/
sqa.py
53 lines (45 loc) · 1.26 KB
/
sqa.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
# -*- coding:utf-8 -*-
"""Utilities of Simulated [Quantum] Annealing.
"""
from numpy import *
#run sqa algorithm with params
def run(kT,Ginit,m,mat,rep):
"""run simulated QA.
@param kT
@param Ginit
@param m
@param mat
@param rep
@return array([qarr,Earr])
"""
h = mat[0]
J = mat[1]
c = mat[2]
N = len(h)
tau = 0.99
Gfin = 0.01
qarr = []
Earr = []
# simulated quantum annealing simulator using quantum monte carlo & metropolis
for j in range(rep):
G = Ginit
q = []
for i in range(m):
q.append(random.choice([-1,1],N))
while G > Gfin:
for i in range(N*m):
x = random.randint(N)
y = random.randint(m)
dE = (2*q[y][x]*(h[x]+q[y][(N+x-1)%N]*J[x][(N+x-1)%N]+q[y][(x+1)%N]*J[x][(x+1)%N]))*1.0/m
dE += -q[y][x]*(q[(m+y-1)%m][x]+q[(y+1)%m][x])*log(tanh(G/kT/m))*1.0/kT
if exp(-dE/kT)>random.rand():
q[y][x] = -q[y][x]
G*=tau
E = 0
for a in range(N):
E += h[a]*q[0][a]
for b in range(a+1,N):
E += J[a][b]*q[0][a]*q[0][b]
qarr.append(q)
Earr.append(E+c)
return(array([qarr,Earr]))