tf.nn.normalize_moments( counts, mean_ss, variance_ss, shift, name=None )
See the guide: Neural Network > Normalization
Calculate the mean and variance of based on the sufficient statistics.
Tensorcontaining a the total count of the data (one value).
Tensorcontaining the mean sufficient statistics: the (possibly shifted) sum of the elements to average over.
Tensorcontaining the variance sufficient statistics: the (possibly shifted) squared sum of the data to compute the variance over.
Tensorcontaining the value by which the data is shifted for numerical stability, or
Noneif no shift was performed.
name: Name used to scope the operations that compute the moments.