The concept of entropy of random variables first defined by Shannon has been generalized later in various ways by mathematicians who so obtained new measures of uncertainty, again for random variables. Recently, the author suggested another extension which provides a meaningful definition for the entropy of deterministic functions, both in the sense of Shannon and of Renyi. These measures of uncertainty are different from those which are utilized by physicists in the study of chaotic dynamics, like the Kolmogorov entropy for instance. The aim of this paper is to go a step further, and to derive measures of uncertainty for operators, by using exactly the same rationale. After a short background on the entropies of deterministic functions, one obtains successively the entropy of a constant square matrix operator, the entropy of a varying square matrix operator, the entropy of the kernel of an integral transformation, and the entropy of differential operators defined by square matrices. Then one carefully exhibits the relation which exists between these results and the quantum mechanical entropy first introduced by Von Neumann, and one so obtains a new generalized quantum mechanical entropy which applies to matrics which are not necessarily density matrices. Finally, some illustrative examples for future applications are outlined.