def selection_cross_prob():
count = 11
attempts_size = 30
min_values = []
success_count = []
median_values = []
for k in range(0, count, 1):
results = []
for _ in range(0, attempts_size):
solver = nq.Solver_8_queens(cross_prob=k / 10)
best_fit, epoch_num, visualization = solver.solve(max_epochs=500)
if best_fit == 1.0:
results.append(epoch_num)
if len(results) != 0:
min_values.append(np.min(results))
success_count.append(
int((attempts_size - len(results)) / attempts_size * 100))
median_values.append(np.median(results))
else:
min_values.append(501)
success_count.append(
int((attempts_size - len(results)) / attempts_size * 100))
median_values.append(501)
# print('size; ', size)
# print('success: ', len(results))
# print('min: ', np.min(results))
# print('median: ', np.median(results))
# print('average: ', np.average(results))
# print('max: ', np.max(results))
# print('\n')
pl.subplot(311)
pl.plot(range(0, count, 1), min_values, 'r-', range(0, count, 1),
min_values, 'ro')
pl.subplot(312)
pl.plot(
range(0, count, 1),
median_values,
'g-',
range(0, count, 1),
median_values,
'go',
)
pl.subplot(313)
pl.plot(
range(0, count, 1),
success_count,
'b-',
range(0, count, 1),
success_count,
'bo',
)
pl.show()
def func (x, y):
#best_fit= None
for i in range(len(x)):
for j in range(len(x[0])):
solver= nq.Solver_8_queens(pop_size=100, cross_prob=x[i][j], mut_prob=y[i][j])
best_fit, epoch_num, visualization = solver.solve()
z[i,j]= best_fit
return z
def selection_poll_size():
count = 41
attempts_size = 30
min_values = []
success_count = []
median_values = []
for size in range(2, count, 2):
results = []
for _ in range(0, attempts_size):
solver = nq.Solver_8_queens(pop_size=size)
best_fit, epoch_num, visualization = solver.solve(max_epochs=500)
if best_fit == 1.0:
results.append(epoch_num)
min_values.append(np.min(results))
success_count.append(
int((attempts_size - len(results)) / attempts_size * 100))
median_values.append(np.median(results))
# print('size; ', size)
# print('success: ', len(results))
# print('min: ', np.min(results))
# print('median: ', np.median(results))
# print('average: ', np.average(results))
# print('max: ', np.max(results))
# print('\n')
pl.subplot(311)
pl.plot(range(2, count, 2), min_values, 'r-', range(2, count, 2),
min_values, 'ro')
pl.subplot(312)
pl.plot(
range(2, count, 2),
median_values,
'g-',
range(2, count, 2),
median_values,
'go',
)
pl.subplot(313)
pl.plot(
range(2, count, 2),
success_count,
'b-',
range(2, count, 2),
success_count,
'bo',
)
pl.show()
#!/usr/bin/env python3
# -*- coding: utf-8 -*_
"""
This script uses nqueens.py module for solving 8-queens problem by using
a generic algorithm
"""
import sys
print('Python version:', sys.version)
import nqueens as nq
solver = nq.Solver_8_queens()
best_fit, epoch_num, visualization = solver.solve()
print("Best solution:")
print("Fitness:", best_fit)
print("Iterations:", epoch_num)
print(visualization)
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
This script uses nqueens.py module for solving 8-queens problem by using
a genetic algorithm
"""
import sys
print('Python version:', sys.version)
import nqueens as nq
solver=nq.Solver_8_queens(0.3,0.4,160)
best_sol,best_fit, epoch_num, visualization = solver.solve(0.9,100)
print("Best solution:",best_sol)
print("Fitness:", best_fit)
print("Iterations:", epoch_num)
print(visualization)