Change power from 0.66 to 1.0, like natural gradient.

This commit is contained in:
Daniel Povey 2022-05-27 16:45:42 +08:00
parent 89fad8cc5a
commit eed864a3db

View File

@ -639,7 +639,7 @@ class Cain(Optimizer):
this_delta_ref = grad / denom
this_delta = grad / (denom ** 1.5)
this_delta = grad / (exp_avg_sq + eps)
renorm_scale = ((this_delta_ref**2).mean() / ((this_delta**2).mean() + eps)) ** 0.5