectdetection_coral_adapter.py: WARNING:root:No work in 60.0 seconds, watchdog shutting down TPUs.
09:23:04:objectdetection_coral_adapter.py: WARNING:root:No work in 60.0 seconds, watchdog shutting down TPUs.
09:23:09:objectdetection_coral_adapter.py: WARNING:root:No work in 60.0 seconds, watchdog shutting down TPUs.
09:23:14:objectdetection_coral_adapter.py: WARNING:root:No work in 60.0 seconds, watchdog shutting down TPUs.
09:23:19:objectdetection_coral_adapter.py: WARNING:root:No work in 60.0 seconds, watchdog shutting down TPUs.
|
|
|
|
|
The shutdown messages are only a problem if there is, in fact, work to do and it’s not getting done. It’s designed to start back up quickly on demand.
|
|
|
|
|
And that is fair, just was a WARNING type message, so thought it needed some explanation. Any thoughts on the device manager changes? Is that considered normal do you know? If not, maybe I could dig a big more.
I'll try disabling the auto/start as was suggested.
|
|
|
|
|
Yeah, the more recent code no longer has that as a warning. I’m not sure when it will roll out however. I’m afraid I’m not going to be much help for what to expect with device manager.
|
|
|
|
|
Just wanted to report back on this after a few days. I unchecked CPAI starting with BI as recommended, and it's remained stable so far. Currently still showing TPU (TF-Lite). Not sure if the OP is seeing that same thing?
I did see a number of log entries saying:
07:36:05:objectdetection_coral_adapter.py: no TPU interpreters: Falling back to CPU detection
But I think this is expected behavior when there is no work and it finally does go idle. As @mailseth2 eluded to previously, it restarts again. I'll report back if I see anything else.
Thanks everyone!
|
|
|
|
|
Good to hear it’s working. I wouldn’t expect it to ever fall back to CPU. Weird that it says that there are no TPUs available. I’ve tried to build it so it is easy to catch errors and restart if needed, but my TPU setup has been too reliable to actually test and build it in ‘real world’ conditions.
|
|
|
|
|
Well if you need a Windows/Corel USB tester, then please let me know!
|
|
|
|
|
If you want to write the code to get it to work (and keep it working), I’d be happy to incorporate it! There shouldn’t be too much trickery. Just some try/except statements to handle failures better and a lot of trial and error to make sure it’s right. I can write fast code, but am not the greatest at getting it running on your computer…
|
|
|
|
|
Server version: 2.6.2
System: Windows
Operating System: Windows (Microsoft Windows 11 version 10.0.22631)
20:58:09:Preparing to install module 'FaceProcessing'
20:58:09:Downloading module 'FaceProcessing'
20:58:09:Unable to download module 'FaceProcessing' from https://www.codeproject.com/KB/articles/5348853/FaceProcessing-1.10.2.zip. Error: No bytes downloaded
modified 23-Apr-24 12:35pm.
|
|
|
|
|
We missed deploying that file.
I've fixed it and you should be good to go in about 5 minutes.
"Mistakes are prevented by Experience. Experience is gained by making mistakes."
|
|
|
|
|
|
Installing the Coral Module fails with:
Unable to download module 'ObjectDetectionCoral' from https://www.codeproject.com/KB/articles/5348853/ObjectDetectionCoral-2.2.2.zip. Error: No bytes downloaded
Looks like if you change the version to 2.2.0 you can manually download it.
Server version: 2.6.2
System: Windows
Operating System: Windows (Microsoft Windows 10.0.19045)
CPUs: Intel(R) Core(TM) i7-6700 CPU @ 3.40GHz (Intel)
1 CPU x 4 cores. 8 logical processors (x64)
GPU (Primary): Microsoft Remote Display Adapter (Microsoft)
Driver: 10.0.19041.3636
System RAM: 32 GiB
Platform: Windows
BuildConfig: Release
Execution Env: Native
Runtime Env: Production
Runtimes installed:
.NET runtime: 7.0.5
.NET SDK: Not found
Default Python: Not found
Go: Not found
NodeJS: Not found
Rust: Not found
Video adapter info:
Microsoft Remote Display Adapter:
Driver Version 10.0.19041.3636
Video Processor
Intel(R) HD Graphics 530:
Driver Version 27.20.100.9664
Video Processor Intel(R) HD Graphics Family
System GPU info:
GPU 3D Usage 2%
GPU RAM Usage 0
Global Environment variables:
CPAI_APPROOTPATH = <root>
CPAI_PORT = 32168
modified 23-Apr-24 12:35pm.
|
|
|
|
|
Came here looking for solutions to this download issue.
I've managed to get the v2.2.0.zip to download, but I'm unsure of the installation methods. Any hints you can point me to?
Cheers, -Ian
|
|
|
|