24 март 2018 г. - QAT: Quantization and training of neural
networks for efficient integer-arithmetic-only
inference, CVPR
Описание:
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.
Добавлено на ленту времени:
Дата: