Asymptotic-complexity-testing

R package developers currently use ad-hoc tests of asymptotic computational complexity via empirical timings of functions and visual diagnostic plots. A complexity of an algorithm state, how fast the algorithm is (how many elementary operations are performed) with respect to the input data set. For algorithm classification is usually used the so called asymptotic complexity. The asymptotic complexity can be computed as a count of all elementary operations, or more easily a count of all operations modifying data, or even only as a count of all comparisons.