Forums - pytorch QAT to dlc conversion failed

3 posts / 0 new
or Register
Last post
pytorch QAT to dlc conversion failed
neeraj.kasera
Join Date: 20 Sep 22
Posts: 1
Posted: Thu, 2022-09-22 23:19
Top
I have trained a pytorch QAT model and converted it to torchscript.
 
Now I need to convert QAT torchscript model to .dlc.
 
But while runnung snpe-pytorch-to-dlc command, I am getting below error
TypeError : Unsupported Op type qnn.quantize.
 
Please let me know way to support QAT model for dlc conversion.
  • Up0
  • Down0
Re: pytorch QAT to dlc conversion failed
weihuan
Join Date: 12 Apr 20
Posts: 270
Posted: Sat, 2022-09-24 09:36
Top

Dear customer,

Coud you please help to add debug option os --debug to conversion steps and then share more logs to us for checking which layer caused?

BR.

Wei

  • Up0
  • Down0
Re: pytorch QAT to dlc conversion failed
sanjjey.a.sanjjey
Join Date: 17 May 22
Posts: 55
Posted: Tue, 2022-09-27 04:19
Top

Hi,

Please find which layer is not supported while doing conversion.Write JSON file for the layer which is not supported.After that use --udo_config_paths <JSON_file> to conversion step.

For reference please check below links

https://developer.qualcomm.com/sites/default/files/docs/snpe/udo_package...

Tools https://developer.qualcomm.com/sites/default/files/docs/snpe/tools.html#...

 

  • 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.