ec.app.majority

Class CA

java.lang.Object

ec.app.majority.CA

All Implemented Interfaces:

java.io.Serializable

public class CA
extends java.lang.Object
implements java.io.Serializable

CA.java Implements a 1-dimensional toroidal CA for purposes of doing the binary majority classification problem. You need to supply the automaton size and the rule neighborhood size at constructor time. The CA itself consists of an array of integers, each 0 or 1. You can clear this array, set it to preset values, or randomize it. You provide the rule as an array of ints, all 0 or 1 as well. The rules are specified in the following order. Let's say that you have a neighborhood of 3, consiting of cells LCR, where L is the cell "left" of the target cell (left is less than, right is greater than). Then the order of the rules are for the neighborhoods 0: 000, 1: 001, 2: 010, 3: 011, 4: 100, 5: 101, 6: 110, 7: 111. In other words, the neighborhood is interpreted as a binary number and that's the index into the rule.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
CA(int width, int neighborhood)

Method Summary

Modifier and Type	Method and Description
void	clear(boolean toOnes)
boolean	converged()
int[]	getRule()
int[]	getVals()
void	randomize(EvolutionState state, int thread)
void	setRule(int[] r)
void	setVals(int[] vals)
void	step(int steps, boolean stopWhenConverged)

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

CA

public CA(int width,
          int neighborhood)

Method Detail

getVals

public int[] getVals()

getRule

public int[] getRule()

setRule

public void setRule(int[] r)

setVals

public void setVals(int[] vals)

clear

public void clear(boolean toOnes)

converged

public final boolean converged()

randomize

public void randomize(EvolutionState state,
                      int thread)

step

public void step(int steps,
                 boolean stopWhenConverged)