Intel® Distribution of OpenVINO™ Toolkit
Community assistance about the Intel® Distribution of OpenVINO™ toolkit, OpenCV, and all aspects of computer vision-related on Intel® platforms.

HDDL using two VPUs and get the [REQUEST BUSY]

Gao__Jack
Beginner
549 Views

Hello,

I refer to the benchmark.cpp to use two VPUs with multi InferRequest on my object detect project.

When I use one VPU with four InferRequest, it's OK.

If I use two VPUs, it get the [REQUEST BUSY] error.

I use the SetCompletionCallback to get the finished inferrequest.

Is it a bug or I missing something?

Thanks

Jack Gao

 

0 Kudos
3 Replies
JesusE_Intel
Moderator
549 Views

Hi Jack Gao,

Thanks for reaching out. Could you please help answer the following questions?

  • Which OS are you using?
  • Which version of the OpenVINO toolkit are you using? Latest OpenVINO 2019 R3.1?
  • Which HDDL device are you using?
  • Which model did you use with the benchmark app?

Regards,

Jesus

0 Kudos
Gao__Jack
Beginner
549 Views

Hi Jesus,

  • Which OS are you using?
  • Windows
  • Which version of the OpenVINO toolkit are you using? Latest OpenVINO 2019 R3.1?
  • Latest Openvino 2019 R3.1
  • Which HDDL device are you using?
  • Our partner gave us a device with 2 myriad X.
  • Which model did you use with the benchmark app?
  • Our model. I tried it on benchmark app but it have some problem on my code which I use the same method as benchmark app

Thanks,

Jack Gao

0 Kudos
JesusE_Intel
Moderator
549 Views

Hi Jack,

Did you complete the Additional Installation Steps for the Intel® Vision Accelerator Design with Intel® Movidius™ VPUs? Could you try to run the benchmark app with one of our pre-trained models first?

python3 /opt/intel/openvino/deployment_tools/tools/model_downloader/downloader.py --name face-detection-retail-0004 ./benchmark_app -d HDDL -m intel/face-detection-retail-0004/FP16/face-detection-retail-0004.xml

Regards,

Jesus

0 Kudos
Reply