topic Re: How to understand the output of mvNCProfile ? in Intel® Distribution of OpenVINO™ Toolkit

How to understand the output of mvNCProfile ?

idata — Wed, 07 Feb 2018 14:38:20 GMT

Some states of GoogLeNet:

Detailed Per Layer Profile
Layer      Name                                 MFLOPs    Bandwidth MB/s        time(ms)
========================================================================================
0          conv1/7x7_s2                        236.028           2505.00            5.63
1          pool1/3x3_s2                          1.806           1441.66            1.06
2          pool1/norm1                           0.000            712.67            0.54
3          conv2/3x3_reduce                     25.690            404.11            0.97
4          conv2/3x3                           693.633            316.67           11.55
5          conv2/norm2                           0.000            797.05            1.44

So how to find the bottlenecks of the network from the states above?

And I notice the "Size Limitations" with caffe:

  Compiled Movidius “graph” file < 320 MB; Intermediate layer buffer size < 100 MB
  Scratch Memory size < 112 KB

So how to compute the Intermediate layer buffer size and Scratch Memory size ? How to make sure that my network is consonant with the "Size Limitations" ?

Re: How to understand the output of mvNCProfile ?

idata — Fri, 09 Feb 2018 02:40:59 GMT

@z_huabao In mvNCProfile's output you can see the processing time and processing bandwidth of each layer. Using this information, you can then tune your network to find a balance of speed and accuracy.

Regarding the size limitations, you can compute the intermediate layer buffer size using Caffe and dividing the values by 2 since we are using fp16, however this is very tedious. The NCSDK will warn you if your model exceeds these size limits, so you don't have to worry if that is what you are concerned about.

Re: How to understand the output of mvNCProfile ?

idata — Fri, 09 Feb 2018 11:06:46 GMT

@Tome_at_Intel thanks for ans, can you tell more details:

Is it good or bad when the value of MFLOPs is big?

Is it good or bad when the value of Bandwidth is big?

Re: How to understand the output of mvNCProfile ?

idata — Sat, 10 Feb 2018 01:53:51 GMT

@z_huabao It depends on what you are trying to do. Larger MFLOPS and bandwidth numbers aren't necessarily bad but if performance is a concern, these numbers could give a hint as to where the slowdowns are.