I need to work on post training quantization of neural networks for master thesis that is compatible on edge GPU and FPGAs, the current research trend as for quantization aware