degra02 / gal-solver

Linear Algebra Program

gal-solver's Issues

Use operator methods overload instead of dedicated functions

In all the objects that supports some sort of sum/multiplication/ecc use operator overloading explained here that make structs support operations of this sort:

Matrix a = Matrix(...);
Matrix b = Matrix(...);
Matrix c = a + b;

I will send a pull request soon.

Function destroy_all()

Create a function that automatically destroys all elements that use Heap memory

Float approximation

There seem to be approximation problems because of the float values.
It is not something that needs to be fixed asap, since the user-input values are (most of the times), limited to only a few decimal numbers.

Approximation

Check lambda division in the rref

Use Epsilon

define EPS 0.00001

GRANDE DEGRA

sei il mio eroe

Important change with the functioning of the library

Due to numerical approximations that cannot be controlled by the user (or us programmer), it would be advisable to change the whole functioning of the library. This means that all float types should be converted into a new "fraction" structure, containing a numerator and a denominator, both integer numbers, so removing any errors that could come up with the whole float approximation thing.

Finish vector functions