Returns a vector containing the largest representable floating-point integer values less than or equal to the values of the corresponding elements of the given vector.