topic Hello, in Intel® Distribution of OpenVINO™ Toolkit

Global Response Normalization difference between Myriad and CPU

Tolbert__Donnie — Thu, 28 Mar 2019 23:36:20 GMT

I've noticed a difference between the output of the GRN layer using the CPU plugin and the Myriad plugin. The CPU plugin behaves as I would expect, by dividing the channel values by the L2 norm of the channel vector, but the Myriad plugin appears to divide by the absolute value of each channel value, creating a vector of +1 and -1 values. Here are two notebooks showing my test results:

CPU: https://gist.github.com/donniet/0d0e8f604eb30aeb0215968dc4e20071

Myriad: https://gist.github.com/donniet/fffdc95526ffa5fa625068db4cdb0507

I've tested on a few devices so I believe the problem is in the plugin.

Python Version: 3.6.7

OpenVINO version: 2018.5.455

Operating System: Ubuntu 18.04

Have other's noticed this or can you reproduce it?

-Donnie

Dear Donnie:

Shubha_R_Intel — Wed, 03 Apr 2019 19:26:10 GMT

Dear Donnie:

Can you kindly try with OpenVino 2019 R1 ? Many improvements and fixes have been made - it was just released yesterday.

Thanks for using OpenVino !

Shubha

Hello,

Tolbert__Donnie — Wed, 01 May 2019 11:24:43 GMT

Hello,

I recently upgraded to the newest version, 2019 R1 133 and I have the same issue with the Myriad plugin and Global Response Normalization. The above gists produce the same results as the previous version.

Can the internal devs at intel reproduce the issue?

Thanks,

-Donnie

Dear Tolbert, Donnie,

Shubha_R_Intel — Wed, 01 May 2019 18:41:45 GMT

Dear Tolbert, Donnie,

I will have internal OpenVino developers look at this issue.

Thanks,

Shubha