Using Vitis AI Runtime APIs - 1.4 English

Vitis AI User Guide (UG1414)

Document ID
Release Date
1.4 English

Vitis AI Runtime (VART) enables applications to use the unified high-level runtime API for both cloud and edge. Therefore, making cloud-to-edge deployments seamless and efficient.

To use Vitis AI Runtime, use the following operations:

  1. Create a DpuRunner object by calling create_runner.
  2. Query the DpuRunner for the tensor format by calling get_input_tensors.
  3. Query the DpuRunner for the shape and name of the input tensors it expects for its loaded AI model by get_output_tensors.
  4. Execute the runner by calling execute_async.
  5. Wait for the runner job to complete by calling wait

For more details about the programming with VART, see the Programming with Vitis AI Runtime (VART).