|
A little cool program I made that generates various fractals. My original was based on OWL, however I have lost the compiler. Thus I had to rewire it for Builder that?s why there are two source files and executables one (P9.exe) requires dlls I no longer have.
Download Program US Mirror 307 KB
Download Source & Program US Mirror 1.4 MB
Subject: problems to get to your C++ examples
Date/Time: 2004-11-13 12:58:20
Hi,
I wandered to your page looking for a least squares method program. I got then interested in your C++ classes, but it seems they've got somehow broken links (at least the NeuralNetwork and LeastSquares ones). I'd be thankful if you notified me when they're fixed and/or send them to me by e-mail.
Btw, it's nice you've got so easy to use feedback system :)
Greetz
mateusz
|
From: Mateusz
Subject: RE: problems to get to your C++ examples
Date/Time: 2004-11-14 10:44:22
Dear Mateusz
I haven't gotten round to formalising them yet.
However the classes are included in the least squares program source code I've attached what you need and a page from the least squares manual that should given you enough to get it working. You'll probably have to remove quite a bit of stuff especial any mention to ErrorSeries if you're not using Borland C++ Builder 5 or above. Although, it should be straight forward to do get the code working in any c++ complier. In addition, some of the classes are missing their destructors if your bothered about memory leaks.
To create your own dataset do this
int nopoints = 10 // eg 10 point data set dataset *xdata = new dataset;
xdata->x = new row(nopoints);
xdata->y = new row(nopoints);
xdata->erry = new row(nopoints); //error in y
//-- add your data
(*xdata->x)[0] = 1
(*xdata->x)[0] = 1
(*xdata->erry)[0] = 0.1
// .... 1-8 range
(*xdata->x)[9] = 10
(*xdata->x)[9] = 20.1
(*xdata->erry)[9] = 0.2
To make your fit class either create one. see pdf for creating one (taken from help in Least Squares program).
E.g. Use General Polynomial.
int xx = 1; // xx = 1 for y = a*x + b , xx = 2 for a*x^2 + b*x +c etc; bfit *xline = new polynom(xx);
xline->downer = xdata; //
xline->fit();
Job done
Hopes this helps
Adam
Attached Files:
APErrorSeries.cpp 10 k
APErrorSeries.dfm 1 k
APErrorSeries.h 3 k
AdamsConst.h 0 k
AdamsStd.cpp 3 k
Engine.cpp 49 k
Engine.h 11 k
Fit Routine_.pdf
Dir
|
From: Adam
Ask a question
|
