24 marzo 2018 anni - QAT: Quantization and training of neural networks for efficient integer-arithmetic-only inference, CVPR

Descrizione:

QAT employs the affine mapping of integers to real values with two constant parameters: Scale and Zeropoint. It first subtracts the Zero-point parameter from data (weights/activation), then divides the data by scale parameter, and finally getting the quantized results with rounding operation and affine mapping.

Aggiunto al nastro di tempo:

Data: