torch / optim

A numeric optimization package for Torch.
Other
197 stars 152 forks source link

In SGD, why set default dampening to momentum? #163

Open YurongYou opened 7 years ago

YurongYou commented 7 years ago

https://github.com/torch/optim/blob/656c42af1f996e4a5d6aae3b9aeac831ca162241/sgd.lua#L35

lucasb-eyer commented 6 years ago

This actually de-couples the dampening and momentum. Look at how it is used in the code, all other frameworks use "mom" instead of "damp" hard-coded.