33
/pt/
AIzaSyB4mHJ5NPEv-XzF7P6NDYXjlkCWaeKw5bc
November 1, 2025
3909867
288434
1
Public Timelines
FAQ Receber premium

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

Descrição:

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.

Adicionado na linha do tempo:

29 mar 2020
1
0
809

Data:

24 mar 2018 ano
Agora
~ 7 years and 7 months ago