-
Notifications
You must be signed in to change notification settings - Fork 1
/
execute_cubic_instance_quantum_annealing.py
166 lines (151 loc) · 4.96 KB
/
execute_cubic_instance_quantum_annealing.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
161
162
163
164
165
import json
import networkx as nx
from dwave.cloud import Client
import time
import ast
import copy
from dwave import embedding
def Start_DWave_connection(device):
client = Client.from_config()
DWave_solver = client.get_solver(device)
A = DWave_solver.undirected_edges
connectivity_graph = nx.Graph(list(A))
return connectivity_graph, DWave_solver
def run(h, j, params):
while (True):
try:
sampleset = solver.sample_ising(h, j, answer_mode='raw', **params)
return sampleset.samples
except Exception as e:
print(e)
time.sleep(1)
continue
def adapt_native_embedding(embedding_dict_in):
embedding_dict = copy.deepcopy(embedding_dict_in)
out = {}
for a in embedding_dict:
out[a] = [embedding_dict[a]]
return out
def remove_zero_terms(h, J):
to_remove = []
for n in h:
if h[n] == 0.0:
to_remove.append(n)
to_remove_J = []
for e in J:
if J[e] == 0.0:
to_remove_J.append(e)
for n in to_remove:
del h[n]
for e in to_remove_J:
del J[e]
return h, J
def merge_cubic_terms(h_easy_parallel, J_easy_parallel, medium_parallel, Target, used_embeddings):
filtered_used_embeddings = []
for (parallel_emb_index, parallel_Ising) in enumerate(medium_parallel):
pre_h_easy_parallel = copy.deepcopy(h_easy_parallel)
pre_J_easy_parallel = copy.deepcopy(J_easy_parallel)
missing_nodes = []
missing_edges = []
valid_INDICATOR = True
for I in parallel_Ising:
lin = I[0]
quad = I[1]
modified_h = {}
for k in lin:
modified_h[k[0]] = lin[k]
for var in modified_h:
if Target.has_node(var) == False:
valid_INDICATOR = False
missing_nodes.append(var)
if var in h_easy_parallel:
weight = h_easy_parallel[var]
h_easy_parallel[var] = weight+modified_h[var]
else:
h_easy_parallel[var] = modified_h[var]
for edge in quad:
if Target.has_edge(*edge) == False:
valid_INDICATOR = False
missing_edges.append(edge)
IND1 = edge in J_easy_parallel
edge_flipped = tuple(reversed(edge))
IND2 = edge_flipped in J_easy_parallel
if IND1 == True:
weight = J_easy_parallel[edge]
J_easy_parallel[edge] = quad[edge]+weight
elif IND2 == True:
weight = J_easy_parallel[edge_flipped]
J_easy_parallel[edge_flipped] = quad[edge]+weight
else:
J_easy_parallel[edge] = quad[edge]
node_fails = []
edge_fails = []
for node in missing_nodes:
if h_easy_parallel[node] != 0:
node_fails.append(node)
for edge in missing_edges:
edge_flipped = tuple(reversed(edge))
IND1 = edge in J_easy_parallel
IND2 = edge_flipped in J_easy_parallel
if IND1 == True:
if J_easy_parallel[edge] != 0:
edge_fails.append(edge)
elif IND2 == True:
if J_easy_parallel[edge_flipped] != 0:
edge_fails.append(edge_flipped)
if len(node_fails+edge_fails) != 0:
valid_INDICATOR = False
else:
valid_INDICATOR = True
if valid_INDICATOR == False:
h_easy_parallel = pre_h_easy_parallel
J_easy_parallel = pre_J_easy_parallel
if valid_INDICATOR == True:
filtered_used_embeddings.append([used_embeddings[parallel_emb_index], medium_parallel[parallel_emb_index]])
h_easy_parallel, J_easy_parallel = remove_zero_terms(h_easy_parallel, J_easy_parallel)
return h_easy_parallel, J_easy_parallel, filtered_used_embeddings
file = open("parallel_embeddings/easy.txt", "r")
parallel_embeddings = ast.literal_eval(file.read())
file.close()
QA_device = "Advantage_system4.1"
Target, solver = Start_DWave_connection(QA_device)
#Annealing time and anneal schedule
AT = 100
s = 0.5
duration_frac = 0.5
heavy_hex_rep_idx = 0
duration = AT*duration_frac
start = (AT-duration)/2.0
end = ((AT-duration)/2.0)+duration
anneal_schedule = [[0, 0], [start, s], [end, s], [AT, 1]]
params = {"num_reads": 1000, "anneal_schedule": anneal_schedule}
IBMQ_device = "ibm_washington"
file = open("problem_instances/"+IBMQ_device+"_"+str(heavy_hex_rep_idx)+".txt", "r")
instance = ast.literal_eval(file.read())
file.close()
h = {}
for a in instance[0]:
h = {**h, **a}
J = {}
for a in instance[1]:
J = {**J, **a}
file = open("parallel_embeddings/medium_"+str(heavy_hex_rep_idx)+".txt", "r")
cubic_term_reductions = ast.literal_eval(file.read())[0]
file.close()
used_embeddings = []
parallel_embedded_ising_j = {}
parallel_embedded_ising_h = {}
for embedding_dict in parallel_embeddings:
try:
embedding_dict_adapt = adapt_native_embedding(embedding_dict)
embedded_h, embedded_j = embedding.embed_ising(h, J, embedding_dict_adapt, Target)
used_embeddings.append(embedding_dict)
parallel_embedded_ising_j = {**parallel_embedded_ising_j, **embedded_j}
parallel_embedded_ising_h = {**parallel_embedded_ising_h, **embedded_h}
except Exception as e:
pass
parallel_embedded_ising_h, parallel_embedded_ising_j, filtered_used_embeddings = merge_cubic_terms(parallel_embedded_ising_h, parallel_embedded_ising_j, cubic_term_reductions, Target, parallel_embeddings)
samples = run(parallel_embedded_ising_h, parallel_embedded_ising_j, params)
file = open("results.json", "w")
json.dump(samples, file)
file.close()