Differentiable approximation to round.

Larger alphas correspond to closer approximations of the round function. If alpha is close to zero, this function reduces to the identity.

This is described in Sec. 4.1. in the paper

"Universally Quantized Neural Compression"
Eirikur Agustsson & Lucas Theis

x tf.Tensor. Inputs to the rounding function.
alpha Float or tf.Tensor. Controls smoothness of the approximation.
eps Float. Threshold below which soft_round will return identity.