The Lambert W function is the inverse of z = w * tf.exp(w). Lambert W is a
complex-valued function, but here it returns only the real part of the image
of the function. It also only returns the principal branch (also known as
A Tensor with type float32 or float64.
A name for the operation (optional).
The Lambert W function evaluated at z. A Tensor with same shape
and same dtype as z.