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.
6434 Discussions

OpenVINO Inference Engine uses synchronous or asynchronous inference mode

nat98
New Contributor I
1,088 Views

Hello there,

 

I am doing some benchmark/comparison for my original (unoptimized) model and the OpenVINO (optimized) model. The performance evaluation of the optimized model is done using DL Workbench.

 

Now, I would like to know that, what is the default inference mode when I am measuring throughput/accuracy/latency in DL Workbench? Is it synchronous or asynchronous inference mode?

 

Thanks.

 

Regards,

nat98

0 Kudos
1 Solution
Wan_Intel
Moderator
949 Views

Hi Nat98,

Thanks for reaching out to us.

 

For your information, the default inference mode when measuring throughput/accuracy/latency in DL Workbench is the asynchronous inference mode.

 

 

Regards,

Wan


View solution in original post

0 Kudos
4 Replies
Wan_Intel
Moderator
950 Views

Hi Nat98,

Thanks for reaching out to us.

 

For your information, the default inference mode when measuring throughput/accuracy/latency in DL Workbench is the asynchronous inference mode.

 

 

Regards,

Wan


0 Kudos
artyom_tugaryov
Employee
936 Views

Hello @nat98 ,

 

DL Workbench runs benchmark application with default inference mode to measure performance of a model. The default inference mode in benchmark application is the asynchronous mode.

To measure accuracy of a model DL Workbench uses  accuracy_checker with the asynchronous mode by default.

 

nat98
New Contributor I
922 Views

Hello Wan_Intel and artyom_tugaryov,

 

Thanks for answering my question and thanks for both of your replies.

 

Regards,

Nathaniel Tan

0 Kudos
Wan_Intel
Moderator
875 Views

Hi Nat98,

You are welcome.

 

This thread will no longer be monitored since this issue has been resolved. 

If you need any additional information from Intel, please submit a new question.

 

 

Regards,

Wan


0 Kudos
Reply