Forums - question about snpe-throughput-net-run

2 posts / 0 new
Last post
question about snpe-throughput-net-run
963090831
Join Date: 10 May 22
Posts: 1
Posted: Thu, 2022-07-14 00:42

I noticed snpe-throughput-net-run has no parameter like --set_output_tensors, does it generate all outputs nodes? I have this confusion because I find that this parameter in snpe-net-run can cause a huge gap in inference time. I want to know if the inference time of snpe-throughput-net-run is equivalent to snpe-net-run without --set_output_tensors.

I also want to ask where can I find the log output when set --debug about  snpe-throughput-net-run? I set the --storage_dir but i only noticed a small .pb file, and i don't know how to read it.

How to solve the problems? Thanks for the help!!

  • Up0
  • Down0
weihuan
Join Date: 12 Apr 20
Posts: 270
Posted: Sat, 2022-07-16 07:36

Dear customer,

The core scheme for snpe-throughput-net-run is same as snpe-net-run, but has bits of logs output. You can't generat the output logs even you add option of --debug. Please move to snpe-net-run if you hopes to know more debugging info.

Regarding the mode accuracy issue, you also can try the tool of snpe-net-run and debug your issue model.

BR.
Wei

 

  • 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.