Activity Not Available

Commits : Listings

  Analyzed about 2 years ago based on code collected about 4 years ago.
Showing page 1 of 1
Commit Message Contributor Files Modified Lines Added Lines Removed Code Location Date
3 new projects created! XMath, GUI and RSA. Starting the implementation of Project XMath, with complex numbers and a big integer square rooting function. Project GA on standby in the meanwhile! Anon32 More... over 4 years ago
Implemented next generation schemes. These schemes define the way a population evolves during each generation. Generational and Steady-State keep the population size each generation, differing in the amount of the old population replaced by the offspring. The new scheme Evolutionary is a real representation of the population dynamics, and in this scheme the population size is variable. Also started 3 more projects: Anon32 More... over 4 years ago
Added validation to the Selection class constructors. Modified the fitness and phenotype type. Fitness is now represented by a double, which permits a wider range of values, and the phenotype is now represented by an Object, which provides more representation than the String type. Anon32 More... over 4 years ago
Implemented more selection algorithms: Anon32 More... over 4 years ago
.classpath and .project files added to svn:ignore, since they are specific to each member of the repository. The coderszone should only contain files containing code, not files created by Eclipse, or other IDE. Anon32 More... over 4 years ago
Eliminated the .classpah and .project file, which should be specific to each member using this repository. Those files will be ignored from now on! Anon32 More... over 4 years ago
Modified the nextGeneration algorithm. A selection object is now responsible for selecting the most fitted (for mating) and the least fitted (who will not survive). The concept of death algorithm was implemented. Each selection algorithm (roulette, sus, tournament, ...) has an equivalent death algorithm (death roulette, suds, death tournament, ...). A death algorithm selects the least fitted individuals of the population, being the perfect inverse of the selection algorithm. Also added a new population stat: invert sum, which is the sum of the inverse of each individual fitness. Anon32 More... over 4 years ago
Changed the way selection works. Now each population has it's own Selection object, which allows for 2 population use different selection algorithms. This also is more useful for when considering multi-tasking. Each population can evolve in it's own thread, without conflicts (this aspect will be refined). Also started the preparations for the tournament selection implementation. Anon32 More... over 4 years ago
Roulette-wheel selection implemented, and made the default selection algorithm. Some minor changes in the Environment and Population class in order to implement the Selection operator. Sum of fitnesses stat added. Anon32 More... over 4 years ago
More functionality added to the population. Next generation implemented. Also created the Selection class (to simulate natural selection) and the Environment class (to simulate the environment of the populations). An environment contains one or more populations, and has a mutation and crossover rate, specific to that environment. The nextGeneration algorithm uses three abstract methods (selection of the individuals to the matingPool, creation of the offspring from that matingPool, and the insertion of that offspring back into the population). The Selection is in it's initial phase, but will evolve and contain the most used selection algorithms (roullette, tournament, SUS, etc...). Anon32 More... over 4 years ago
Added some basic population behavior: adding and removing a given individual. This simulates migration. Two populations can now exchange individuals using this methods. Added also the updated flag, which signals changes in the population. With changes in population comes changes in it's stats. A performance perk has been added in the stat getter methods, using the updated flag. Anon32 More... over 4 years ago
Recombination interface implemented. Population class created, using generics. Still lacking methods to change the population itself. Created a method for calculating the max, min and average fitness of the population. Anon32 More... over 4 years ago
Folder structure reorganization. All projects in packages in the src folder. First project (Genetic Algoritms) in package ga! Anon32 More... over 4 years ago
Reorganizing folders structure! Anon32 More... over 4 years ago
Individual class and Mutate interface! The first step in implementing Genetic Algorithms!! Anon32 More... over 4 years ago
Initial directory structure. Anon32 More... over 4 years ago
 
 
 

Creative Commons License Copyright © 2013 Black Duck Software, Inc. and its contributors, Some Rights Reserved. Unless otherwise marked, this work is licensed under a Creative Commons Attribution 3.0 Unported License . Ohloh ® and the Ohloh logo are trademarks of Black Duck Software, Inc. in the United States and/or other jurisdictions. All other trademarks are the property of their respective holders.