Clarification for step_size?

[chigkim]

From readme, "step_size : number of training iterations per half cycle. Authors suggest setting step_size = (2-8) x (training iterations in epoch) . Default 2000."
Does it mean step_size should be "np.ceil(x_train.shape[0]/batch_size/2)" or "2*np.ceil(x_train.shape[0]/batch_size)"?

[maxfrei750]

The number of training iterations is clearly np.ceil(x_train.shape[0]/batch_size), so I'd say 2*np.ceil(x_train.shape[0]/batch_size)
is correct if you want to achieve

2 x (training iterations in epoch)