bbell03 / geneticalgo Goto Github PK

* self.dimen - items to select from 7
* self.maxweight - max capacity of bag 120
* self.population - array of 100 genomes, initialized empty
* self.fit - array of 100 corresponding fitness values
* self.weight - each of 7 items weight
* self.importance - each of 7 items importance value

def crossover(parent1, parent2) -- FRINGE OP
  * takes two parent genomes and returns two children genomes

def generator()
  * takes no argument and returns a randomly generated genome where
  a genome is represented by an array of 7 bits representing a binary
  choice for each item there is to choose from

def sort_list(tup_list)
  * takes a list of tuples and returns the list sorted in
  descending order by the second tuple element's value
  (in this case fitness)

def fitness(self, genome) -- FITNESS FUNCTION
  * takes a reference to self and a genome and returns the fitness
  value of that genome by summing its fitness values

def pool(self) -- mate and cull GA, SOLUTION TEST
  * takes a reference to self, contains the main functionality of the GA
  which creates a population of different knapsack placements and culls
  it by 50% leaving only the fittest according to fitness value,
  then mating these fittest to replenish the 50% culled.

def mutation(self, genome) -- FRINGE OP
  * takes a reference to self and implements a 10% chance of mutation in
  which one of the 7 bits in a genome array are flipped.