Intel® Integrated Performance Primitives
Community support and discussions relating to developing high-performance vision, signal, security, and storage applications.
Announcements
This community is designed for sharing of public information. Please do not share Intel or third-party confidential information here.
6628 Discussions

Standard deviation in MeanStdDev doesn't return the result according to the documentation

Duverney-Prêt__Pierr
220 Views

Hello,

It seems than the MeanStdDev function doesn't calculate the standard deviation following the documentation. After doing a simple test I noticed that the result value is divided by the length of the imput vector and not "length - 1" as it was written on the documentation.

I used the following vector for my test: vec[10] = {0, 1, 2, 3, 4, 5, 6, 7, 8, 9}

I'm using IPP version 2019.0.5

Is it a known issue?

0 Kudos
2 Replies
Gennady_F_Intel
Moderator
220 Views

Hello,

the quick checking with the current ( latest) IPP 2020 shown the following results with ippsMeanStdDev_64f function.

>_tese.exe
         ========================================================================
                 version of IPP is:   ippIP AVX2 (l9) 2020.0.0 Gold (r0x0f66e7e3) 2020.0.0.258402275
         ========================================================================
... ippsMeanStdDev_64f PASSED ...
...Mean == 4.500000, ... StdDev == 3.027650 ...
Do you see the same outputs?

-Gennady

 

Duverney-Prêt__Pierr
220 Views

My outputs are:  Mean = 4.500000, StdDev = 2.8722813232690143 .

I'm expected the same outputs as yours so I will update my version.

Thanks for your help =)

Cheers,

Pierre

Reply