-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathoptimization_plot.py
160 lines (133 loc) · 7.15 KB
/
optimization_plot.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
from logging import warn
import matplotlib.pyplot as plt
import numpy as np
from qaoa_xy_mixer import qaoa_xy
import matplotlib.ticker as ticker
from noise_models import create_noise_model_random_x
import qiskit.providers.aer.noise as noise
plt.rcParams.update({
"text.usetex": True,
"font.family": "serif"})
##########################################################################################
##################### ###################### QAOA optimization ###########################
##########################################################################################
cities_no = [3, 4]
layers_no = 8
tsp_instances = 40
exp_no = 1
gamma = 0.002
noise_model_str = "randx"+str(gamma)
layer_init = 8
repetition_qaoa_per_layer = 3
scheme = "01"
model = "xy"
noise_model = create_noise_model_random_x(gamma)
y_min = {3: -0.005, 4: -0.005}
y_max = {3: .06, 4: 0.09}
def opt_plot(ax, data_en_pure, data_en_nopost, data_en_post, city_no):
data_en_pure = np.asarray(data_en_pure)
data_en_nopost = np.asarray(data_en_nopost)
data_en_post = np.asarray(data_en_post)
for ind, en in enumerate(data_en_post):
if en == 0.:
print(ind, en)
assert np.max(data_en_pure) < y_max[city_no], "should be {} {}".format(np.max(data_en_pure), y_max[city_no])
assert np.max(data_en_post) < y_max[city_no], "should be {} {}".format(np.max(data_en_post), y_max[city_no])
assert np.max(data_en_nopost) < y_max[city_no] , "should be {} {}".format(np.max(data_en_nopost), y_max[city_no])
assert y_min[city_no] < np.min(data_en_pure), "{} should be {}".format(y_min[city_no], np.min(data_en_pure))
assert y_min[city_no] < np.min(data_en_nopost), "{} should be {}".format(y_min[city_no], np.min(data_en_nopost))
assert y_min[city_no] < np.min(data_en_post), "{} should be {}".format(y_min[city_no], np.min(data_en_post))
# print(data_en_post/data_en_nopost)
print(data_en_post)
ax.scatter(data_en_pure, data_en_post, c="r", marker="+", s=10, lw=0.5, label="mid")
ax.scatter(data_en_pure, data_en_nopost, c="k", marker="x", s=6, lw=0.5, label="no-mid")
# ax.hlines(np.min(data_en_nopost), y_min[city_no], y_max[city_no], linestyle="-", color="r", linewidth=.2)
f_nopost = dict()
f_post = dict()
normalization = "full"
f_pure = lambda x, diag_vals, hamildict: np.real(qaoa_xy(x, noise.NoiseModel(), diag_vals, hamiltonian=hamildict,normalization=normalization)[0])
f_nopost = lambda x, diag_vals, hamildict: np.real(qaoa_xy(x, noise_model, diag_vals, hamiltonian=hamildict,normalization=normalization)[0])
f_post = lambda x, diag_vals, hamildict: np.real(qaoa_xy(x, noise_model, diag_vals, midmeasure='yes', hamiltonian=hamildict, scheme=scheme, normalization=normalization)[0])
fig, axes = plt.subplots(2, 3, figsize=(6,4), sharex="row", sharey="row")
data_x = range(1, tsp_instances+1)
for city_ind, city_no in enumerate(cities_no):
print(city_no)
data_en_pure = []
data_en_nopost = []
for m in range(tsp_instances):
filename = "data_noise_qaoa_xy/optimization/tsp{}-model{}-noise{}-exp{}-scheme{}".format(city_no, model, noise_model_str, m, scheme)
hamildict = np.load('./tsp_data_xy/tsp{}/tsp_dict_{}_{}.npz'.format(city_no, model, m+1))
for j in [0,2]:
for l in range(len(hamildict[:,j])):
if hamildict[:,j][l] > 0:
hamildict[:,j][l] -= 1
diag_vals = np.load('./tsp_data_xy/tsp{}/qubo_{}.npz'.format(city_no, m+1))
angle = np.load(filename + "_pure.npy")
data_en_pure.append(f_pure(angle, diag_vals, hamildict))
angle = np.load(filename + "_nopost.npy")
data_en_nopost.append(f_nopost(angle, diag_vals, hamildict))
data_en_post = []
for m in range(tsp_instances):
filename = "data_noise_qaoa_xy/optimization/tsp{}-model{}-noise{}-exp{}-scheme{}".format(city_no, model, noise_model_str, m, scheme)
hamildict = np.load('./tsp_data_xy/tsp{}/tsp_dict_{}_{}.npz'.format(city_no, model, m+1))
for j in [0,2]:
for l in range(len(hamildict[:,j])):
if hamildict[:,j][l] > 0:
hamildict[:,j][l] -= 1
diag_vals = np.load('./tsp_data_xy/tsp{}/qubo_{}.npz'.format(city_no, m+1))
angle = np.load(filename + "_nopost.npy")
data_en_post.append(f_post(angle, diag_vals, hamildict))
opt_plot(axes[city_ind,0], data_en_pure, data_en_nopost, data_en_post, city_no)
data_en_post = []
for m in range(tsp_instances):
filename = "data_noise_qaoa_xy/optimization/tsp{}-model{}-noise{}-exp{}-scheme{}".format(city_no, model, noise_model_str, m, scheme)
hamildict = np.load('./tsp_data_xy/tsp{}/tsp_dict_{}_{}.npz'.format(city_no, model, m+1))
for j in [0,2]:
for l in range(len(hamildict[:,j])):
if hamildict[:,j][l] > 0:
hamildict[:,j][l] -= 1
diag_vals = np.load('./tsp_data_xy/tsp{}/qubo_{}.npz'.format(city_no, m+1))
angle = np.load(filename + "_post.npy")
data_en_post.append(f_post(angle, diag_vals, hamildict))
opt_plot(axes[city_ind,1], data_en_pure, data_en_nopost, data_en_post, city_no)
data_en_post = []
for m in range(tsp_instances):
filename = "data_noise_qaoa_xy/optimization/tsp{}-model{}-noise{}-exp{}-scheme{}".format(city_no, model, noise_model_str, m, scheme)
hamildict = np.load('./tsp_data_xy/tsp{}/tsp_dict_{}_{}.npz'.format(city_no, model, m+1))
for j in [0,2]:
for l in range(len(hamildict[:,j])):
if hamildict[:,j][l] > 0:
hamildict[:,j][l] -= 1
diag_vals = np.load('./tsp_data_xy/tsp{}/qubo_{}.npz'.format(city_no, m+1))
angle = np.load(filename + "_nopost_post.npy")
data_en_post.append(f_post(angle, diag_vals, hamildict))
opt_plot(axes[city_ind,2], data_en_pure, data_en_nopost, data_en_post, city_no)
lgnd = axes[1,2].legend(loc=4, borderpad=0.2)
for handle in lgnd.legendHandles:
handle.set_sizes([15.0])
# axes[0].set_ylim(10e-2, 2)
axes[0,0].set_ylabel("noisy energy\n3 cities")
axes[1,0].set_ylabel("noisy energy\n4 cities")
axes[0,0].set_title("error mitigation")
axes[0,1].set_title("optimization")
axes[0,2].set_title("two-step opt.")
axes[1,0].set_xlabel("pure energy")
axes[1,1].set_xlabel("pure energy")
axes[1,2].set_xlabel("pure energy")
for ax_ind, ax in enumerate(axes.flatten()):
ax.set_aspect('equal')
if ax_ind < 3:
city_no = 3
else:
city_no = 4
ax.plot([y_min[city_no], y_max[city_no]], [y_min[city_no], y_max[city_no]], 'k-', alpha=0.75, zorder=0, linewidth=.2)
# ax.plot([y_min[city_no], y_max[city_no]], [0.85*y_min[city_no], 0.85*y_max[city_no]], 'r-', alpha=0.75, zorder=0, linewidth=.2)
ax.yaxis.set_major_formatter(ticker.FuncFormatter(lambda y, _: '{:.3f}'.format(y)))
ax.xaxis.set_major_formatter(ticker.FuncFormatter(lambda y, _: '{:.3f}'.format(y)))
ax.set_axisbelow(True)
for axis in [ax.xaxis, ax.yaxis]:
axis.grid(True, which="both", linewidth=.1)
ax.set_ylim(y_min[city_no], y_max[city_no])
ax.set_xlim(y_min[city_no], y_max[city_no])
plt.subplots_adjust()
plt.savefig("plots/qaoa_optimization.pdf", bbox_inches="tight")