cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
gguasti
Moderator
Moderator
762 Views
Registered: ‎11-29-2007

Vitis AI 1.4 is live

0 Kudos
8 Replies
syw0122
Observer
Observer
698 Views
Registered: ‎01-14-2020

can you link any example for U250? i cannot find

0 Kudos
gguasti
Moderator
Moderator
627 Views
Registered: ‎11-29-2007

0 Kudos
marcus0
Visitor
Visitor
594 Views
Registered: ‎05-13-2021

Does Vitis AI 1.4 support old version Vivado?(maybe 2019.1-2020.2) Because I had seen the old DPUTRD being remove from Xilinx

0 Kudos
gguasti
Moderator
Moderator
579 Views
Registered: ‎11-29-2007

Vivado 2021.1 should be used.

You can find the TRD design here: https://github.com/Xilinx/Vitis-AI/tree/master/dsa

0 Kudos
marcus0
Visitor
Visitor
567 Views
Registered: ‎05-13-2021

Noted, if I am using old version of Vivado can I still find old version DPU IP?

0 Kudos
Km
Participant
Participant
506 Views
Registered: ‎05-13-2021

Hi @gguasti 

I had problems with quantizing YoloV3 model with older Vitis AI version, you mentioned to give a try with Vitis AI 1.4, however I am still having issues (different erros). Would you mind assisting on the erros. The link on that thread is here 

0 Kudos
shashwat_98
Observer
Observer
380 Views
Registered: ‎06-05-2018

Hey @gguasti ,

I upgraded from vitis ai 1.3.0 to vitis ai 1.4.0....but for some experiments I want to revert to vitis ai 1.3.0.....How can I do that? Is that possible?

Regards,

Shashwat

0 Kudos
gguasti
Moderator
Moderator
83 Views
Registered: ‎11-29-2007

Hi,

yes you can always go back to 1.3.0: we should use the "tags" to identify the Vitis AI release.

Open a terminal and move to Vitis-AI folder; please type 

git tag

now you can revert to an older Vitis-AI release by typing

git checkout v1.3

 

gguasti_0-1631028187021.png

Please notice that you also need the v1.3 docker release.

In the same way, you can return to v1.4

0 Kudos