Big-O of while loop with nested-if
问题 I have this algorithm, and I am trying to calculate its complexity. A = {{B_1}, {B_2}, {B_3}, ..., {B_n}} t_n = 0 //for every set in A while A != empty Calculate f(t_n) for each element in A using t_n. n' = argmin(A) //n' is the element (set) in A with smallest f(t_n) t_n' = t_n' + 1 if (t_n' == Constant) B_n' = B_n' - {k} //k is an element in B_n' if(B_n' == empty) A = A - {B_n'} A is a set, and if the condition (t_n' == Constant) and if(B_n' == empty) are true , we remove an element (set)