|
Name
|
Description
|
|
TRSBinaryGeneticProgramming
|
The TRSBinaryGeneticProgramming component is the descendant class of the RSGeneticProgramming.TRSCustomBinaryGeneticProgramming component which publishes all the properties needed to use genetic algorithms in your programs.
Genetic programming is a computer science method, inspired by evolutionary biology, for automatically solving problems, without having to know or define the form or structure of optimum problem structure beforehand. You define the basic building blocks (functions, constants, and variables) of the problem and then the component does the rest. Genetic programming solves problems by evolving a group or population of candidate individuals through successive generations, selecting fitter (or better) child individuals for each generation, until a solution is found. It uses evolutionary biology techniques such as inheritance, mutation, selection, and crossover (also called recombination).
The TRSBinaryGeneticProgramming class contains a population of individuals (TRSGPPopulation<TGAFloat>), which contain genetic program trees as DNA. These DNA are used to represent the solution to the search problem. They represent the functions, constants, and variables of each candidate solution (or individual), represented by a TRSGPIndividual<TGAFloat> class.
After setting up the Instructions that your genetic programs can use, you set the RSGeneticProgramming.TRSBinaryGeneticProgramming.InitialPopulation , define a fitness function ( RSGeneticProgramming.TRSBinaryGeneticProgramming.OnEvaluateFitness event) to properly "score" each individual, and then Evolve your solution.
 Notes to Implementers
|
The TRSBinaryGeneticProgramming class redeclares the Instructions property and OnConstantMutation event to use instantiated generics types that look like non-generic types. Unfortunately, when a published property contains a generic, even an instantiated one (e.g., TRSGPInstructions<TGAFloat>), Delphi will not publish the property and make it available at design-time.
|
|
|
TRSCustomBinaryGeneticProgramming
|
The TRSCustomBinaryGeneticProgramming component is the base class for including genetic programming (GP) in your programs that use floating point values and use functions of at most arity = 2. It instantiates the RSGenerics.GeneticProgramming.TRSCustomGeneticProgramming<T> generics class using TGAFloat and ensures the correct TRSGPInstruction<T> and TRSGPNode<T> classes are created for it. This class creates nodes and instructions with functions of at most arity = 2, it is slightly faster and uses less memory than the RSGeneticProgramming.TRSCustomGeneticProgramming class, which is the more general solution.
Genetic programming is a computer science method, inspired by evolutionary biology, for automatically solving problems, without having to know or define the form or structure of optimum problem structure beforehand. You define the basic building blocks (functions, constants, and variables) of the problem and then the component does the rest. Genetic programming solves problems by evolving a group or population of candidate individuals through successive generations, selecting fitter (or better) child individuals for each generation, until a solution is found. It uses evolutionary biology techniques such as inheritance, mutation, selection, and crossover (also called recombination).
The TRSCustomBinaryGeneticProgramming class contains a population of individuals (TRSGPPopulation<TGAFloat>), which contain genetic program trees as DNA. These DNA are used to represent the solution to the search problem. They represent the functions, constants, and variables of each candidate solution (or individual), represented by a TRSGPIndividual<TGAFloat> class.
After setting up the Instructions that your genetic programs can use, you set the InitialPopulation , define a fitness function ( OnEvaluateFitness event) to properly "score" each individual, and then Evolve your solution.
Notes to Implementers
|
The TRSCustomBinaryGeneticProgramming class redeclares the Instructions property and OnConstantMutation event to use instantiated generics types that look like non-generic types. Unfortunately, when a published property contains a generic, even an instantiated one (e.g., TRSGPInstructions<TGAFloat>), Delphi will not publish the property and make it available at design-time.
|
|
|
TRSCustomGeneticProgramming
|
The TRSCustomGeneticProgramming component is the base class for including genetic programming (GP) in your programs that use floating point values. It instantiates the RSGenerics.GeneticProgramming.TRSCustomGeneticProgramming<T> generics class using TGAFloat and ensures the correct TRSGPInstruction<T> and TRSGPNode<T> classes are created for it. This class creates nodes and instructions with arbitrary number of child nodes (arity > 0). If your problem requires at most functions of arity = 2, it is recommended to use the RSGeneticProgramming.TRSCustomBinaryGeneticProgramming class as it is slightly faster and uses slightly less memory.
Genetic programming is a computer science method, inspired by evolutionary biology, for automatically solving problems, without having to know or define the form or structure of the optimum problem structure beforehand. You define the basic building blocks (functions, constants, and variables) of the problem and then the component does the rest. Genetic programming solves problems by evolving a group or population of candidate individuals through successive generations, selecting fitter (or better) child individuals for each generation, until a solution is found. It uses evolutionary biology techniques such as inheritance, mutation, selection, and crossover (also called recombination).
The TRSCustomGeneticProgramming class contains a population of individuals (TRSGPPopulation<TGAFloat>), which contain genetic program trees as DNA. These DNA are used to represent the solution to the search problem. They represent the functions, constants, and variables of each candidate solution (or individual), represented by a TRSGPIndividual<TGAFloat> class.
After setting up the Instructions that your genetic programs can use, you set the InitialPopulation , define a fitness function ( OnEvaluateFitness event) to properly "score" each individual, and then Evolve your solution.
Notes to Implementers
|
The TRSCustomGeneticProgramming class redeclares the Instructions property and OnConstantMutation event to use instantiated generics types that look like non-generic types. Unfortunately, when a published property contains a generic, even an instantiated one (e.g., TRSGPInstructions<TGAFloat>), Delphi will not publish the property and make it available at design-time.
|
|
|
TRSCustomGeneticProgrammingExecutor
|
The TRSCustomGeneticProgrammingExecutor component is the base class for reading and executing a genetic program. Use this class when you have already evolved a solution to your problem and need to execute the genetic program. It instantiates the RSGenerics.GeneticProgramming|TRSCustomGeneticProgrammingExector<T> generics class using TGAFloat and ensures the correct TRSGPInstruction<T> are created for it.
Notes to Implementers
|
The TRSCustomGeneticProgrammingExecutor class redeclares the Instructions property to use instantiated generics types that look like non-generic types. Unfortunately, when a published property contains a generic, even an instantiated one (e.g., TRSGPInstructions<TGAFloat>), Delphi will not publish the property and make it available at design-time.
|
|
|
TRSCustomGeneticProgrammingVariant
|
The TRSCustomGeneticProgrammingVariant component is the base class for including genetic programming (GP) in your programs that use Variant values. It instantiates the RSGenerics.GeneticProgramming.TRSCustomGeneticProgramming<T> generics class using Variant and ensures the correct TRSGPInstruction<T> and TRSGPNode<T> classes are created for it. This class creates nodes and instructions with arbitrary number of child nodes (arity > 0).
Genetic programming is a computer science method, inspired by evolutionary biology, for automatically solving problems, without having to know or define the form or structure of the optimum problem structure beforehand. You define the basic building blocks (functions, constants, and variables) of the problem and then the component does the rest. Genetic programming solves problems by evolving a group or population of candidate individuals through successive generations, selecting fitter (or better) child individuals for each generation, until a solution is found. It uses evolutionary biology techniques such as inheritance, mutation, selection, and crossover (also called recombination).
The TRSCustomGeneticProgrammingVariant class contains a population of individuals (TRSGPPopulation<Variant>), which contain genetic program trees as DNA. These DNA are used to represent the solution to the search problem. They represent the functions, constants, and variables of each candidate solution (or individual), represented by a TRSGPIndividual<Variant> class.
After setting up the Instructions that your genetic programs can use, you set the InitialPopulation , define a fitness function ( OnEvaluateFitness event) to properly "score" each individual, and then Evolve your solution.
Tip
|
The TRSCustomGeneticProgrammingVariant component uses more memory and will be generally slower than the TRSCustomGeneticProgramming component (which uses doubles as opposed to variants). On the other hand, variants can pass around many more types of values and generally be more flexible. However, it is important to remember that your functions must satisfy the closure property, i.e., all functions in the instructions should be able to accept any other function or terminal as inputs.
|
Notes to Implementers
|
The TRSCustomGeneticProgrammingVariant class redeclares the Instructions property and OnConstantMutation event to use instantiated generics types that look like non-generic types. Unfortunately, when a published property contains a generic, even an instantiated one (e.g., TRSGPInstructions<Variant>), Delphi will not publish the property and make it available at design-time.
|
|
|
TRSCustomGeneticProgrammingVariantExecutor
|
The TRSCustomGeneticProgrammingVariantExecutor component is the base class for reading and executing a genetic program using variants. Use this class when you have already evolved a solution to your problem and need to execute the genetic program. It instantiates the RSGenerics.GeneticProgramming|TRSCustomGeneticProgrammingVariantExector<T> generics class using TGAFloat and ensures the correct TRSGPInstruction<T> are created for it.
Notes to Implementers
|
The TRSCustomGeneticProgrammingVariantExecutor class redeclares the Instructions property to use instantiated generics types that look like non-generic types. Unfortunately, when a published property contains a generic, even an instantiated one (e.g., TRSGPInstructions<Variant>), Delphi will not publish the property and make it available at design-time.
|
|
|
TRSGeneticProgramming
|
The TRSGeneticProgramming component is the descendant class of the RSGeneticProgramming.TRSCustomGeneticProgramming component which publishes all the properties needed to use genetic algorithms in your programs.
Genetic programming is a computer science method, inspired by evolutionary biology, for automatically solving problems, without having to know or define the form or structure of optimum problem structure beforehand. You define the basic building blocks (functions, constants, and variables) of the problem and then the component does the rest. Genetic programming solves problems by evolving a group or population of candidate individuals through successive generations, selecting fitter (or better) child individuals for each generation, until a solution is found. It uses evolutionary biology techniques such as inheritance, mutation, selection, and crossover (also called recombination).
The TRSGeneticProgramming class contains a population of individuals (TRSGPPopulation<TGAFloat>), which contain genetic program trees as DNA. These DNA are used to represent the solution to the search problem. They represent the functions, constants, and variables of each candidate solution (or individual), represented by a TRSGPIndividual<TGAFloat> class.
After setting up the Instructions that your genetic programs can use, you set the RSGeneticProgramming.TRSGeneticProgramming.InitialPopulation , define a fitness function ( RSGeneticProgramming.TRSGeneticProgramming.OnEvaluateFitness event) to properly "score" each individual, and then Evolve your solution.
Notes to Implementers
|
The TRSGeneticProgramming class redeclares the Instructions property and OnConstantMutation event to use instantiated generics types that look like non-generic types. Unfortunately, when a published property contains a generic, even an instantiated one (e.g., TRSGPInstructions<TGAFloat>), Delphi will not publish the property and make it available at design-time.
|
|
|
TRSGeneticProgrammingExecutor
|
The TRSGeneticProgrammingExecutor component is the descendant class of the RSGeneticProgramming.TRSCustomGeneticProgrammingExecutor component which publishes all the properties needed for reading and executing a genetic program. Use this class when you have already evolved a solution to your problem and need to execute the genetic program. It instantiates the RSGenerics.GeneticProgramming|TRSCustomGeneticProgrammingExector<T> generics class using TGAFloat and ensures the correct TRSGPInstruction<T> are created for it.
Notes to Implementers
|
The TRSGeneticProgrammingExecutor class redeclares the Instructions property to use instantiated generics types that look like non-generic types. Unfortunately, when a published property contains a generic, even an instantiated one (e.g., TRSGPInstructions<TGAFloat>), Delphi will not publish the property and make it available at design-time.
|
|
|
TRSGeneticProgrammingVariant
|
The TRSGeneticProgrammingVariant component is the descendant class of the RSGeneticProgramming.TRSCustomGeneticProgrammingVariant component which publishes all the properties needed to use genetic algorithms in your programs.
Genetic programming is a computer science method, inspired by evolutionary biology, for automatically solving problems, without having to know or define the form or structure of optimum problem structure beforehand. You define the basic building blocks (functions, constants, and variables) of the problem and then the component does the rest. Genetic programming solves problems by evolving a group or population of candidate individuals through successive generations, selecting fitter (or better) child individuals for each generation, until a solution is found. It uses evolutionary biology techniques such as inheritance, mutation, selection, and crossover (also called recombination).
The TRSGeneticProgrammingVariant class contains a population of individuals (TRSGPPopulation<Variant>), which contain genetic program trees as DNA. These DNA are used to represent the solution to the search problem. They represent the functions, constants, and variables of each candidate solution (or individual), represented by a TRSGPIndividual<Variant> class.
After setting up the Instructions that your genetic programs can use, you set the RSGeneticProgramming.TRSGeneticProgrammingVariant.InitialPopulation , define a fitness function ( RSGeneticProgramming.TRSGeneticProgrammingVariant.OnEvaluateFitness event) to properly "score" each individual, and then Evolve your solution.
Tip
|
The TRSGeneticProgrammingVariant component uses more memory and will be generally slower than the TRSGeneticProgramming component (which uses doubles as opposed to variants). On the other hand, variants can pass around many more types of values and generally be more flexible. However, it is important to remember that your functions must satisfy the closure property, i.e., all functions in the instructions should be able to accept any other function or terminal as inputs.
|
Notes to Implementers
|
The TRSGeneticProgrammingVariant class redeclares the Instructions property and OnConstantMutation event to use instantiated generics types that look like non-generic types. Unfortunately, when a published property contains a generic, even an instantiated one (e.g., TRSGPInstructions<Variant>), Delphi will not publish the property and make it available at design-time.
|
|
|
TRSGeneticProgrammingVariantExecutor
|
The TRSGeneticProgrammingVariantExecutor component is the descendant class of the RSGeneticProgramming.TRSCustomGeneticProgrammingVariantExecutor component which publishes all the properties needed for reading and executing a genetic program. Use this class when you have already evolved a solution to your problem and need to execute the genetic program. It instantiates the RSGenerics.GeneticProgramming|TRSCustomGeneticProgrammingExector<T> generics class using Variants and ensures the correct TRSGPInstruction<T> are created for it.
Notes to Implementers
|
The TRSGeneticProgrammingVariantExecutor class redeclares the Instructions property to use instantiated generics types that look like non-generic types. Unfortunately, when a published property contains a generic, even an instantiated one (e.g., TRSGPInstructions<Variant>), Delphi will not publish the property and make it available at design-time.
|
|
|
TRSGPBinaryFloatNode
|
Defines the floating point node for tree-based genetic programming which can contain at most 2 children. Each node represents one function or terminal (constant, variable) in a genetic program. It has one Parent (or nil if root node) and 0 or more Child nodes (terminals have no children).
Notes to Implementers
|
The TRSGPBinaryFloatNode class was created so that LoadValueFromXML. SaveValueToXML. amd ValueToString methods could be defined (using FloatToStr and StrToFloat).
|
|
|
TRSGPFloatInstruction
|
Defines one instruction (function, variable, or constant) in the genetic program.
 Note
|
When the instruction executes, it needs no information beyond the nodes and the values passed in, so it may be multi-threaded as long as the values are copied and unless the OnExecute event does something unsafe.
|
Notes to Implementers
|
The TRSGPFloatInstruction class was created so that LoadValueFromXML and SaveValueToXML methods could be defined (using FloatToStr and StrToFloat). In addition, the OnExecute event handler could be declared published using a instantiated generic type that looks non-generic. Unfortunately, when a published property contains a generic, even an instantiated one (e.g., TRSGPInstruction<TGAFloat>), Delphi will not publish the property and make it available at design-time.
|
|
|
TRSGPFloatInstructions
|
Represents type TRSGPFloatInstructions.
|
|
TRSGPFloatNode
|
Defines the floating point node for tree-based genetic programming which can contain an arbitrary number of children. Each node represents one function or terminal (constant, variable) in a genetic program. It has one Parent (or nil if root node) and 0 or more Child nodes (terminals have no children).
Note
|
For the TRSGPFloatNode class, the children are stored in a list. If you restrict your instructions to 2 or less children, it is recommended to use the RSGeneticProgramming.TRSGPBinaryFloatNode as it uses slightly less memory and is slightly faster.
|
Notes to Implementers
|
The TRSGPFloatNode class was created so that LoadValueFromXML. SaveValueToXML. amd ValueToString methods could be defined (using FloatToStr and StrToFloat).
|
|
|
TRSGPVariantInstruction
|
Defines one instruction (function, variable, or constant) in the genetic program.
Note
|
When the instruction executes, it needs no information beyond the nodes and the values passed in, so it may be multi-threaded as long as the values are copied and unless the OnExecute event does something unsafe.
|
Notes to Implementers
|
The TRSGPVariantInstruction class was created so that LoadValueFromXML and SaveValueToXML methods could be defined. In addition, the OnExecute event handler could be declared published using a instantiated generic type that looks non-generic. Unfortunately, when a published property contains a generic, even an instantiated one (e.g., TRSGPInstruction<Variant>), Delphi will not publish the property and make it available at design-time.
|
|
|
TRSGPVariantInstructions
|
Represents type TRSGPVariantInstructions.
|
|
TRSGPVariantNode
|
Defines the Variant node for tree-based genetic programming which can contain an arbitrary number of children. Each node represents one function or terminal (constant, variable) in a genetic program. It has one Parent (or nil if root node) and 0 or more Child nodes (terminals have no children).
Notes to Implementers
|
The TRSGPVariantNode class was created so that LoadValueFromXML. SaveValueToXML. amd ValueToString methods could be defined
|
|
Top
|
