When quantizing dlc model by command 'snpe-dlc-quantize', there are '--enable_hta' and '--enable_htp' params.
What's the difference between HTA & HTP?
Thanks.
BR
When quantizing dlc model by command 'snpe-dlc-quantize', there are '--enable_hta' and '--enable_htp' params.
What's the difference between HTA & HTP?
Thanks.
BR
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.
Dear developer,
HTP=HMX+HVX The HMX and HVX are all DSP module
HHTA is a speicifc accelerator which has been remove after SM8350.
BR.
Wei
Dear Weihuan,
I have few queries could you please answer them
Query 1 : How to differentiate HVX and HMX with respect to the Deep Learning model inference?
Query 2 : How to control these modules in QNN SDK to run DL model?
Query 3 : If we have two DL models, can we split them to HVX and HMX ?
Query 4 : If we have 4 deep learning models can we split them to 4 HVX threads/cores?
It would be kind of you to answer our queries.
Thanks and Regards,
Vyom Mishra