Optimizing the Model - 1.4.1 English

Vitis AI User Guide (UG1414)

Document ID
UG1414
Release Date
2021-12-13
Version
1.4.1 English
Note: Optimizing the model is an optional step.

The Vitis AI optimizer provides the ability to optimize neural network models. Currently, the Vitis AI optimizer includes only one tool called the Vitis AI pruner (VAI pruner), which prunes redundant connections in neural networks and reduces the overall required operations. The pruned models produced by the VAI pruner can be further quantized by the VAI quantizer and deployed to an FPGA.

Figure 1. Vitis AI Optimizer

The VAI pruner supports four deep learning frameworks: TensorFlow, PyTorch, Caffe, and Darknet. The corresponding tool names are vai_p_tensorflow, vai_p_pytorch, vai_p_caffe, and vai_p_darknet, where the "p" in the middle stands for pruning.

For more information, see the Vitis AI Optimizer User Guide (UG1333).

The Vitis AI optimizer requires a commercial license to run. Contact a Xilinx sales representative for more information.