pony_gp is an implementation of Genetic Programming (GP -- see http://geneticprogramming.com).
The purpose of pony_gp is to describe how the GP algorithm works and to be compatible with
emscripten, in particular, to be converted to WebAssembly (see the
branch Website). The intended use is for teaching. The aim is to allow the developer to quickly
start using and developing. The design is supposed to be simple, self contained and use core
C libraries. The original project, written in python, can be found here:
https://github.com/flexgp/pony_gp
Find an equation that produces the given outputs from the given inputs. Example output:
Reading: ../data/fitness_cases.csv, Headers: {a, b, y}, Number of Exemplars: 121
GP Settings:
[[Population Size: 1000, Max Depth: 3, Elite Size: 2, Generations: 2, Tournament Size: 3, Seed: 0.000000, Crossover Probability: 0.800000, Mutation Probability: 0.200000, Verbose: 0, Config: ../data/configs.ini, Functions: {+, *, /, -}, Terminals: {0, 1, a, b}, Arities: {+: 2.000000, *: 2.000000, /: 2.000000, -: 2.000000, 0: 0.000000, 1: 0.000000, a: 0.000000, b: 0.000000}, Fitness Cases: {[-5.000000, -5.000000], [-5.000000, -4.000000], [-5.000000, -3.000000], [-5.000000, -2.000000], [-5.000000, -1.000000], [-5.000000, 0.000000], [-5.000000, 1.000000], [-5.000000, 2.000000], [-5.000000, 3.000000], [-5.000000, 4.000000], [-5.000000, 5.000000], [-4.000000, -5.000000], [-4.000000, -4.000000], [-4.000000, -3.000000], [-4.000000, -2.000000], [-4.000000, -1.000000], [-4.000000, 0.000000], [-4.000000, 1.000000], [-4.000000, 2.000000], [-4.000000, 3.000000], [-4.000000, 4.000000], [-4.000000, 5.000000], [-3.000000, -5.000000], [-3.000000, -4.000000], [-3.000000, -3.000000], [-3.000000, -2.000000], [-3.000000, -1.000000], [-3.000000, 0.000000], [-3.000000, 1.000000], [-3.000000, 2.000000], [-3.000000, 3.000000], [-3.000000, 4.000000], [-3.000000, 5.000000], [-2.000000, -5.000000], [-2.000000, -4.000000], [-2.000000, -3.000000], [-2.000000, -2.000000], [-2.000000, -1.000000], [-2.000000, 0.000000], [-2.000000, 1.000000], [-2.000000, 2.000000], [-2.000000, 3.000000], [-2.000000, 4.000000], [-2.000000, 5.000000], [-1.000000, -5.000000], [-1.000000, -4.000000], [-1.000000, -3.000000], [-1.000000, -2.000000], [-1.000000, -1.000000], [-1.000000, 0.000000], [-1.000000, 1.000000], [-1.000000, 2.000000], [-1.000000, 3.000000], [-1.000000, 4.000000], [-1.000000, 5.000000], [0.000000, -5.000000], [0.000000, -4.000000], [0.000000, -3.000000], [0.000000, -2.000000], [0.000000, -1.000000], [0.000000, 0.000000], [0.000000, 1.000000], [0.000000, 2.000000], [0.000000, 3.000000], [0.000000, 4.000000], [0.000000, 5.000000], [1.000000, -5.000000], [1.000000, -4.000000], [1.000000, -3.000000], [1.000000, -2.000000], [1.000000, -1.000000], [1.000000, 0.000000], [1.000000, 1.000000], [1.000000, 2.000000], [1.000000, 3.000000], [1.000000, 4.000000], [1.000000, 5.000000], [2.000000, -5.000000], [2.000000, -4.000000], [2.000000, -3.000000], [2.000000, -2.000000], [2.000000, -1.000000], [2.000000, 0.000000], [2.000000, 1.000000], [2.000000, 2.000000], [2.000000, 3.000000], [2.000000, 4.000000], [2.000000, 5.000000], [3.000000, -5.000000], [3.000000, -4.000000], [3.000000, -3.000000], [3.000000, -2.000000], [3.000000, -1.000000], [3.000000, 0.000000], [3.000000, 1.000000], [3.000000, 2.000000], [3.000000, 3.000000], [3.000000, 4.000000], [3.000000, 5.000000], [4.000000, -5.000000], [4.000000, -4.000000], [4.000000, -3.000000], [4.000000, -2.000000], [4.000000, -1.000000], [4.000000, 0.000000], [4.000000, 1.000000], [4.000000, 2.000000], [4.000000, 3.000000], [4.000000, 4.000000], [4.000000, 5.000000], [5.000000, -5.000000], [5.000000, -4.000000], [5.000000, -3.000000], [5.000000, -2.000000], [5.000000, -1.000000], [5.000000, 0.000000], [5.000000, 1.000000], [5.000000, 2.000000], [5.000000, 3.000000], [5.000000, 4.000000], [5.000000, 5.000000]}, Targets: {50.000000, 41.000000, 34.000000, 29.000000, 26.000000, 25.000000, 26.000000, 29.000000, 34.000000, 41.000000, 50.000000, 41.000000, 32.000000, 25.000000, 20.000000, 17.000000, 16.000000, 17.000000, 20.000000, 25.000000, 32.000000, 41.000000, 34.000000, 25.000000, 18.000000, 13.000000, 10.000000, 9.000000, 10.000000, 13.000000, 18.000000, 25.000000, 34.000000, 29.000000, 20.000000, 13.000000, 8.000000, 5.000000, 4.000000, 5.000000, 8.000000, 13.000000, 20.000000, 29.000000, 26.000000, 17.000000, 10.000000, 5.000000, 2.000000, 1.000000, 2.000000, 5.000000, 10.000000, 17.000000, 26.000000, 25.000000, 16.000000, 9.000000, 4.000000, 1.000000, 0.000000, 1.000000, 4.000000, 9.000000, 16.000000, 25.000000, 26.000000, 17.000000, 10.000000, 5.000000, 2.000000, 1.000000, 2.000000, 5.000000, 10.000000, 17.000000, 26.000000, 29.000000, 20.000000, 13.000000, 8.000000, 5.000000, 4.000000, 5.000000, 8.000000, 13.000000, 20.000000, 29.000000, 34.000000, 25.000000, 18.000000, 13.000000, 10.000000, 9.000000, 10.000000, 13.000000, 18.000000, 25.000000, 34.000000, 41.000000, 32.000000, 25.000000, 20.000000, 17.000000, 16.000000, 17.000000, 20.000000, 25.000000, 32.000000, 41.000000, 50.000000, 41.000000, 34.000000, 29.000000, 26.000000, 25.000000, 26.000000, 29.000000, 34.000000, 41.000000, 50.000000}]]
Generation: 0, Duration: ~0.0000, fit ave: -958.37+/-105862.475, size ave: 5.24+/-51.991 depth ave: 1.19+/-12.623, max size: 15, max depth: 3, max fit: -0.000000, best solution: Genome: {b*b-a*0+a*1*a+0}, Fitness: -0.0000
Generation: 1, Duration: ~1.0000, fit ave: -550.46+/-1667.830, size ave: 4.80+/-47.394 depth ave: 1.18+/-12.368, max size: 15, max depth: 3, max fit: -0.000000, best solution: Genome: {b*b-a*0+a*1*a+0}, Fitness: -0.0000
Best solution on the training data: Genome: {b*b-a*0+a*1*a+0}, Fitness: -0.0000
Best solution on the test data: Genome: {b*b-a*0+a*1*a+0}, Fitness: -0.0000
Change the parameters from the configs.ini file to your desired
parameters if you wish.
The input(s) with their respective output is in the file data/fitness_case.csv. The
exemplars are generated from y = a^2 + b^2 from range [-5, 5]
To implement a system-dependant time function, modify the function get_time in misc_util.c.
C99 and CMake 3.5+
usage: ./pony_gp --config <CONFIG> --fc <FITNESS_CASES>
[-p <POPULATION_SIZE>] [-m <MAX_DEPTH>] [-e <ELITE_SIZE>]
[-g <GENERATIONS>] [--ts <TOURNAMENT_SIZE>] [-s <SEED>]
[--cp <CROSSOVER_PROBABILITY>] [--mp <MUTATION_PROBABILITY>]
[--tts <TEST_TRAIN_SPLIT>] [-v <VERBOSE>] [-h]
Required arguments:
--config <CONFIG> Config path (INI format). Overridden by CLI-arguments.
--fc <FITNESS_CASES> Fitness cases path. The exemplars of input and the
corresponding output used to train and test individual
solutions. Inputs must be in alphabetical order (not case
sensitive).
Optional arguments:
-h, --help Show this help message and exit.
-p <POPULATION_SIZE> --population_size <POPULATION_SIZE>
Population size is the number of individual solutions
-m <MAX_DEPTH> --max_depth <MAX_DEPTH>
Max depth of tree. Partly determines the search space
of the solutions.
-e <ELITE_SIZE> --elite_size <ELITE_SIZE>
Elite size is the number of best individual solutions
that are preserved between generations.
-g <GENERATIONS> --generations <GENERATIONS>
Number of generations. The number of iterations of the
search loop.
--ts <TOURNAMENT_SIZE> --tournament_size <TOURNAMENT_SIZE>
Tournament size. The number of individual solutions
that are compared when determining which solutions are
inserted into the next generation (iteration) of the
search loop.
-s <SEED> --seed <SEED>
Random seed. For replication of runs of the EA. The
search is stochastic and and replication of the
results are guaranteed the random seed.
--cp <CROSSOVER_PROBABILITY> --crossover_probability <CROSSOVER_PROBABILITY>
Crossover probability, [0.0, 1.0]. The probability of
two individual solutions to be varied by the crossover
operator.
--mp <MUTATION_PROBABILITY> --mutation_probability <MUTATION_PROBABILITY>
Mutation probability, [0.0, 1.0]. The probability of
an individual solutions to be varied by the mutation
operator.
--tts <TEST_TRAIN_SPLIT> --test_train_split <TEST_TRAIN_SPLIT>
Test-train data split, [0.0, 1.0]. The ratio of fitness
cases used for training individual solutions.
-v <VERBOSE> --verbose <VERBOSE>
Set to 1 for verbose printing. Otherwise, 0.
Runs for generations
Initial tree nr: number nodes: number of nodes in tree max_depth: max tree depth tree: symbols in tree
Generation: generation number, duration: evaluation time, fit_ave: average fitness of the generation, size_ave: average number of nodes in the generation amongst all solutions, depth_ave: average max_tree depth,max_size: maximum number of nodes, max_depth: maximum depth, max_fit: maximum fitness best_solution: 'genome': {individual formula/tree}, 'fitness': fitness of genome
Best solution on train data: 'genome': {individual formula/tree}, 'fitness': fitness of genome
Best solution on test data: 'genome': {individual formula/tree}, 'fitness': fitness of genome
Run. Generator type can be anything. Config and fitness case file paths are required.
cmake -G <Generator type>
make
./pony_gp -config <path/to/config/file> -fc <path/to/fitness/cases>
For information on the types of generators, run cmake --help