注釈
Click here to download the full example code
mset of the Henon map¶
[1]A. Shudo and K. S. Ikeda, Complex Trajectory Description for Chaotic Tunneling, Progress of Theoretical Physics Supplement 139, 246 (2000).
import numpy as np
import matplotlib.pyplot as plt
import warnings
warnings.filterwarnings('ignore')
#np.seterr(invalid='ignore')
twopi = 2*np.pi
def get_Mset(evolve, pinit, tmax, xr=[0,twopi], yr =[-3,3],sample=500):
qr = np.linspace(xr[0], xr[1], sample)
qi = np.linspace(yr[0], yr[1], sample)
if yr[0]*yr[1]<0:
qi = np.linspace(yr[0], 0, sample//2,endpoint=False)
qi = np.append(qi, np.linspace(0, yr[1], sample//2))
QR,QI = np.meshgrid(qr,qi)
q0 = QR + 1.j*QI
p0 = np.zeros(q0.shape, dtype=np.complex128) + pinit
x = np.array([np.copy(q0), np.copy(p0)])
initial = np.copy(q0)
x = evolves(x, tmax, evolve)
index = (np.abs(x[0]) > 1e100) & (np.abs(x[1]) > 1e100)
#x[0][index] = np.nan + 1.j*np.nan
#x[1][index] = np.nan + 1.j*np.nan
x[0][index] = np.inf
x[1][index] = np.inf
prod = x[1].imag*np.roll(x[1].imag,1)
index = (prod > 0)
return [initial, q0[~index], x]
def plot_contour(ax,levels, init, lset):
#vmax = lset[1][~np.isnan(lset[1])].imag.max()
ax.contour(init.real, init.imag, lset[1].imag, levels, colors='k',alpha=0.5,lw=1)
def fmap(x, c):
q,p = x
p = p - (c - q**2)
q = q + p #+ np.sin(p)
return np.array([q, p])
def evolves(x, tmax, evolve):
for i in range(tmax):
x = evolve(x)
return x
tmax = 6
c=0.8
evolve = lambda x: fmap(x, c)
init, mset, lset = get_Mset(evolve, 0, tmax, sample=2000, xr=[-2,2], yr=[-2,2])
fig, ax = plt.subplots(1,1)
ax.plot(mset.real, mset.imag, ',k')
levels = np.linspace(-2,2,20)
plot_contour(ax, levels, init, lset)
levels = [-np.inf, 0, np.inf]
colors = ["#00ff7f", "#00ffff"]
ax.contour(init.real, init.imag, lset[1].imag, levels, colors='k',linewidths=1)
ax.contourf(init.real, init.imag, lset[1].imag, levels, colors=colors,alpha=0.5, linewidths=3)
ax.set_xlabel(r"$\mathrm{Re}[x_0]$")
ax.set_ylabel(r"$\mathrm{Im}[x_0]$")
plt.show()
Total running time of the script: ( 0 minutes 1.905 seconds)