High-Fidelity Conversion of Floating-Point Networks for Low-Precision Inference using Distillation with Limited Data


Share

High-Fidelity Conversion of Floating-Point Networks for Low-Precision Inference using Distillation with Limited Data