def _get_best_chromosome(chromosomes, data, count_clusters):
"""!
@brief Returns the current best chromosome.
@param[in] chromosomes (list): Chromosomes that are used for searching.
@param[in] data (list): Input data that is used for clustering process.
@param[in] count_clusters (uint): Amount of clusters that should be allocated.
@return (list, float, list) The best chromosome, its fitness function value and fitness function values for
all chromosomes.
"""
# Calc centers
centres = ga_math.get_centres(chromosomes, data, count_clusters)
# Calc Fitness functions
fitness_functions = genetic_algorithm._calc_fitness_function(
centres, data, chromosomes)
# Index of the best chromosome
best_chromosome_idx = fitness_functions.argmin()
# Get chromosome with the best fitness function
return chromosomes[best_chromosome_idx], fitness_functions[
best_chromosome_idx], fitness_functions
def _select(chromosomes, data, count_clusters, select_coeff):
"""!
@brief Performs selection procedure where new chromosomes are calculated.
@param[in] chromosomes (numpy.array): Chromosomes
"""
# Calc centers
centres = ga_math.get_centres(chromosomes, data, count_clusters)
# Calc fitness functions
fitness = genetic_algorithm._calc_fitness_function(
centres, data, chromosomes)
for _idx in range(len(fitness)):
fitness[_idx] = math.exp(1 + fitness[_idx] * select_coeff)
# Calc probability vector
probabilities = ga_math.calc_probability_vector(fitness)
# Select P chromosomes with probabilities
new_chromosomes = np.zeros(chromosomes.shape, dtype=np.int)
# Selecting
for _idx in range(len(chromosomes)):
new_chromosomes[_idx] = chromosomes[ga_math.get_uniform(
probabilities)]
return new_chromosomes