Forums - Changing NN input to batch input

4 posts / 0 new
Last post
Changing NN input to batch input
orenhakira
Join Date: 8 Jul 21
Posts: 1
Posted: Tue, 2021-08-03 04:22

Hi.

I am trying to convert a certain nn from onnx to dlc so it will get a batch sized 2 as input.

The original input is 1x128x256x3 and I want to convert it so that the input will be 2x128x256x3.

The original output is 1x512 and the new output is 2x512.

For a given image, I executed the original network and got feature vector and then tried to concatenate the image twice and pass it

 as input for the new network. The output feature vectors are different. I expected that both feature vectors from the new network will be

 the same as the feature vector from the original network.

Should it really be the same feature vectors?

Thanks!

  • Up0
  • Down0
dwenceri484
Join Date: 19 Aug 21
Posts: 1
Posted: Thu, 2021-08-19 04:01

HI,all

I am also concerned about this issue.

 

My device is XIAOMI 11 with Snapdragon 888 devices. But the android app provided by SNPE example can run model in the DSP or AIP. I also can not find related information in the SNPE release notes.

 

geometry dash 

  • Up0
  • Down0
corbitt.jame
Join Date: 22 Jun 22
Posts: 1
Posted: Thu, 2022-06-23 05:19

Hello everybody. 

 

And also me faced with this one problem and I dont know what to do with this. Maybe, here is the solution of this issue. 

  • Up0
  • Down0
ruzhongl
Join Date: 7 Jun 21
Posts: 15
Posted: Thu, 2022-06-23 07:40

the test step looks good. yes, you should get same output.

I guess the issue should come from HTP core. In graph optimization stage, the compiler may wrong for special op. 

 

As I known, 2 known bugs for batch > 1.

For v66 arch (8250,8150), a known bug in Resize optimization and issue fixed in latest SNPE SDK. 

For v68(888, 8 Gen 1/1+), a known bug in bert-like models in batch > 1 and issue fixed in latest SNPE SDK.

They may some corner case for batch > 1, we need to look into the model.

  • Up0
  • Down0
or Register

Opinions expressed in the content posted here are the personal opinions of the original authors, and do not necessarily reflect those of Qualcomm Incorporated or its subsidiaries (“Qualcomm”). The content is provided for informational purposes only and is not meant to be an endorsement or representation by Qualcomm or any other party. This site may also provide links or references to non-Qualcomm sites and resources. Qualcomm makes no representations, warranties, or other commitments whatsoever about any non-Qualcomm sites or third-party resources that may be referenced, accessible from, or linked to this site.