Forums - Benchmark for batch input model

2 posts / 0 new
or Register
Last post
Benchmark for batch input model
Jacob_Ke
Join Date: 24 Aug 17
Posts: 10
Posted: Mon, 2018-08-06 19:43
Top

Hi,

How can i benchmark the model which have more than one batch input size in snpe 1.17.0?

if it can work, can you provide .json example?

Thanks

  • Up0
  • Down0
Re: Benchmark for batch input model
gesqdn-forum
Join Date: 4 Nov 18
Posts: 184
Posted: Tue, 2019-09-24 03:33
Top

Hi  Jacob_Ke,
The benchmarking tool "snpe-bench.py" provided Qualcomm Neural Processing SDK internally uses "snpe-net-run".
"snpe-net-run" performs the inference on the provided input image and gives you the output.
Here the benchmarking tool calculates the inference time of each layer in the model and the complete model inference time too.

We don't think we can provide a batch input to a model for prediction.
Kindly let us know if you need more details on this.

  • 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.