The plot from my keras tuner shows that the 'best step' is always zero. Does this mean that the first iteration of each trial was the best (lowest MSE) or am I misunderstanding?
Many thanks
Asked
Active
Viewed 39 times
