gene

Documentation

Overview

Package gene provides the basis for a single gene in GEP.

Index

• func CheckEqual(g1 *Gene, g2 *Gene) error
• type FuncWeight
• type Gene
  • func New(x string, funcType functions.FuncType) *Gene
  • func RandomNew(headSize, tailSize, numTerminals, numConstants int, functions []FuncWeight, ...) *Gene
  • func (g *Gene) Dup() *Gene
  • func (g *Gene) EvalBool(in []bool) bool
  • func (g *Gene) EvalInt(in []int) int
  • func (g *Gene) EvalMath(in []float64) float64
  • func (g *Gene) EvalVectorInt(in []VectorInt) VectorInt
  • func (g *Gene) Expression(grammar *grammars.Grammar, helpers grammars.HelperMap) (string, error)
  • func (g *Gene) Mutate()
  • func (g Gene) String() string
  • func (g *Gene) SymbolCount(sym string) int
• type VectorInt

Functions

func CheckEqual

func CheckEqual(g1 *Gene, g2 *Gene) error

CheckEqual is used for testing purposes only (exported to use in genome_test.go).

Types

type FuncWeight

type FuncWeight struct {
	Symbol string
	Weight int
}

FuncWeight contains the symbol name and its weight to be used during a run of the GEP algorithm. A symbol with weight 5, for example, will be five times more likely to be used than a symbol with weight 1.

type Gene

type Gene struct {
	// Symbols is the slice of strings being used in this gene's expression.
	Symbols []string
	// Constants is the slice of floats available for use by this gene.
	Constants []float64

	SymbolMap map[string]int // do not use directly.  Use SymbolCount() instead.
	// contains filtered or unexported fields
}

Gene contains all the information needed to represent a single gene in a GEP expression.

func New

func New(x string, funcType functions.FuncType) *Gene

New creates a new gene based on the Karva string representation.

func RandomNew

func RandomNew(headSize, tailSize, numTerminals, numConstants int, functions []FuncWeight, funcType functions.FuncType) *Gene

RandomNew generates a new, random gene for further manipulation by the GEP algorithm. The headSize, tailSize, numTerminals, and numConstants determine the respective properties of the gene, and functions provide the available functions and their respective weights to be used in the creation of the gene.

func (*Gene) Dup

func (g *Gene) Dup() *Gene

Dup duplicates the gene into the provided destination gene.

func (*Gene) EvalBool

func (g *Gene) EvalBool(in []bool) bool

EvalBool evaluates the gene as a boolean expression and returns the result. "in" represents the boolean inputs available to the gene.

func (*Gene) EvalInt

func (g *Gene) EvalInt(in []int) int

EvalInt evaluates the gene as a floating-point expression and returns the result. in represents the int inputs available to the gene.

func (*Gene) EvalMath

func (g *Gene) EvalMath(in []float64) float64

EvalMath evaluates the gene as a floating-point expression and returns the result. in represents the float64 inputs available to the gene.

func (*Gene) EvalVectorInt

func (g *Gene) EvalVectorInt(in []VectorInt) VectorInt

EvalVectorInt evaluates the gene as a floating-point expression and returns the result. in represents the int inputs available to the gene.

func (*Gene) Expression

func (g *Gene) Expression(grammar *grammars.Grammar, helpers grammars.HelperMap) (string, error)

Expression builds up the expression tree and returns the resulting string. While building, it keeps track of any helper functions that are needed.

func (*Gene) Mutate

func (g *Gene) Mutate()

Mutate mutates a gene by performing a single random symbol exchange within the gene.

func (Gene) String

func (g Gene) String() string

String returns the Karva representation of the gene.

func (*Gene) SymbolCount

func (g *Gene) SymbolCount(sym string) int

SymbolCount returns the count of the number of times the symbol is actually used in the Gene. Note that this count is typically different from the number of times the symbol appears in the Karva expression. This can be a handy metric to assist in the fitness evaluation of a Gene.

type VectorInt

type VectorInt = functions.VectorInt