Develop deep-learning AI inference applications on Xilinx® hardware platforms, including both Edge devices and Alveo™ accelerator cards.

Optimize the neural network models in the applications by pruning the redundant connections and reducing overall required operations.

Implement efficient and high-quality neural networks using a unified interface in your applications.

Understand Vitis AI's critical deep learning processing unit (DPU) designed to support different device families and interfaces.