None
EN
Comment on Neural Network Model Quantization On Mobile by Alexis R. Ware
['Roberto Lopez Mendez', 'Technical Paper Link', 'Tom Katsioulas', 'Alex', 'Matt Bailey', 'Alexis R. Ware', 'Yağız Boz', 'Murugavel Ganesan', 'F. Chen', 'B.S. Deepaksubramanyan']
Comments for Semiconductor Engineering
The second section discusses and compares the main two quantization approaches in TensorFlow Lite (TFLite): Post-Training Quantization (PTQ) and Quantization Aware Training (QAT).
We will look in more detail at these approaches in Quantization Aware Training vs Post-Training Quantization.
Quantization Aware Training (QAT) performs quantization by retraining the model, while Post-Training Quantization (PTQ) applies quantization without retraining the model.
As with PTQ, QAT brings clear benefits in terms of model compression by reducing model size by 4x (32-bit/8-bit).
In this way the model learns parameters which are robust to quantization loss and models the accuracy of a quantized model.