Forums - Whether Snapdragon610 GPU support both Float and Int precision?

2 posts / 0 new
or Register
Last post
Whether Snapdragon610 GPU support both Float and Int precision?
amal.antony
Join Date: 29 Nov 22
Posts: 3
Posted: Thu, 2023-01-05 00:55
Top

Im using Thundercomm TurboX610 for running the ML models.
Do the GPU runtime support both the Int and Float precision for all its computations?

Can anyone share the documents related to this?

  • Up0
  • Down0
Re: Whether Snapdragon610 GPU support both Float and Int...
weihuan
Join Date: 12 Apr 20
Posts: 270
Posted: Sat, 2023-01-07 23:41
Top

Dear developer,

SNPE GPU runtime don't support fixed points for inference. The data will be dequantized to float event if you input the quantized model and fixed inputs.

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.