Create a single-precision floating-point number from an unsigned integer corresponding to an IEEE 754 binary representation.
