def triangulate(
corner_points,
max_refloops=20,
theta=0.8,
smooth_steps=50,
quality=0.9,
verbose=False,
**params,
):
from .criterion import avg_quality
from .initial import cdt
from .mark import adaptive_theta
from .refine import rgb
from .smooth import cpt
from .solve import laplace
m = process(
initial=cdt,
solve=laplace,
mark=adaptive_theta,
refine=rgb,
smooth=cpt,
criterion=avg_quality,
corner_points=corner_points,
max_refloops=max_refloops,
theta=theta,
smooth_steps=smooth_steps,
quality=quality,
verbose=verbose,
**params,
)
m.draw = types.MethodType(lambda self, **kwargs: draw(self, **kwargs), m)
m.show = types.MethodType(lambda self, **kwargs: show(self, **kwargs), m)
return m
if __name__ == "__main__":
from skfem.visuals.matplotlib import draw, plot, show
draw(m)
for itr in reversed(range(9)):
basis = Basis(m, e)
K = asm(laplace, basis)
M = asm(mass, basis)
#f = asm(load, basis)
I = m.interior_nodes()
Lh, eigen_uh = solve(*condense(K, M, I=I),
solver=solver_eigen_scipy_sym(sigma=20, k=1))
eigv = Lh[0]
print(eigv)
eigen_uh = eigen_uh.flatten()
if itr > 0:
m = m.refined(adaptive_theta(eval_estimator(m, eigen_uh, eigv)))
draw(m)
show()
if __name__ == "__main__":
draw(m)
plot(m, eigen_uh, shading='gouraud')
show()
