Fetching the paper…

Trained Quantization Thresholds for Accurate and Efficient Fixed-Point Inference of Deep Neural Networks · Around