AI Dynamics

Global AI News Aggregator

About

Quantization-Aware Training: Keeping Weights as Floats During Training

Yeah you can’t convert weights to int for training because you need to calculate the gradients during training. There are lots of different techniques for QAT, maybe this implementation helps: https://
pytorch.org/blog/quantizat
ion-aware-training/

→ View original post on X — @rasbt