pytorch-pruning-2step
pytorch-pruning-2step copied to clipboard
hou-yz
2-stage pruning to favor distributed inference (local device compute half of the model, upload the feature for further computing on stronger devices or cloud).
Results
1
pytorch-pruning-2step issues
Sort by
recently updated
recently updated
newest added
trafficstars
您好，请问您论文中的在不同计算比 r 情况下的端到端延迟的那张图。是如何绘制的呢？尤其是横坐标的设置。以及根据您论文中的公式。得出 ：End to End lantency = (r +1) 服务器端延迟+传输工作量/R 是否正确？
