Of course most languages have library functions for this, but suppose I want to do it myself.
Suppose that the float is given like in a C or Java program (except for
For that you have to understand the standard IEEE 754 in order for proper binary representation. After that you can use Float.intBitsToFloat or Double.longBitsToDouble.
http://en.wikipedia.org/wiki/IEEE_754