BriefGPT.xyz
Nov, 2020
针对MobileNet的Subtensor量化
Subtensor Quantization for Mobilenets
HTML
PDF
Thu Dinh, Andrey Melnikov, Vasilios Daskalopoulos, Sek Chai
TL;DR
本文研究深度神经网络量化的问题，针对不同的架构提出了一些不同的替代方案，并在ImageNet数据集上进行了图像分类实验，结果表明后量化准确率与浮点数版本在0.7％以内。
Abstract
quantization
for
deep neural networks
(DNN) have enabled developers to deploy models with less memory and more efficient low-power inference. However, not all DNN designs are friendly to
→