def sched(s, to_load, graph, n):
if to_load == []:
param = find_params(graph, n, s)
obj = param[0].primal
if best == []:
best.append(obj)
sbest.append(s)
elif obj < best[-1]:
best.append(obj)
sbest.append(s)
else:
for v in to_load:
if (type(v) == tuple and v[0] in s and v[1] in s) or type(v) == int:
t = deepcopy(s)
t.append(v)
branch = [x for x in to_load if x != v]
sched(t, branch, graph, n)
def sched(graph, n):
m = len(graph)
inv_graph = invert_dict(graph)
for v in permutations(range(1, n+1)):
for e in permutations(range(1, m+1)):
s = [v[i] for i in range(n)]
edg = [inv_graph[e[i]] for i in range(m)]
s = s + edg
param = find_params(graph, n, s)
obj = param[0].primal
if best == []:
best.append(obj)
sbest.append(s)
print best
print sbest
elif obj < best[-1]:
best.append(obj)
sbest.append(s)
print best
print sbest
four_clique = {(1,2):1, (1,3):2, (1,4):3, (2,3):4, (2,4):5, (3,4):6}
####################
if __name__ == '__main__':
# set the input graph here
graph = four_clique
# the next few lines compute the number of vertices
vertices = []
for e in graph:
if e[0] not in vertices:
vertices.append(e[0])
if e[1] not in vertices:
vertices.append(e[1])
n = len(vertices)
to_load = vertices + graph.keys()
# best is a list to hold the progression of best costs found
best = []
# sbest is a list to hold the progression of corresponding best schedules
sbest = []
sched(graph, n)
print 'progression of best objective values and accompanying schedule'
print best
print sbest
print
print 'Best schedule found is ', sbest[-1]
print 'Parameters for this schedule are'
param = find_params(graph, n, sbest[-1])
print_params(graph, n, param)