Module tutorial5

source code

tutorial5

Contains all parameters for the evolutionary run, grammar rules, constraints, and specifics about the terminal and function set of the trees in tutorial5. In this example, we evolve hybrid system: a model containing discrete values and logic operators and numerical operations. More specifically, we will evolve if_then_else types rules that will determine the application of different polynomials... from 20 sets of testing data (20 different values for x and y). Considering the constraints for building the trees, the root node will have 3 children, and these will be ordered ADF defining branches (we simply won't use ADF terminals nodes) describing the structure of the if then else statement. The solution found should be the equivalent of this expression: if x>y then cos(x) else sin(y)

A typical way to run the tutorial would be to:

• modify the settings file so that:

  >>>     # setup for running tutorial 5
          functions = tutorial5.functions
          crossover_mapping=tutorial5.crossover_mapping
          nb_eval=tutorial5.nb_eval
          ideal_results=tutorial5.GetIdealResultsData()
          terminals =tutorial5.terminals
          Strongly_Typed_Crossover_degree=tutorial5.Strongly_Typed_Crossover_degree
          Substitute_Mutation=tutorial5.Substitute_Mutation
          treeRules = tutorial5.treeRules
          adfOrdered = tutorial5.adfOrdered
          FitnessFunction = tutorial5.FitnessFunction

• use a different fitness function for the evaluation of a hybrid system (we use both discrete and continuous values now!) We create 1 new fitness function:
  • FinalFitness3 is a modification of FinalFitness that take as input sets of If Then Else types of trees and return the sum of their fitnesses. These fitness functions are called in 3 different places. We need to update the code that call them. These fitness functions are called in FitnessFunction in the tutorial5 module.

    >>>    def FitnessFunction(my_tree):
    >>>        return evalfitness.FinalFitness3(evalfitness.EvalTreeForAllInputSets(my_tree,xrange(nb_eval)))

• run the evolution in the main method, and make sure that the root node parameter only have three children.e.g.

  >>>    import evolver
  if __name__ == "__main__":
  >>>    dbname=r'C:\pop_db'
  >>>    evolver.EvolutionRun(2000,(0,3,'root'),3,8,'AddHalfNode',100, 0.00001 ,0.5,0.49,7,0.8,dbname,True)

  This means that we define:

• a population size of 2000 individuals,
• a root node with 3 children
• a minimum tree depth of 3 (because the constraints of the tree would not work for a depth <3)
• a maximum tree depth of 8
• the use of a Ramped half and Half Koza tree building method
• a maximum number of runs of 100 generations before we stop
• a stoping fitness criteria of 0.1 (if the fitness<=0.1, solution found)
• a crossover probability of 0.5
• a mutation probability of 0.49
• a reproduction probability of 0.01 (automatically deduced from the 2 previous values)
• the size of the tournament selection
• the probability of selecting the fitttest during tournament selection
• a database of name and path dbname
• a run done in verbose mode (printing the fittest element of each generation)

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.