Fetching the paper…

Channel-wise Hessian Aware trace-Weighted Quantization of Neural Networks · Around