zarray.append(p.current_position[2])
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
fig = plt.figure()
ax = fig.add_subplot(111, projection='3d')
ax.scatter(xarray, yarray, zarray, zdir=u'z', s=20, c='b', marker='o')
plt.show()
# printParticles()
while i < max_iterations:
for j, p in enumerate(particles):
fitness = k_means.execute(p.current_position.tolist())
error = 1-fitness
if fitness > p.fitness:
p.fitness = fitness
p.best_position = p.current_position
if fitness > global_best.fitness:
global_best = p
v = p.velocity + velocity_parameter_1 * random.uniform(1, 10) * (p.best_position - p.current_position) \
+ velocity_parameter_2 * random.uniform(1, 10) * (global_best.current_position - p.current_position)
p.current_position = p.current_position + v
i += 1
__author__ = 'felipe'
from kmeans import KMeans
from random import uniform
km = KMeans()
for num_clusters in range(2, 31):
average_precision = 0
best_precision = 0
worst_precision = 9999999999999
for execs in range(0, 3532):
params = [num_clusters, ]
for j in range(0, num_clusters):
for i in range(0, km.dimens):
params.extend([uniform(km.mini[i], km.maxi[i])])
precision = km.execute(params)
average_precision += precision
best_precision = max(precision, best_precision)
worst_precision = min(precision, worst_precision)
# print(execs)
average_precision /= 3532
print(str(num_clusters) + "," + str(average_precision) + "," + str(best_precision) + "," + str(worst_precision))
