Forums - Different inference speed Android App vs snpe-throughput-net-run

2 posts / 0 new
Last post
Different inference speed Android App vs snpe-throughput-net-run
omer.sarioglu
Join Date: 27 Oct 21
Posts: 1
Posted: Sat, 2021-11-20 02:19
Hello to everyone.

When I run the speed tests of the model I created with Qualcomm's own tools, I get different results, and when I run it with java on Android, I get different results.
 
I checked the GPU utilization. While testing with `snpe-throughput-net-run` I see 90%, but with the Android app it does not go above 15%. Are there any additional steps (multi-threading, priority settings etc.) that need to be done? Because I couldn't find any information about it in the documentation. I would be very happy if someone could help.
 
Best regards.
  • Up0
  • Down0
zhiguol
Join Date: 16 Dec 19
Posts: 25
Posted: Sun, 2021-11-21 23:55

Dear customer, 

Could you please provide more details of your test? Like:

1. The SNPE version?

2. The reproduce step you have made to reproduce this issue?

3. Hardware version?

We can check from our side to have a look.

  • 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.