Forums - Losing accuracy on DLC conversion

2 posts / 0 new
or Register
Last post
Losing accuracy on DLC conversion
clovett
Join Date: 30 May 16
Posts: 17
Posted: Mon, 2022-04-11 14:13
Top

I have some ONNX models that convert to DLC and quantize really well (meaning they run fast and keep high accuracy) but I have some that don't.  Do you have any whitepapers or anything on what layer types work well, and what to avoid so that the DLC and quantization work well?  Also does the DLC conversion try and leverage the 4 threads on the DSP or do we have to design the model in a way that maximises parallelism potential?  Any whitepapers on that topic?  

  • Up0
  • Down0
Re: Losing accuracy on DLC conversion
clovett
Join Date: 30 May 16
Posts: 17
Posted: Thu, 2022-04-14 11:54
Top

I think we've narrowed down the problem to the use of blinear interpolation.  Without that things work better.

  • 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.