Genetic Algorithm Implementation in C

A configurable genetic algorithm implementation in C. Demonstrates core GA concepts: selection, crossover, mutation, and fitness evaluation.

Getting Started

Clone the repository and compile:

gcc GeneticAlgorithm.c -o genetic_algorithm -lm

Run:

./genetic_algorithm

Configuration

All parameters are set in config.txt:

pop=100
minGeneValue=1
maxGeneValue=3
size=8
numParents=3
mutationRate=100
maxGens=100000
verbosity=1
fitnessFunc=fitnessCalculator

Parameter	Description
pop	Population size (number of individuals)
minGeneValue	Minimum value a gene can have
maxGeneValue	Maximum value a gene can have
size	Number of genes per individual
numParents	Number of fittest individuals selected for breeding
mutationRate	Higher = fewer mutations (1/mutationRate chance per gene)
maxGens	Maximum generations before stopping
verbosity	0 = silent, 1 = show progress
fitnessFunc	Which fitness function to use

Fitness Functions

Function	Goal	Max Fitness
fitnessCalculator	All genes equal 1	size
palindromeCalculator	Genes form a palindrome	size/2
ascendingCalculator	Genes in ascending order	size-1
alternatingCalculator	Genes alternate min/max	size
noRepeatCalculator	No adjacent genes match	size-1
sumTargetCalculator	Gene sum equals size*2	size*2
primeCalculator	Genes form a prime number	-2

Code Structure

Function	Description
printIndividual()	Prints an individual's genes and fitness
printPopulation()	Prints fitness scores for all individuals
initialisePopulation()	Creates random initial population
naturalSelection()	Selects fittest individuals as parents
getFitness()	Calculates fitness for entire population
reproduction()	Creates next generation via crossover and mutation
geneticAlgorithm()	Main loop orchestrating the evolution
readConfig()	Loads settings from config.txt
getMaxFitness()	Returns max fitness for the selected function

Adding Custom Fitness Functions

1. Add your function following this signature:

   int myFitnessFunc(int* population, int size, int index){
       // Access genes with: *(population + index*size + i)
       // Return fitness score (higher = better)
   }

2. Add max fitness calculation in getMaxFitness()

3. Add the function mapping in main()

4. Use it in config.txt: fitnessFunc=myFitnessFunc

Example Output

Gen 523|Genes|12131213 fitness: 6
Result (Gen 524):12131312 fitness: 7

Real-World Applications

Genetic algorithms are useful when the search space is too large for brute force and there's no gradient to follow:

• Scheduling (shifts, timetables, routes)
• Engineering optimization (antenna design, structural layout)
• Neural network architecture search
• Game AI strategy tuning
• Drug discovery