Intel® Edge Software Hub
Get answers from community peers to your questions about building Edge Software Hub solutions for edge compute node.
Announcements
Welcome to the Intel Community! If you like the answer to your question, please mark it as 'Accepted Solution' to help others.

The Edge Software Vision Package for Red Hat Enterprise Linux is now available here.

Learn more about developing on Intel® Hardware and Software here.
421 Discussions

Issue with INT8 quantization of PCB Defect detection model on EII 3.0.1

sagi-scalers-ai
Beginner
727 Views
System information (version)
EII => 3.0.1
Operating System => Ubuntu 20.04
Problem classification => INFO

 

We are trying to use EII on 4th Gen Intel Xeon(Sapphire Rapids). We are able to custom build EII with latest OpenVINO to leverage AMX optimizations using the steps from Issue with current EII 3.0.1 in leveraging AMX optimization on 4th Gen Intel Xeon(Sapphire Rapids) .

 

But we see that the PCB defect detection demo on EII is not using INT8 quantized model and this might not effectively use the capability of AMX optimization on Sapphire Rapids.

 

Please guide us on the steps for INT8 quantization of PCB defect detection model.

Labels (1)
0 Kudos
1 Solution
JesusE_Intel
Moderator
699 Views

Hi sagi-scalers-ai,

 

Thanks for reaching out, I am in the process of re-installing EII to investigate further. I believe the models for the PCB Defect detection are provided in FP16 and FP32. You could try to quantize the model to INT8 using the OpenVINO toolkit using the Post-Training Optimization Tool (POT) default quantization as this may not require an annotated dataset.

 

https://docs.openvino.ai/latest/pot_default_quantization_usage.html

 

Regards,

Jesus

 

 

View solution in original post

0 Kudos
4 Replies
JesusE_Intel
Moderator
700 Views

Hi sagi-scalers-ai,

 

Thanks for reaching out, I am in the process of re-installing EII to investigate further. I believe the models for the PCB Defect detection are provided in FP16 and FP32. You could try to quantize the model to INT8 using the OpenVINO toolkit using the Post-Training Optimization Tool (POT) default quantization as this may not require an annotated dataset.

 

https://docs.openvino.ai/latest/pot_default_quantization_usage.html

 

Regards,

Jesus

 

 

0 Kudos
JesusE_Intel
Moderator
642 Views

Hi sagi-scalers-ai,


I can confirm that all the IR models included in Edge Insights for Industrial are either in FP16 or FP32 format.

Where you able to quantize the model using OpenVINO?


Regards,

Jesus


0 Kudos
JesusE_Intel
Moderator
618 Views

If you need any additional information, please submit a new question as this thread will no longer be monitored.


0 Kudos
FaizanK
Beginner
599 Views

Hi, 

where can i find older version of EII like 2.6.3, or reference implementation of textile defect classifcation?

0 Kudos
Reply