Mixed-Precision Neural Network Quantization via Learned Layer-wise Importance

Open in new window