math-lib-float

This library provides Float & TrigoFloat modules which export the following funcitons.

Float module

1. new - int -> int -> Float.t
2. inverse - Float.t -> Float.t
3. add - Float.t -> Float.t -> Float.t
4. sub - Float.t -> Float.t -> Float.t
5. lt - Float.t -> Float.t -> bool
6. lte - Float.t -> Float.t -> bool
7. gte - Float.t -> Float.t -> bool
8. gt - Float.t -> Float.t -> bool
9. mul - Float.t -> Float.t -> Float.t
10. div - Float.t -> Float.t -> Float.t
11. modulo - Float.t -> Float.t -> Float.t

TrigoFloat module

1. sinus - Float.t * nat -> Float.t
2. cosinus - Float.t * nat -> Float.t

The floating poing numbers are represented by a pair (a,b) where n = a * 10^b. The scientifc notation considers that the a is defined as a float betwen -10 and 10) and b is a nat. Since there is no native float type, the a is a nat and can be greater than 10. All operations are applied in order to keep the number as a pair (a,b) where n = a * 10^b.

Basde on floating-point representation, this library introduces an implementation of trigonometric functions (cosinus, sinus). The implementation of sinus is based on Chebychev polynoms interpolation.

Tests

A makefile is provided to launch tests.

make test