INTRODUCTION

Last year when I was working with a friend for some atomic physicists, I had to 
do a lot of fitting to cos^n-distributions (sputtering of gold on gold), a very
boring and time consuming job. The thought of making a program to do all this 
for me therefore quickly showed up, when I was considering a project.

The header file includes 2 base classes, one for data type, and one for fit 
type. From these to I have derived classes, so far only one fit type and 3 data
classes.
The structure is:

base class Data; (abstract class)
	derived class LinearData;
	derived class ExponentialData;
	derived class CosineData;
	
base class Fit; (abstract class)
	derived class StraightLineFit;
	

LinearData: Fits to functions y=ax+b
ExponentialData: Fits to Y=a*exp(bx)
CosineData: Fits to y=a*(cos(x))^n , the input must be in radians.

The program will save the original data + the fitted formula in a file.
To see the result, go in to gnuplot and write the command:

gnuplot> plot formula, "filename"

( As an ex. I used
gnuplot> plot 0.101921 * x + -0.0503571, "data/OhmsLaw.dat"
to see how good my 5LC-students data from the fall confirmed Ohm's Law.)

I have included some of my data files, so you can see how it works. (Then you
can also use this data as an input and save time!)
As mentioned, OhmsLaw.dat is a fit to a straight line, 
then there is exp.dat, and 2 cosine datafiles: cos.dat and AuSputt.dat.

I have decided only to include one main program. To change to exponential or 
cosine, simply replace all 'LinearData' with 'ExponentialData' or 'CosineData'.
(And maybe change the name, too!)