They just release an arm64 version yesterday which I think the arm64-2.1.0 should work for you. I am not a Docker expert so I am not 100% sure
CodeProject.AI Version 2.0
- Thread starter MikeLud1
- Start date
Not sure if it was just me playing with settings but when I first switched I was having issues too. I guess I finally got my DS where I wanted. I have a flag out front and umbrellas out back that move with wind and was getting many different identities, flag was bird, dog, person, etc. Umbrellas were persons. But remember that when i first set up DS too. A few weeks later, It seems CP.AI is doing pretty good. I have not tried any custom modeling yet. Wanted to get my camera's AI alerts right first. I have went back on some footage to see if any triggers were missed and so far none.
I will say I had a few times where AI just stopped too, but I am one to always update BI so I am blaming myself for not always rebooting, because it seems to work once I do a full reboot of my BI PC, but that has only happened a couple of times.
If you need any BI screens let me know.
Yea I don't think that will work as it is for an ARM 64-bit Processor not a x86 (Intel/AMD) processor
if you simple want to try it out or do some benchmarks, there are multiple images out there with Tensorflow and a BI compatible API =>
My CPAI seems to be dead in the water and I don't know why. It's been installed and running very successfully for several months. The environment is Windows 10, BI 5.7.4.2 and CPAI 2.0.8.0. Starting 2 days ago, CPAI didn't respond and today when I pulled up localhost:32168, it did not respond. I went to the SERVICES app and tried a manual start of the CPAI service and got an error 1053 that the service did not respond in a timely manner. I tried a REPAIR on the installation but no change.
Since then, I've uninstalled it, rebooted, re-installed v2.0.8.0. Now, the service doesn't show at all in the SERVICES app. I'd be grateful for any hints on where to go next.
Since then, I've uninstalled it, rebooted, re-installed v2.0.8.0. Now, the service doesn't show at all in the SERVICES app. I'd be grateful for any hints on where to go next.
That's so fun as I already go that running on my setup to accept alert image processing from BI. That app's log file is able to successfully detected objects but BI never liked the results and always threw it out. So I gave up once I heard that CPAI was close with the coral support. The big down side of the project when utilized with BI was that BI sends a whole image without bounding boxes of where the motion happened. Why this is an issue is that the model used for the coral only accepts an image size of 640 x 640 pixels. When the python code scales down the image say for a 4K camera you lose all the detail that could be used for a more accurate detection. I had to turn down the confidence to 10-20% to get atleast some level of successful detection. I know that Frigate gets around this by cutting out the bounding boxes within an image of where motion was detected then scaling that down and passing it to the model. This has a lot less loss of detail when running it through the coral model.
Ok to end a long story, I guess I'll just modify the raspberry pi or ARM dockerfile that is posed on DockerHub for the CodeProject AI server and see if I can get it running on an x86-64bit machine.
Try this:
1. Uninstall CP:AI
2. Delete the Codeproject Directory completely - "C:\Program Files\Codeproject"
3. Delete "C:\ProgramData\CodeProject"
4. Restart The Computer
5. Reinstall CodeProject along with the modules that you wish to install.
6. Restart Computer one more time
7. Run BI and CP:AI
I have been doing this since the beginning with this software package and very rarely do I have problems when this is followed as the uninstall/install procedure.
Last edited:
I forgot about that one..... I corrected my original post Thanks Mike
Thanks @kaltertod and @MikeLud1 But, still no joy. All the uninstalls and re-installs seems to go perfectly and the Program Files/CodeProject directory is created again. But it never runs. I see no sign of CPAI in Device Manager or in the Services list. In the Blue Iris AI screen the "Start Now" button (above the CP orange banner) is greyed-out and does not respond.
Is there a log somewhere that I can retrieve?
When successful in the past, i never had to do anything else. But I notice a setup.bat in the CodeProject directory and a CodeProject.AI.Server.exe executable in the Server subdirectory. Would executing these do anything. I have tried starting the latter and a DOS screen with commands appear and disappears but no sign of CPAI actually starting. I've never heard of anyone having issues actually starting the service, especially since it was running before on this system.
I appreciate your help in debugging this. Please send any and all suggestions my way. Thanks!
CodeProject.AI Update: Version 2.1.0 Docker images are released
Does the following thread hold any relevance to your situation: CodeProject.AI Server: AI the easy way.
That report looks identical to my issue. Thanks so much for the tip. I'm away from my BI/CPAI machine tonight but will investigate Saturday. Thanks again.
UPDATE: My symptoms were the same but the solution offered {deleteing the PROGRAMDATA json} does not work for me. In fact, I don't even have a C:\ProgramData\CodeProject directory. It does not appear in the SERVICES list. I've posted in the CP.AI forum since my similar problem did not resolve as others have.
Last edited:
CCTVCam
Known around here
- Sep 25, 2017
- 2,830
- 3,708
Just started to use CPAI to take the load off my cameras. Seems very good so far.
However, I have a couple of questions;
1. I'm cpu only but have loads of power - currently using 5% only! However, I don't seem to have a choice of YOLO models. I appear to be locked to the basic YOLO 5. Is this a feature due to a lack of a gpu?
2. Super resolution - how does it work in BI (I've just noticed a zoom icon when I click over live view although it doesn't zoom), and what are people's experiences with the current build? Does it improve resolution / quality?
However, I have a couple of questions;
1. I'm cpu only but have loads of power - currently using 5% only! However, I don't seem to have a choice of YOLO models. I appear to be locked to the basic YOLO 5. Is this a feature due to a lack of a gpu?
2. Super resolution - how does it work in BI (I've just noticed a zoom icon when I click over live view although it doesn't zoom), and what are people's experiences with the current build? Does it improve resolution / quality?
CodeProject.AI Update: Version 2.1.1 Windows Install was just released
V2.1.1
www.codeproject.com
V2.1.1
- Improved Raspberry Pi support. A new, fast object detection module with support for the Coral.AI TPU, all within an Arm64 Docker image
- All modules can now be installed / uninstalled (rather than having some modules fixed and uninstallble).
- Installer is streamlined: Only the server is installed at installation time, and on first run we install Object Detection (Python and .NET) and Face Processing (which can be uninstalled).
- Reworking of the Python module SDK. Modules are new child classes, not aggregators of our module runner.
- Reworking of the modulesettings file to make it simpler and have less replication
- Improved logging: quantity, quality, filtering and better information
- Addition of 2 modules: ObjectDetectionTFLite for Object Detection on a on Raspberry Pi using Coral, and Cartoonise for some fun
- Improvements to half-precision support checks on CUDA cards
- Modules are now versioned and our module registry will now only show modules that fit your current server version.
- Various bug fixes
- Shared Python runtimes now in runtimes.
- All modules moved from the AnalysisLayer folder to the modules folder
- Tested on CUDA 12
CodeProject.AI Server: AI the easy way.
Version 2.6.5. Our fast, free, self-hosted Artificial Intelligence Server for any platform, any language
www.codeproject.com
I just saw your post on the CPAI chat that v2.1.1 only works with CPU. Any idea when it'll support GPU?CodeProject.AI Update: Version 2.1.1 Windows Install was just released
V2.1.1
- Improved Raspberry Pi support. A new, fast object detection module with support for the Coral.AI TPU, all within an Arm64 Docker image
- All modules can now be installed / uninstalled (rather than having some modules fixed and uninstallble).
- Installer is streamlined: Only the server is installed at installation time, and on first run we install Object Detection (Python and .NET) and Face Processing (which can be uninstalled).
- Reworking of the Python module SDK. Modules are new child classes, not aggregators of our module runner.
- Reworking of the modulesettings file to make it simpler and have less replication
- Improved logging: quantity, quality, filtering and better information
- Addition of 2 modules: ObjectDetectionTFLite for Object Detection on a on Raspberry Pi using Coral, and Cartoonise for some fun
- Improvements to half-precision support checks on CUDA cards
- Modules are now versioned and our module registry will now only show modules that fit your current server version.
- Various bug fixes
- Shared Python runtimes now in runtimes.
- All modules moved from the AnalysisLayer folder to the modules folder
- Tested on CUDA 12
CodeProject.AI Server: AI the easy way.
Version 2.6.5. Our fast, free, self-hosted Artificial Intelligence Server for any platform, any language
This is only for the Object Detection (YOLOv5 .NET) module, other modules are working fine. It should be a easy fix for Chris so most likely tomorrow.
I really thought that the Windows update would have coral support for all architecture, not just the raspberry pi.
I see you like to run YOLOv5 .NET instead of the YOLOv5 6.2. Is there a particular benefit of that?