For Edge (DPUv2) - 1.1 English

Vitis AI User Guide (UG1414)

Document ID
Release Date
1.1 English
  1. Download the samples from host to the target using scp with the following command.
    $scp -r ~/Vitis-AI/vart root@[IP_OF_BOARD]:~/
  2. Download the vitis_ai_runtime_r1.1_image_video.tar.gz from host to the target using scp with the following command.
    $scp vitis_ai_runtime_r1.1_image_video.tar.gz root@[IP_OF_BOARD]:~/
  3. Unzip the vitis_ai_runtime_r1.1_image_video.tar.gz package.
    $tar -xzvf vitis_ai_runtime_r1.1_image_video.tar.gz -C vart
  4. Enter the directory of samples in the target board. Take resnet50 as an example.
    #cd ~/vart/samples/resnet50
  5. Run the example.
    #./resnet50 model_dir_for_zcu102
    Note: If the above executable program does not exist, run the following command to compile and generate the corresponding executable program.
    #bash –x

The following table shows the run commands for all the Vitis AI samples.

Table 1. Launching Commands for Vitis AI Samples on ZCU102
ID Example Name Command
1 resnet50 ./resnet50 model_dir_for_zcu102
2 resnet50_mt_py python3 1 model_dir_for_zcu102
3 inception_v1_mt_py

python3 1 model_dir_for_zcu102

4 pose_detection ./pose_detection video/pose.mp4 model_dir_for_zcu102
5 video_analysis ./video_analysis video/structure.mp4 model_dir_for_zcu102
6 adas_detection ./adas_detection video/adas.avi model_dir_for_zcu102
7 segmentation ./segmentation video/traffic.mp4 model_dir_for_zcu102