Scilab Function frexp - dissect floating-point numbers into base 2 exponent and mantissa
Calling Sequence
- [f,e]=frexp(x)
Parameters
- x
: real vector or matrix
- f
: array of real values, usually in the range 0.5 <= abs(f) < 1.
- e
: array of integers that satisfy the equation: x = f.*2.^e
Description
This function corresponds to the ANSI C function frexp(). Any zeros in x produce
f=0 and e=0.
Examples
[f,e]=frexp([1,%pi,-3,%eps])
See Also