|
Currently I'm using "platerecognizer". But BI doesn't deal with the results since a couple of months now. The only result is "DayPlate" but although the text is returned correctly, it fails to cancel the alerts for my own car. In a former version this worked too.
Now I want to test this CPAI solution which is much better, especially because I don't like to send such data into cloud storages.
I have cuDNN installer downloaded. Will test it later.
Here is the system info because of the GPU:
|
|
|
|
|
With your GPU I recommend using CUDA 11.8 not CUDA 12.0
|
|
|
|
|
I got it to run with cuDNN installation. But all sample plates were read with wrong results.
Unfortunately currently not reliable to detect at least German license plates with my 5 MP cameras.
"platerecognizer.com" can read it, but here BI fails with "cancel alert" rule as mentioned above. Can't verify if LPR would work if the results of test images aren't correct.
|
|
|
|
|
Try uninstalling the License Plate Reader module then reinstall.
|
|
|
|
|
The installation was ok, any errors in log about cuDNN or CUDA. It worked, but the results were wrong. "Badges" from admission office, "blanks" and "-" are detected as 8, ...
And the samples are really good from my 5 MP cameras. I also enabled OCR improvement, no changes.
It seems not to work with German license plates.
Send you PM with samples.
|
|
|
|
|
Got a new error message in latest version, one I don't recall seeing before. Not crashing level of problem but thought I would share...
08:19:37:Object Detection (YOLOv5 6.2): [RuntimeError] : Traceback (most recent call last):
File "C:\Program Files\CodeProject\AI\modules\ObjectDetectionYOLOv5-6.2\detect.py", line 141, in do_detection
det = detector(img, size=640)
File "C:\Program Files\CodeProject\AI\runtimes\bin\windows\python37\venv\lib\site-packages\torch\nn\modules\module.py", line 1190, in _call_impl
return forward_call(*input, **kwargs)
File "C:\Program Files\CodeProject\AI\runtimes\bin\windows\python37\venv\lib\site-packages\torch\autograd\grad_mode.py", line 27, in decorate_context
return func(*args, **kwargs)
File "C:\Program Files\CodeProject\AI\runtimes\bin\windows\python37\venv\lib\site-packages\yolov5\models\common.py", line 705, in forward
y = self.model(x, augment=augment) # forward
File "C:\Program Files\CodeProject\AI\runtimes\bin\windows\python37\venv\lib\site-packages\torch\nn\modules\module.py", line 1190, in _call_impl
return forward_call(*input, **kwargs)
File "C:\Program Files\CodeProject\AI\runtimes\bin\windows\python37\venv\lib\site-packages\yolov5\models\common.py", line 515, in forward
y = self.model(im, augment=augment, visualize=visualize) if augment or visualize else self.model(im)
File "C:\Program Files\CodeProject\AI\runtimes\bin\windows\python37\venv\lib\site-packages\torch\nn\modules\module.py", line 1190, in _call_impl
return forward_call(*input, **kwargs)
File "C:\Program Files\CodeProject\AI\runtimes\bin\windows\python37\venv\Lib\site-packages\yolov5\models\yolo.py", line 209, in forward
return self._forward_once(x, profile, visualize) # single-scale inference, train
File "C:\Program Files\CodeProject\AI\runtimes\bin\windows\python37\venv\Lib\site-packages\yolov5\models\yolo.py", line 121, in _forward_once
x = m(x) # run
File "C:\Program Files\CodeProject\AI\runtimes\bin\windows\python37\venv\lib\site-packages\torch\nn\modules\module.py", line 1190, in _call_impl
return forward_call(*input, **kwargs)
File "C:\Program Files\CodeProject\AI\runtimes\bin\windows\python37\venv\Lib\site-packages\yolov5\models\yolo.py", line 75, in forward
wh = (wh * 2) ** 2 * self.anchor_grid[i] # wh
RuntimeError: The size of tensor a (60) must match the size of tensor b (56) at non-singleton dimension 2
modified 5-Mar-24 11:42am.
|
|
|
|
|
Thanks very much for your report. Could you please share your System Info tab from your CodeProject.AI Server dashboard?
Thanks,
Sean Ewington
CodeProject
|
|
|
|
|
What size model are you using, and how much RAM is your GPU using? (I'm wondering if this is a memory issue)
cheers
Chris Maunder
|
|
|
|
|
Hello,
i am running a docker instance of CPAI. The plain docker container containes OCoral 2.1.3. The update works perfect, but after a reboot of the container the version switched back to 2.1.3. Does anybody else have that issue?
Server version: 2.5.4
System: Docker
Operating System: Linux (Ubuntu 22.04)
CPUs: 1 CPU. (Arm64)
System RAM: 8 GiB
Platform: RaspberryPi
BuildConfig: Release
Execution Env: Docker
Runtime Env: Production
.NET framework: .NET 7.0.16
Default Python: 3.10
Go Version:
Video adapter info:
System GPU info:
GPU 3D Usage 0%
GPU RAM Usage 0
Global Environment variables:
CPAI_APPROOTPATH = <root>
CPAI_PORT = 32168
My stack file:
version: "3.8"
services:
ai-server:
image: codeproject/ai-server:rpi64-2.5.4
container_name: CodeprojectAI
privileged: true
ports:
- "32168:32168"
volumes:
- /dev/bus/usb:/dev/bus/usb
- /data/compose/codeprojectai/data:/app/data
- /data/compose/codeprojectai/ai:/etc/codeproject/ai
- /data/compose/codeprojectai/modules:/app/modules
Thanks
Tbs
modified 6-Mar-24 11:09am.
|
|
|
|
|
Thanks for the report. Does this happen every time you reboot the container? Does it happen to other modules?
Thanks,
Sean Ewington
CodeProject
|
|
|
|
|
Hello Sean,
it happens after each reboot - I didn't saw this behaviour to other modules so far.
|
|
|
|
|
Try
version: "3.8"
services:
ai-server:
image: codeproject/ai-server:rpi64-2.5.4
container_name: CodeprojectAI
privileged: true
ports:
- "32168:32168"
volumes:
- /dev/bus/usb:/dev/bus/usb
- codeproject_ai_data_gpu:/etc/codeproject/ai
- codeproject_ai_modules_gpu:/app/modules
volumes:
codeproject_ai_data_gpu:
codeproject_ai_modules_gpu:
cheers
Chris Maunder
modified 14-Mar-24 12:18pm.
|
|
|
|
|
Hey Chris,
thanks for that idea - but unfortunately still same issue. After a reboot it switched back to 2.1.3 version of Coral.
Thank you
Tbs
modified 14-Mar-24 12:18pm.
|
|
|
|
|
Hi all,
A vast, big THANK YOU to those who made Codeproject.ai work with Debian!
I've got version 2.5.4 installed, however I cannot install any module, I keep getting "Call Failed" every time I click on "Install" for any module. Nothing gets downloaded, so I can't manually install the module.
What am I doing wrong? Any advice would be greatly appreciated. Thank you!
System Info:
Server version: 2.5.4
System: Linux
Operating System: Linux (Debian GNU/Linux 12)
CPUs: Intel(R) Xeon(R) CPU E3-1220 v5 @ 3.00GHz (Intel)
1 CPU x 4 cores. 4 logical processors (x64)
GPU (Primary): Tesla P4 (8 GiB) (NVIDIA)
Driver: 550.54.14, CUDA: 12.4 (up to: 12.4), Compute: 6.1, cuDNN:
System RAM: 8 GiB
Platform: Linux
BuildConfig: Release
Execution Env: Native (SSH)
Runtime Env: Production
.NET framework: .NET 7.0.15
Default Python: 3.11
Go Version:
Video adapter info:
Matrox Electronics Systems Ltd. G200eR2 (rev 01):
Driver Version
Video Processor
System GPU info:
GPU 3D Usage 3%
GPU RAM Usage 1.5 GiB
Global Environment variables:
CPAI_APPROOTPATH = <root>
CPAI_PORT = 32168
Logs:
9:51:08:System: Linux
09:51:08:Operating System: Linux (Debian GNU/Linux 12)
09:51:08:CPUs: Intel(R) Xeon(R) CPU E3-1220 v5 @ 3.00GHz (Intel)
09:51:08: 1 CPU x 4 cores. 4 logical processors (x64)
09:51:08:GPU (Primary): Tesla P4 (8 GiB) (NVIDIA)
09:51:08: Driver: 550.54.14, CUDA: 12.4 (up to: 12.4), Compute: 6.1, cuDNN:
09:51:08:System RAM: 8 GiB
09:51:08:Platform: Linux
09:51:08:BuildConfig: Release
09:51:08:Execution Env: Native (SSH)
09:51:08:Runtime Env: Production
09:51:08:.NET framework: .NET 7.0.15
09:51:08:Default Python: 3.11
09:51:08:Go Version:
09:51:08:App DataDir: /etc/codeproject/ai
09:51:08:Video adapter info:
09:51:08: Matrox Electronics Systems Ltd. G200eR2 (rev 01):
09:51:08: Driver Version
09:51:08: Video Processor
09:51:08:STARTING CODEPROJECT.AI SERVER
09:51:08:RUNTIMES_PATH = /usr/bin/codeproject.ai-server-2.5.4/runtimes
09:51:08:PREINSTALLED_MODULES_PATH = /usr/bin/codeproject.ai-server-2.5.4/preinstalled-modules
09:51:08:MODULES_PATH = /usr/bin/codeproject.ai-server-2.5.4/modules
09:51:08:PYTHON_PATH = /bin/linux/%PYTHON_NAME%/venv/bin/python3
09:51:08:Data Dir = /etc/codeproject/ai
09:51:08:Server version: 2.5.4
09:51:08:ModuleRunner Start
09:51:08:Starting Background AI Modules
09:51:14:Current Version is 2.5.4
09:51:14:A new version 2.5.6 is available
10:03:07:API server is offline (NetworkError when attempting to fetch resource.)
10:03:49:API server is offline (NetworkError when attempting to fetch resource.)
10:03:54:API server is offline (NetworkError when attempting to fetch resource.)
modified 4-Mar-24 15:48pm.
|
|
|
|
|
Thanks very much for the report. Once we have it available, we'd like you to try 2.5.6, which may resolve this issue.
Thanks,
Sean Ewington
CodeProject
|
|
|
|
|
Just reinstalled the TrainingYoloV5 from scratch and it is failing to start...
10:06:57:Preparing to install module 'TrainingObjectDetectionYOLOv5'
10:06:57:Downloading module 'TrainingObjectDetectionYOLOv5'
10:06:59:Installing module 'TrainingObjectDetectionYOLOv5'
10:06:59:TrainingObjectDetectionYOLOv5: Setting verbosity to quiet
10:06:59:TrainingObjectDetectionYOLOv5: Hi Docker! We will disable shared python installs for downloaded modules
10:06:59:TrainingObjectDetectionYOLOv5: No schemas installed
10:06:59:TrainingObjectDetectionYOLOv5: (No schemas means: we can't detect if you're in light or dark mode)
10:06:59:TrainingObjectDetectionYOLOv5: Installing CodeProject.AI Analysis Module
10:06:59:TrainingObjectDetectionYOLOv5: ======================================================================
10:06:59:TrainingObjectDetectionYOLOv5: CodeProject.AI Installer
10:06:59:TrainingObjectDetectionYOLOv5: ======================================================================
10:06:59:TrainingObjectDetectionYOLOv5: 5.01 TiB of 8.00 TiB available on Docker
10:06:59:TrainingObjectDetectionYOLOv5: Installing xz-utils...
10:06:59:TrainingObjectDetectionYOLOv5: WARNING: apt does not have a stable CLI interface. Use with caution in scripts.
10:06:59:TrainingObjectDetectionYOLOv5: WARNING: apt does not have a stable CLI interface. Use with caution in scripts.
10:07:06:TrainingObjectDetectionYOLOv5: General CodeProject.AI setup
10:07:06:TrainingObjectDetectionYOLOv5: Setting permissions on downloads folder...Done
10:07:06:TrainingObjectDetectionYOLOv5: Setting permissions on runtimes folder...Done
10:07:06:TrainingObjectDetectionYOLOv5: Setting permissions on persisted data folder...Done
10:07:06:TrainingObjectDetectionYOLOv5: GPU support
10:07:06:TrainingObjectDetectionYOLOv5: CUDA (NVIDIA) Present: No
10:07:19:TrainingObjectDetectionYOLOv5: ROCm (AMD) Present: (attempt to install rocminfo... ) No
10:07:19:TrainingObjectDetectionYOLOv5: MPS (Apple) Present: No
10:07:20:TrainingObjectDetectionYOLOv5: Reading module settings.......Done
10:07:20:TrainingObjectDetectionYOLOv5: Processing module TrainingObjectDetectionYOLOv5 1.6.3
10:07:20:TrainingObjectDetectionYOLOv5: Installing Python 3.8
10:07:20:TrainingObjectDetectionYOLOv5: Python 3.8 is already installed
10:07:39:TrainingObjectDetectionYOLOv5: Ensuring PIP in base python install... done
10:07:47:TrainingObjectDetectionYOLOv5: Upgrading PIP in base python install... done
10:07:47:TrainingObjectDetectionYOLOv5: Virtual Environment already present
10:07:47:TrainingObjectDetectionYOLOv5: Checking for Python 3.8...(Found Python 3.8.18) All good
10:07:57:TrainingObjectDetectionYOLOv5: Upgrading PIP in virtual environment... done
10:08:07:TrainingObjectDetectionYOLOv5: Installing updated setuptools in venv... Done
10:08:11:TrainingObjectDetectionYOLOv5: Searching for libcurl4...All good.
10:08:11:TrainingObjectDetectionYOLOv5: deb http://security.ubuntu.com/ubuntu focal-security main
10:08:31:TrainingObjectDetectionYOLOv5: Searching for libssl1.1...installing... Done
10:08:46:TrainingObjectDetectionYOLOv5: Downloading Standard YOLO models...Expanding... Done.
10:08:47:TrainingObjectDetectionYOLOv5: Moving contents of models-yolo5-pt.zip to assets...done.
10:08:47:TrainingObjectDetectionYOLOv5: Installing Python packages for Training for YoloV5 6.2
10:08:47:TrainingObjectDetectionYOLOv5: Installing GPU-enabled libraries: If available
10:08:51:TrainingObjectDetectionYOLOv5: Searching for python3-pip...All good.
10:09:03:TrainingObjectDetectionYOLOv5: Ensuring PIP compatibility... Done
10:09:03:TrainingObjectDetectionYOLOv5: Python packages will be specified by requirements.txt
10:09:10:TrainingObjectDetectionYOLOv5: - Installing urllib3, the HTTP client for Python...Already installed
10:09:18:TrainingObjectDetectionYOLOv5: - Installing matplotlib, the python plotting package...Already installed
10:09:25:TrainingObjectDetectionYOLOv5: - Installing NumPy, a package for scientific computing...Already installed
10:09:33:TrainingObjectDetectionYOLOv5: - Installing OpenCV, the Open source Computer Vision library...Already installed
10:09:40:TrainingObjectDetectionYOLOv5: - Installing Pillow, a Python Image Library...Already installed
10:09:48:TrainingObjectDetectionYOLOv5: - Installing PyYAML, a library for reading configuration files...Already installed
10:09:55:TrainingObjectDetectionYOLOv5: - Installing request, the HTTP library...Already installed
10:10:03:TrainingObjectDetectionYOLOv5: - Installing SciPy, a library for mathematics, science, and engineering...Already installed
10:10:10:TrainingObjectDetectionYOLOv5: - Installing Torch, for Tensor computation and Deep neural networks...Already installed
10:10:18:TrainingObjectDetectionYOLOv5: - Installing TorchVision, for Computer Vision based AI...Already installed
10:10:25:TrainingObjectDetectionYOLOv5: - Installing tdqm, the Fast, Extensible Progress Meter...Already installed
10:10:33:TrainingObjectDetectionYOLOv5: - Installing protobuf, extensible mechanisms for serializing structured data...Already installed
10:10:40:TrainingObjectDetectionYOLOv5: - Installing tensorboard, a tool to let you watch Tensors Flow...Already installed
10:10:48:TrainingObjectDetectionYOLOv5: - Installing Pandas, a data analysis / data manipulation tool...Already installed
10:10:55:TrainingObjectDetectionYOLOv5: - Installing Seaborn, a data visualization library based on matplotlib...Already installed
10:11:03:TrainingObjectDetectionYOLOv5: - Installing ipython, for interactive notebooks...Already installed
10:11:10:TrainingObjectDetectionYOLOv5: - Installing psutil, a tool to check system utilization...Already installed
10:11:17:TrainingObjectDetectionYOLOv5: - Installing thop, a tool to count the FLOPs of PyTorch model....Already installed
10:11:25:TrainingObjectDetectionYOLOv5: - Installing Ultralytics YoloV5 package for object detection in images...Already installed
10:11:32:TrainingObjectDetectionYOLOv5: - Installing fiftyone, for building datasets and computer vision models...Already installed
10:11:32:TrainingObjectDetectionYOLOv5: Installing Python packages for the CodeProject.AI Server SDK
10:11:37:TrainingObjectDetectionYOLOv5: Searching for python3-pip...All good.
10:11:48:TrainingObjectDetectionYOLOv5: Ensuring PIP compatibility... Done
10:11:48:TrainingObjectDetectionYOLOv5: Python packages will be specified by requirements.txt
10:11:56:TrainingObjectDetectionYOLOv5: - Installing Pillow, a Python Image Library...Already installed
10:12:03:TrainingObjectDetectionYOLOv5: - Installing Charset normalizer...Already installed
10:12:10:TrainingObjectDetectionYOLOv5: - Installing aiohttp, the Async IO HTTP library...Already installed
10:12:18:TrainingObjectDetectionYOLOv5: - Installing aiofiles, the Async IO Files library...Already installed
10:12:25:TrainingObjectDetectionYOLOv5: - Installing py-cpuinfo to allow us to query CPU info...Already installed
10:12:32:TrainingObjectDetectionYOLOv5: - Installing Requests, the HTTP library...Already installed
10:12:53:TrainingObjectDetectionYOLOv5: Self test: Self-test passed
10:12:53:TrainingObjectDetectionYOLOv5: Module setup time 00:05:34
10:12:53:TrainingObjectDetectionYOLOv5: Setup complete
10:12:53:TrainingObjectDetectionYOLOv5: Total setup time 00:05:54
10:12:53:Module TrainingObjectDetectionYOLOv5 installed successfully.
10:12:53:Installer exited with code 0
10:12:53:Module TrainingObjectDetectionYOLOv5 not configured to AutoStart.
10:13:19:Update TrainingObjectDetectionYOLOv5. Setting AutoStart=true
10:13:19:Restarting Training for YoloV5 6.2 to apply settings change
10:13:19:
10:13:19:Module 'Training for YoloV5 6.2' 1.6.3 (ID: TrainingObjectDetectionYOLOv5)
10:13:19:Valid: True
10:13:19:Module Path: <root>/modules/TrainingObjectDetectionYOLOv5
10:13:19:AutoStart: True
10:13:19:Queue: trainingobjectdetectionyolov5_queue
10:13:19:Runtime: python3.8
10:13:19:Runtime Loc: Local
10:13:19:FilePath: training_objectdetection_YOLOv5_adapter.py
10:13:19:Pre installed: False
10:13:19:Start pause: 1 sec
10:13:19:Parallelism: 2
10:13:19:LogVerbosity:
10:13:19:Platforms: all,!raspberrypi,!orangepi,!jetson
10:13:19:GPU Libraries: installed if available
10:13:19:GPU Enabled: enabled
10:13:19:Accelerator:
10:13:19:Half Precis.: enable
10:13:19:Environment Variables
10:13:19:FIFTYONE_DATABASE_DIRNAME = fiftyone
10:13:19:YOLOv5_AUTOINSTALL = false
10:13:19:YOLOv5_VERBOSE = false
10:13:19:YOLO_DATASETS_DIRNAME = datasets
10:13:19:YOLO_DATASET_ZOO_DIRNAME = zoo
10:13:19:YOLO_MODELS_DIRNAME = assets
10:13:19:YOLO_TRAINING_DIRNAME = train
10:13:19:YOLO_WEIGHTS_DIRNAME = weights
10:13:19:
10:13:19:Started Training for YoloV5 6.2 module
10:13:46:training_objectdetection_YOLOv5_adapter.py: Subprocess ['/app/modules/TrainingObjectDetectionYOLOv5/bin/linux/python38/venv/lib/python3.8/site-packages/fiftyone/db/bin/mongod', '--dbpath', '/app/modules/TrainingObjectDetectionYOLOv5/fiftyone', '--logpath', '/app/modules/TrainingObjectDetectionYOLOv5/fiftyone/log/mongo.log', '--port', '0', '--nounixsocket'] exited with error -4:
10:14:34:detect_adapter.py: YOLOv5.1m summary: 391 layers, 21805053 parameters, 0 gradients
10:14:34:detect_adapter.py: Adding AutoShape...
10:14:39:training_objectdetection_YOLOv5_adapter.py: Unable to import and initialise the fiftyone.zoo package: fiftyone.core.service.DatabaseService failed to bind to port
10:14:42:Module TrainingObjectDetectionYOLOv5 has shutdown
10:14:42:training_objectdetection_YOLOv5_adapter.py: has exited
CPU-info:
Server version: 2.5.6
System: Docker
Operating System: Linux (Ubuntu 22.04)
CPUs: Intel(R) Atom(TM) CPU C2538 @ 2.40GHz (Intel)
1 CPU x 4 cores. 4 logical processors (x64)
System RAM: 8 GiB
Platform: Linux
BuildConfig: Release
Execution Env: Docker
Runtime Env: Production
Runtimes installed:
.NET runtime: 7.0.16
.NET SDK: Not found
Default Python: Not found
Go: Not found
NodeJS: Not found
Video adapter info:
System GPU info:
GPU 3D Usage 0%
GPU RAM Usage 0
Global Environment variables:
CPAI_APPROOTPATH = <root>
CPAI_PORT = 32168
Any recommendation on how to fix this ?
modified 5-Mar-24 11:57am.
|
|
|
|
|
Thanks very much for your report. Have you mapped your folders for your install?
Thanks,
Sean Ewington
CodeProject
|
|
|
|
|
Hi,
i got 2 folders mounted through my docker compose:
volumes:
- /volume1/docker/senseAI/config:/etc/codeproject/ai
- /volume1/docker/senseAI/modules:/app/modules
the module was installed through the GUI so I'm unable to specify any arguments.
Any recommendation if other folders need to be mounted ?
Based on the error returned, issue looks to be related to embedded MongoDB in FiftyOne module.
|
|
|
|
|
Sometimes the MongoDb package does not initialize correctly.
Try restarting the module.
"Mistakes are prevented by Experience. Experience is gained by making mistakes."
|
|
|
|
|
Already done that several times... also reinstalled 3 times already... always same error...
|
|
|
|
|
Hello.
Tried to update my CP (2.2.4 ver) running on Docker on RPI5 to 2.5.6 however 2.5.6 wont start, error log is:
/bin/bash: line 1: pstree: command not found
Unhandled exception. System.Collections.Generic.KeyNotFoundException: The given key 'version' was not present in the dictionary.
at System.Collections.Generic.Dictionary`2.get_Item(TKey key)
at CodeProject.AI.SDK.Common.SystemInfo.GetRuntimesAsync()
at CodeProject.AI.SDK.Common.SystemInfo.InitializeAsync()
at CodeProject.AI.Server.Program.Main(String[] args)
at CodeProject.AI.Server.Program.<Main>(String[] args)
Any hints how to fix this ?
modified 14-Mar-24 12:22pm.
|
|
|
|
|
|
Can I ask you: which of the following do you have installed on your system (that you're aware of):
.NET 7
Python
Go
nodejs
Rust
Thanks,
Sean Ewington
CodeProject
modified 4-Mar-24 16:13pm.
|
|
|
|
|
I have installed:
- Raspbian 12 bookworm (Linux raspberryzero52 6.1.0-rpi8-rpi-2712 #1 SMP PREEMPT Debian 1:6.1.73-1+rpt1 (2024-01-25) aarch64 GNU/Linux)
- Docker version 25.0.3, build 4debf41
Nothing else (except some containers).
|
|
|
|
|
Which Docker image are you running?
"Mistakes are prevented by Experience. Experience is gained by making mistakes."
|
|
|
|
|