How to run sample VART models (Xilinx Vitis AI) on Azure NP 20s VM with Alveo U250 Deployment VM XRT 2021.1 - Ubuntu 20.04 Gen 1 Image ?

Rishabh Anand 1 Reputation point
2022-06-08T13:06:06.807+00:00

Hello,

I have acquired an NP20s VM on azure and the image is Alveo U250 Deployment VM XRT 2021.1 - Ubuntu 20.04 Gen 1.
I am trying to run the sample resnet 50 Vitis AI model by following this - For-Cloud.
I am getting the following error when I run the final command -

/workspace/demo/VART/resnet50 > ./resnet50 /usr/share/vitis_ai_library/models/resnet50/resnet50.xmodel  
WARNING: Logging before InitGoogleLogging() is written to STDERR  
I0608 06:28:11.426890   362 main.cc:292] create running for subgraph: subgraph_conv1  
terminate called after throwing an instance of 'std::runtime_error'  
  what():  Error: failed to create XRM context  
Aborted (core dumped)  

I think this is because of some vitis AI setup issue or the Alveo card issue. I am not sure about which all commands I have to run for the setup as I am using an image that has Xilinx Vitis AI built in but the steps on the README are for new installation.

Not Monitored
Not Monitored
Tag not monitored by Microsoft.
41,539 questions
0 comments No comments
{count} votes

1 answer

Sort by: Most helpful
  1. deherman-MSFT 37,841 Reputation points Microsoft Employee
    2022-06-09T18:45:49.627+00:00

    @Rishabh Anand
    I understand you are running into the above error when using VART models on your Azure VM.

    I believe this issue is more related to the software which is offered on the Azure Marketplace. Support for this software is not provided by Azure but you can reach out to them via the support methods from this link.

    https://www.xilinx.com/support.html

    -------------------------------

    Please don’t forget to "Accept the answer" and “up-vote” wherever the information provided helps you, this can be beneficial to other community members.

    0 comments No comments

Your answer

Answers can be marked as Accepted Answers by the question author, which helps users to know the answer solved the author's problem.