def main():
initial = np.array(list(product([0, 1, -1], [0, 1, -1])))[1:]
initial = normalize(initial)
initial = np.hstack((initial, np.zeros((initial.shape[0], 1))))
filename = "smooth_transition.html"
figure = animate_init_(initial, filename)
final = shear(initial)
generation = 1
figure = animate_next_(final, figure, filename, generation)
complete = False
while not complete:
center = np.average(final, axis=0)
final, complete = rotate_toward(center, np.array([1, 1, 1]), final)
figure = animate_next_(final, figure, filename, generation)
generation = generation + 1
complete = False
while not complete:
center = np.average(final, axis=0)
final, complete = rotate_toward(center, np.array([1, 0, 1]), final)
figure = animate_next_(final, figure, filename, generation)
generation = generation + 1
complete = False
while not complete:
center = np.average(final, axis=0)
final, complete = rotate_toward(center, np.array([0, 0, 1]), final)
figure = animate_next_(final, figure, filename, generation)
generation = generation + 1
complete = False
while not complete:
center = np.average(final, axis=0)
final, complete = rotate_toward(center, np.array([1, 0, 0]), final)
figure = animate_next_(final, figure, filename, generation)
generation = generation + 1
def plot_init_(self):
"""Initialize animation object. Return figure"""
obj = self.objectives
self.figure = animate_init_(obj, self.filename + ".html")
return self.figure
K1[j]=K[0] # The index of the solution with the minimal ASF value
K2[j]=asf_sort[0] # The ASF value related to the K1 index
no_redun_sol = list(dict.fromkeys(K1)) # Remove the redundant solutions (the solutions chosen several times for different weights) from the list K1
# Select pop_size solutions for next generation
if len(no_redun_sol) > pop_size:
no_redun_sol=no_redun_sol[0:pop_size]
## Sorting the individuals and original objective functions based on their ASF values
for i in range(pop_size):
objectives1 = np.vstack((objectives1, objectives[int(no_redun_sol[i])]))
individuals1 = np.vstack((individuals1, individuals[int(no_redun_sol[i])]))
if gen==0:
figure = animate_init_(objectives1, filename + ".html")
#figure= animate_next_(objectives1, figure, filename + ".html",gen)
else:
figure= animate_next_(objectives1, figure, filename + ".html",gen)
# if gen==20
# k-mean cluster Ns solutions
k_means = KMeans(NS)
solution_points = objectives1
k_means.fit(solution_points)
# Closest solutions to the center of each class are selected