Search found 21 matches

by lemontree
Dec 28, 2009 4:09
Forum: Community Discussion
Topic: Is FreeBasic still continued to develop?
Replies: 139
Views: 47412

I like FreeBasic and will continue to use it. I had hoped it would become an easy to use object orientated language that would be ported to forthcoming intel Larrabee chip( http://en.wikipedia.org/wiki/Larrabee_(GPU) ). People get married, have kids and take on many duties and responsibilities in li...
by lemontree
Oct 26, 2009 6:51
Forum: Community Discussion
Topic: time to say goodbye
Replies: 14
Views: 5173

I would never use the terms C++ and stability in the same breath. C was developed to make programming 8-bit microprocessors easier. C++ was for 16 bit microprocessors and there it should have stopped. Up to you if you want to spend years learning C++. It is unlikely that you will be able to do anyth...
by lemontree
Sep 27, 2009 4:52
Forum: Sources, Examples, Tips and Tricks
Topic: Continuous Gray code numerical optimisation
Replies: 3
Views: 1928

I've been trying out the Gray code algorithm for a few days now. Hey, it's really amazing. Nearly all genetic algorithms are convergent in nature. Once they start converging that's it, there is no going back and that makes them less than suitable where you want open-ended evolution or co-evolution. ...
by lemontree
Sep 18, 2009 7:21
Forum: Sources, Examples, Tips and Tricks
Topic: Continuous Gray code numerical optimisation
Replies: 3
Views: 1928

Continuous Gray code numerical optimisation

I learned about this simple and interesting algorithm from this paper: http://www.cs.bham.ac.uk/~jer/papers/ctsgray.pdf function continuousGray(x as double,precision as double=20) as double dim as double y if rnd>.5 then y=x+exp(-precision*rnd) if y>1.0 then y-=2.0 else y=x-exp(-precision*rnd) if y<...
by lemontree
Aug 21, 2009 15:41
Forum: Sources, Examples, Tips and Tricks
Topic: Compressive Sensing
Replies: 3
Views: 1985

I think one of the ideas is that you get the neural net to learn relatively small sized random projections of the (sparse) training data, rather than the full training data itself. Later when you give the neural net some input it will recall a random projection that could be decoded into the much la...
by lemontree
Aug 14, 2009 14:52
Forum: Sources, Examples, Tips and Tricks
Topic: Compressive Sensing
Replies: 3
Views: 1985

I just found an American Scientist article on the subject:
http://www.americanscientist.org/issues ... -best-bits
by lemontree
Aug 14, 2009 13:31
Forum: Sources, Examples, Tips and Tricks
Topic: Compressive Sensing
Replies: 3
Views: 1985

Compressive Sensing

Compressive Sensing is a pretty dramatic new development in Digital Signal Processing that is having a big impact even beyond that field. It's very useful for neural nets and things like that. Anyway you can think of it as a way of compressing readings from a sensor with a very low computational bur...
by lemontree
Mar 28, 2009 14:14
Forum: Archive
Topic: OCT Resonator data compression / extrapolation algorithm
Replies: 2
Views: 1360

OCT Resonator data compression / extrapolation algorithm

Hey I just invented a cool new algorithm to do data compression / extrapolation and pattern generation. This is hot off the presses so I don't really know what it can do yet, but anyway it's yours to try out already:

http://code.google.com/p/lemontree/downloads/list
by lemontree
Dec 17, 2008 6:45
Forum: Archive
Topic: Self Assembly - Tiles on a board
Replies: 4
Views: 1746

I don't know why you get a blank screen. The internet is a hall of smoke and mirrors. The only thing that code is trying to do is create a self assembling tiling that will reproduce a little picture of a house stored in an array. Where it gets difficult is if the tiles represented electronic compone...
by lemontree
Dec 04, 2008 15:38
Forum: Archive
Topic: Self Assembly - Tiles on a board
Replies: 4
Views: 1746

Anyway I am getting much better and faster results using the following initialization for PolyHouse.bas size=initSA(3,4,4,20,20) InitEDAParams(10,100000,4,10,5) Each position on the board receives an X signal from the left and a Y signal from the top. These signals are then concatenated into a combi...
by lemontree
Dec 03, 2008 6:02
Forum: Archive
Topic: Self Assembly - Tiles on a board
Replies: 4
Views: 1746

Self Assembly - Tiles on a board

I have just started trying out code related to computational self assembly. It seems like a very powerful idea, I don't know why it has been so little researched. The general idea is that you have a bunch of tiles that have different affinities for each other and you let them stick together on a boa...
by lemontree
Sep 19, 2008 10:06
Forum: Sources, Examples, Tips and Tricks
Topic: Numerical Optimizer - Find the minimum of a function
Replies: 0
Views: 1627

Numerical Optimizer - Find the minimum of a function

Do I ever stop? Can I ever be stopped? No and yeh, sure. The following optimizer code is almost a drop in replacement for the optimizers in the FBMath library. It is rather slow because it uses both global and local optimization. It should be able to solve some difficult problems. Oh yeh, I am pasti...
by lemontree
May 23, 2008 13:51
Forum: Sources, Examples, Tips and Tricks
Topic: A Bloom filter, for the forgetful.
Replies: 2
Views: 2085

Please do fix my mistakes.
I have just put some more code in the usual place:
http://code.google.com/p/lemontree/downloads/list
A good few things this time. Walsh Hadamard transform wavelet compression and some tomfoolery with neural nets.
by lemontree
May 15, 2008 23:40
Forum: General
Topic: Tools to aid in converting C to FreeBASIC
Replies: 5
Views: 2729

I have tried the D programming language. It is good, my problem with it is that it is a moving target. I don't mind if keywords and templates and other things are added to a language. What gets me are the fundamental changes to the meanings of already defined keywords. I think Walter Bright is an in...
by lemontree
May 12, 2008 7:43
Forum: Sources, Examples, Tips and Tricks
Topic: A Bloom filter, for the forgetful.
Replies: 2
Views: 2085

A Bloom filter, for the forgetful.

"A Bloom filter is a simple space-efficient randomized data structure for representing a set in order to support membership queries. Bloom filters allow false positives but the space savings often outweigh this drawback when the probability of an error is controlled. Bloom filters have been use...