CodeProject.AI Version 2.0

Most of these problems are from NOT letting CP install fully. When you do the install select "remove previously installed modules & data" and then wait for the installer finish the initial setup. BE SURE to leave status window open and watch the progress of each component as it is downloaded and installed. It will take at least 15-20 minutes to install just the 3 basic parts. Patience is a key factor here :)
 
  • Like
Reactions: jrbeddow
prsmith777 said:
Ok I did that and no change
Click to expand...
Also open the below file with notepad and post what is in the file, like the below.

{
"Modules": {
"ObjectDetectionYOLOv5Net": {
"AutoStart": true
},
"ObjectDetectionYOLOv5-6.2": {
"AutoStart": false
},
"FaceProcessing": {
"AutoStart": false
},
"TrainingObjectDetectionYOLOv5": {
"AutoStart": false
},
"ObjectDetectionYOLOv8": {
"AutoStart": false
}
}
}
 
MikeLud1 said:
Also open the below file with notepad and post what is in the file, like the below.

{
"Modules": {
"ObjectDetectionYOLOv5Net": {
"AutoStart": true
},
"ObjectDetectionYOLOv5-6.2": {
"AutoStart": false
},
"FaceProcessing": {
"AutoStart": false
},
"TrainingObjectDetectionYOLOv5": {
"AutoStart": false
},
"ObjectDetectionYOLOv8": {
"AutoStart": false
}
}
}
Click to expand...

Sorry, which file is that?
 
So replacing my modulesettings.json with yours has returned the GPU option.

However, when I test LPR using CP.AI Explorer I get no predictions returned. When I disable GPU and use CPU, I get a good license read.

Here is my version that didn't work.

{
"Modules": {

"ALPR": {
"Name": "License Plate Reader",
"Version": "2.7.5",

/ Publishing info
"Description": "Detects and readers single-line and multi-line licence plates using YOLO object detection and the PaddleOCR toolkit",
"Category": "Computer Vision",
"Platforms": [ "all" ],
"License": "SSPL",
"LicenseUrl": "Server Side Public License (SSPL)",

/ Launch instructions
"AutoStart": true,
"FilePath": "ALPR_adapter.py",
"Runtime": "python3.8",
"RuntimeLocation": "Local", / Can be Local or Shared
"Queue": "alpr_queue", / default is lower(module_id) + '_queue'

"EnvironmentVariables": {
"MIN_COMPUTE_CAPABILITY": "7",
"MIN_CUDNN_VERSION": "7",

"PLATE_CONFIDENCE": 0.7, / Confidence required in detecting a plate in the first place
"PLATE_ROTATE_DEG": 0, / If non-zero, rotate plate before OCR (+ve = counterclockwise)
"AUTO_PLATE_ROTATE": true,
"PLATE_RESCALE_FACTOR": 2,
"OCR_OPTIMIZATION": true,
"OCR_OPTIMAL_CHARACTER_HEIGHT": 60,
"OCR_OPTIMAL_CHARACTER_WIDTH": 36
},

/ GPU options
"InstallGPU": false, / PaddlePaddle-GPU only installs easily on Windows (though with issues on older cards)
"EnableGPU": true, / Will be coerced to false if InstallGPU = false
"AcceleratorDeviceName": null, / = default
"Parallelism": 0, / 0 = Default (number of CPUs - 1)
"HalfPrecision": "enable", / 'Force', 'Enable', 'Disable': whether to force on, allow, or disable half-precision ops
"PostStartPauseSecs": 1, / Generally 1 if using GPU, 0 for CPU


/ Which server version is compatible with each version of this module.
"ModuleReleases": [
{ "ModuleVersion": "1.0", "ServerVersionRange": [ "1.0", "2.0.8" ], "ReleaseDate": "2022-11-01" },
{ "ModuleVersion": "2.1", "ServerVersionRange": [ "2.0.9", "2.0.9" ], "ReleaseDate": "2022-12-01" },
{ "ModuleVersion": "2.2", "ServerVersionRange": [ "2.1", "2.1.12" ], "ReleaseDate": "2023-03-20" },
{ "ModuleVersion": "2.3", "ServerVersionRange": [ "2.1", "2.1.12" ], "ReleaseDate": "2023-04-20", "ReleaseNotes": "Updated module settings", "Importance": "Minor" },
{ "ModuleVersion": "2.4", "ServerVersionRange": [ "2.1", "2.1.12" ], "ReleaseDate": "2023-05-10", "ReleaseNotes": "PaddlePaddle install more reliable", "Importance": "Minor" },
{ "ModuleVersion": "2.5", "ServerVersionRange": [ "2.1", "2.1.12" ], "ReleaseDate": "2023-06-04", "ReleaseNotes": "Updated PaddlePaddle" },
{ "ModuleVersion": "2.6", "ServerVersionRange": [ "2.2", "2.2.4" ], "ReleaseDate": "2023-09-09", "ReleaseNotes": "Updated installer" },
{ "ModuleVersion": "2.7", "ServerVersionRange": [ "2.3.0", "2.3.0" ], "ReleaseDate": "2023-10-01", "ReleaseNotes": "Updated to match new installer SDK. Now works on Raspberry Pi." },
{ "ModuleVersion": "2.7.1", "ServerVersionRange": [ "2.3.1", "2.3.5" ], "ReleaseDate": "2023-10-10", "ReleaseNotes": "Updated to match new installer SDK." },
{ "ModuleVersion": "2.7.2", "ServerVersionRange": [ "2.3.1", "2.4.0" ], "ReleaseDate": "2023-10-10", "ReleaseNotes": "LibSSL install correction." },
{ "ModuleVersion": "2.7.3", "ServerVersionRange": [ "2.3.5", "2.4.0" ], "ReleaseDate": "2023-11-06", "ReleaseNotes": "Installer updates", "Importance": "Minor" },
{ "ModuleVersion": "2.7.4", "ServerVersionRange": [ "2.4.1", "2.4.1" ], "ReleaseDate": "2023-12-06", "ReleaseNotes": "Updated modulesettings schema, new route", "Importance": "Minor" },
{ "ModuleVersion": "2.7.5", "ServerVersionRange": [ "2.4.2", "" ], "ReleaseDate": "2023-12-09", "ReleaseNotes": "Installer updates", "Importance": "Minor" }
],

"RouteMaps": [
{
"Name": "License Plate Reader",
"Route": "vision/alpr",
"Method": "POST",
"Command": "alpr",
"Description": "Detects and reads the characters in license plates detected within an image",
"Inputs": [
{
"Name": "upload",
"Type": "File",
"Description": "The image to ALPR."
}
],

"Outputs": [
{
"Name": "success",
"Type": "Boolean",
"Description": "True if successful."
},
{
"Name": "message",
"Type": "String",
"Description": "A summary of the inference operation."
},
{
"Name": "error",
"Type": "String",
"Description": "(Optional) An description of the error if success was false."
},
{
"Name": "predictions",
"Type": "Object[]",
"Description": "An array of objects with the x_max, x_min, max, y_min bounds of the plate, label, the plate chars and confidence."
},
{
"Name": "count",
"Type": "Integer",
"Description": "The number of objects found."
},
{
"Name": "command",
"Type": "String",
"Description": "The command that was sent as part of this request. Can be detect, list, status."
},
{
"Name": "moduleId",
"Type": "String",
"Description": "The Id of the module that processed this request."
},
{
"Name": "executionProvider",
"Type": "String",
"Description": "The name of the device or package handling the inference. eg CPU, GPU, TPU, DirectML."
},
{
"Name": "canUseGPU",
"Type": "Boolean",
"Description": "True if this module can use the current GPU if one is present."
},
{
"Name": "inferenceMs",
"Type": "Integer",
"Description": "The time (ms) to perform the AI inference."
},
{
"Name": "processMs",
"Type": "Integer",
"Description": "The time (ms) to process the image (includes inference and image manipulation operations)."
},
{
"Name": "analysisRoundTripMs",
"Type": "Integer",
"Description": "The time (ms) for the round trip to the analysis module and back."
}
]
},
{
"Name": "License Plate Reader, Legacy route",
"Route": "image/alpr",
"Method": "POST",
"Command": "alpr",
"Description": "Detects the characters in license plates detected within an image",
"Inputs": [
{
"Name": "upload",
"Type": "File",
"Description": "The image to ALPR."
}
],

"Outputs": [
{
"Name": "success",
"Type": "Boolean",
"Description": "True if successful."
},
{
"Name": "message",
"Type": "String",
"Description": "A summary of the inference operation."
},
{
"Name": "error",
"Type": "String",
"Description": "(Optional) An description of the error if success was false."
},
{
"Name": "predictions",
"Type": "Object[]",
"Description": "An array of objects with the x_max, x_min, max, y_min bounds of the plate, label, the plate chars and confidence."
},
{
"Name": "count",
"Type": "Integer",
"Description": "The number of objects found."
},
{
"Name": "command",
"Type": "String",
"Description": "The command that was sent as part of this request. Can be detect, list, status."
},
{
"Name": "moduleId",
"Type": "String",
"Description": "The Id of the module that processed this request."
},
{
"Name": "executionProvider",
"Type": "String",
"Description": "The name of the device or package handling the inference. eg CPU, GPU, TPU, DirectML."
},
{
"Name": "canUseGPU",
"Type": "Boolean",
"Description": "True if this module can use the current GPU if one is present."
},
{
"Name": "inferenceMs",
"Type": "Integer",
"Description": "The time (ms) to perform the AI inference."
},
{
"Name": "processMs",
"Type": "Integer",
"Description": "The time (ms) to process the image (includes inference and image manipulation operations)."
},
{
"Name": "analysisRoundTripMs",
"Type": "Integer",
"Description": "The time (ms) for the round trip to the analysis module and back."
}
]
}
]
}
}
}
 
MikeLud1 said:
Try using the Object Detection (YOLOv5 .NET) module instead of the Object Detection (YOLOv5 6.2) 1.7.5 module. The Object Detection (YOLOv5 .NET) module will use the iGPU.
Click to expand...
Thanks. I am using the yolov5 .net. However it still shows CPU on the dashboard even after hundreds of triggers. I had this issue with 2.0.8 as well on other systems (2.0.8 otherwise works great). My understanding was that once ai processed its first image with yolov5 .net, the dash would display gpu. Otherwise its was a really smooth install and working perfectly.
Your work on this project along with the others is fantastic and appreciated!
 
  • Like
Reactions: Mad_Marik, David L, hikky_b and 1 other person
prsmith777 said:
So replacing my modulesettings.json with yours has returned the GPU option.

However, when I test LPR using CP.AI Explorer I get no predictions returned. When I disable GPU and use CPU, I get a good license read.

Here is my version that didn't work.

{
"Modules": {

"ALPR": {
"Name": "License Plate Reader",
"Version": "2.7.5",

/ Publishing info
"Description": "Detects and readers single-line and multi-line licence plates using YOLO object detection and the PaddleOCR toolkit",
"Category": "Computer Vision",
"Platforms": [ "all" ],
"License": "SSPL",
"LicenseUrl": "Server Side Public License (SSPL)",

/ Launch instructions
"AutoStart": true,
"FilePath": "ALPR_adapter.py",
"Runtime": "python3.8",
"RuntimeLocation": "Local", / Can be Local or Shared
"Queue": "alpr_queue", / default is lower(module_id) + '_queue'

"EnvironmentVariables": {
"MIN_COMPUTE_CAPABILITY": "7",
"MIN_CUDNN_VERSION": "7",

"PLATE_CONFIDENCE": 0.7, / Confidence required in detecting a plate in the first place
"PLATE_ROTATE_DEG": 0, / If non-zero, rotate plate before OCR (+ve = counterclockwise)
"AUTO_PLATE_ROTATE": true,
"PLATE_RESCALE_FACTOR": 2,
"OCR_OPTIMIZATION": true,
"OCR_OPTIMAL_CHARACTER_HEIGHT": 60,
"OCR_OPTIMAL_CHARACTER_WIDTH": 36
},

/ GPU options
"InstallGPU": false, / PaddlePaddle-GPU only installs easily on Windows (though with issues on older cards)
"EnableGPU": true, / Will be coerced to false if InstallGPU = false
"AcceleratorDeviceName": null, / = default
"Parallelism": 0, / 0 = Default (number of CPUs - 1)
"HalfPrecision": "enable", / 'Force', 'Enable', 'Disable': whether to force on, allow, or disable half-precision ops
"PostStartPauseSecs": 1, / Generally 1 if using GPU, 0 for CPU


/ Which server version is compatible with each version of this module.
"ModuleReleases": [
{ "ModuleVersion": "1.0", "ServerVersionRange": [ "1.0", "2.0.8" ], "ReleaseDate": "2022-11-01" },
{ "ModuleVersion": "2.1", "ServerVersionRange": [ "2.0.9", "2.0.9" ], "ReleaseDate": "2022-12-01" },
{ "ModuleVersion": "2.2", "ServerVersionRange": [ "2.1", "2.1.12" ], "ReleaseDate": "2023-03-20" },
{ "ModuleVersion": "2.3", "ServerVersionRange": [ "2.1", "2.1.12" ], "ReleaseDate": "2023-04-20", "ReleaseNotes": "Updated module settings", "Importance": "Minor" },
{ "ModuleVersion": "2.4", "ServerVersionRange": [ "2.1", "2.1.12" ], "ReleaseDate": "2023-05-10", "ReleaseNotes": "PaddlePaddle install more reliable", "Importance": "Minor" },
{ "ModuleVersion": "2.5", "ServerVersionRange": [ "2.1", "2.1.12" ], "ReleaseDate": "2023-06-04", "ReleaseNotes": "Updated PaddlePaddle" },
{ "ModuleVersion": "2.6", "ServerVersionRange": [ "2.2", "2.2.4" ], "ReleaseDate": "2023-09-09", "ReleaseNotes": "Updated installer" },
{ "ModuleVersion": "2.7", "ServerVersionRange": [ "2.3.0", "2.3.0" ], "ReleaseDate": "2023-10-01", "ReleaseNotes": "Updated to match new installer SDK. Now works on Raspberry Pi." },
{ "ModuleVersion": "2.7.1", "ServerVersionRange": [ "2.3.1", "2.3.5" ], "ReleaseDate": "2023-10-10", "ReleaseNotes": "Updated to match new installer SDK." },
{ "ModuleVersion": "2.7.2", "ServerVersionRange": [ "2.3.1", "2.4.0" ], "ReleaseDate": "2023-10-10", "ReleaseNotes": "LibSSL install correction." },
{ "ModuleVersion": "2.7.3", "ServerVersionRange": [ "2.3.5", "2.4.0" ], "ReleaseDate": "2023-11-06", "ReleaseNotes": "Installer updates", "Importance": "Minor" },
{ "ModuleVersion": "2.7.4", "ServerVersionRange": [ "2.4.1", "2.4.1" ], "ReleaseDate": "2023-12-06", "ReleaseNotes": "Updated modulesettings schema, new route", "Importance": "Minor" },
{ "ModuleVersion": "2.7.5", "ServerVersionRange": [ "2.4.2", "" ], "ReleaseDate": "2023-12-09", "ReleaseNotes": "Installer updates", "Importance": "Minor" }
],

"RouteMaps": [
{
"Name": "License Plate Reader",
"Route": "vision/alpr",
"Method": "POST",
"Command": "alpr",
"Description": "Detects and reads the characters in license plates detected within an image",
"Inputs": [
{
"Name": "upload",
"Type": "File",
"Description": "The image to ALPR."
}
],

"Outputs": [
{
"Name": "success",
"Type": "Boolean",
"Description": "True if successful."
},
{
"Name": "message",
"Type": "String",
"Description": "A summary of the inference operation."
},
{
"Name": "error",
"Type": "String",
"Description": "(Optional) An description of the error if success was false."
},
{
"Name": "predictions",
"Type": "Object[]",
"Description": "An array of objects with the x_max, x_min, max, y_min bounds of the plate, label, the plate chars and confidence."
},
{
"Name": "count",
"Type": "Integer",
"Description": "The number of objects found."
},
{
"Name": "command",
"Type": "String",
"Description": "The command that was sent as part of this request. Can be detect, list, status."
},
{
"Name": "moduleId",
"Type": "String",
"Description": "The Id of the module that processed this request."
},
{
"Name": "executionProvider",
"Type": "String",
"Description": "The name of the device or package handling the inference. eg CPU, GPU, TPU, DirectML."
},
{
"Name": "canUseGPU",
"Type": "Boolean",
"Description": "True if this module can use the current GPU if one is present."
},
{
"Name": "inferenceMs",
"Type": "Integer",
"Description": "The time (ms) to perform the AI inference."
},
{
"Name": "processMs",
"Type": "Integer",
"Description": "The time (ms) to process the image (includes inference and image manipulation operations)."
},
{
"Name": "analysisRoundTripMs",
"Type": "Integer",
"Description": "The time (ms) for the round trip to the analysis module and back."
}
]
},
{
"Name": "License Plate Reader, Legacy route",
"Route": "image/alpr",
"Method": "POST",
"Command": "alpr",
"Description": "Detects the characters in license plates detected within an image",
"Inputs": [
{
"Name": "upload",
"Type": "File",
"Description": "The image to ALPR."
}
],

"Outputs": [
{
"Name": "success",
"Type": "Boolean",
"Description": "True if successful."
},
{
"Name": "message",
"Type": "String",
"Description": "A summary of the inference operation."
},
{
"Name": "error",
"Type": "String",
"Description": "(Optional) An description of the error if success was false."
},
{
"Name": "predictions",
"Type": "Object[]",
"Description": "An array of objects with the x_max, x_min, max, y_min bounds of the plate, label, the plate chars and confidence."
},
{
"Name": "count",
"Type": "Integer",
"Description": "The number of objects found."
},
{
"Name": "command",
"Type": "String",
"Description": "The command that was sent as part of this request. Can be detect, list, status."
},
{
"Name": "moduleId",
"Type": "String",
"Description": "The Id of the module that processed this request."
},
{
"Name": "executionProvider",
"Type": "String",
"Description": "The name of the device or package handling the inference. eg CPU, GPU, TPU, DirectML."
},
{
"Name": "canUseGPU",
"Type": "Boolean",
"Description": "True if this module can use the current GPU if one is present."
},
{
"Name": "inferenceMs",
"Type": "Integer",
"Description": "The time (ms) to perform the AI inference."
},
{
"Name": "processMs",
"Type": "Integer",
"Description": "The time (ms) to process the image (includes inference and image manipulation operations)."
},
{
"Name": "analysisRoundTripMs",
"Type": "Integer",
"Description": "The time (ms) for the round trip to the analysis module and back."
}
]
}
]
}
}
}
Click to expand...
Did you reboot the PC, this is import for the change that I made with the file.
 
fenderman said:
Thanks. I am using the yolov5 .net. However it still shows CPU on the dashboard even after hundreds of triggers. I had this issue with 2.0.8 as well on other systems (2.0.8 otherwise works great). My understanding was that once ai processed its first image with yolov5 .net, the dash would display gpu. Otherwise its was a really smooth install and working perfectly.
Your work on this project along with the others is fantastic and appreciated!
Click to expand...
The issue may be the iGPU driver, it needs to be DirectX 12 compatible. You can check if it is DirectX 12 compatible by running dxdiag
1702782937613.png
 
  • Like
Reactions: David L and fenderman
You must log in or register to reply here.
Top Bottom