Adding Quantizer APIs to Float Models - 1.4.1 English

Vitis AI RNN User Guide (UG1563)

Document ID
UG1563
Release Date
2021-12-03
Version
1.4.1 English

Suppose there is a trained float model and some Python scripts to evaluate the model's accuracy before quantization. The Quantizer API replaces the float module with a quantized module. The normal evaluate function encourages quantized module forwarding. Quantize calibration determines quantization steps of tensors in the evaluation process if quant_mode flag is set to calib. After calibration, evaluate the quantized model by setting quant_mode to test.