def generate(args):
f = open(args.gen_log, 'w')
n_cnt = args.max_n - args.min_n + 1
problems_per_n = args.n_pairs * 1.0 / n_cnt
problems = []
batches = []
n_nodes_in_batch = 0
prev_n_vars = None
for n_var in range(args.min_n, args.max_n + 1):
lower_bound = int((n_var - args.min_n) * problems_per_n)
upper_bound = int((n_var - args.min_n + 1) * problems_per_n)
for problems_idx in range(lower_bound, upper_bound):
n_vars, iclauses, iclause_unsat, iclause_sat = gen_iclause_pair(
args, n_var)
if random.random() < 0.5:
iclauses.append(iclause_unsat)
else:
iclauses.append(iclause_sat)
n_clauses = len(iclauses)
n_cells = sum([len(iclause) for iclause in iclauses])
n_nodes = 2 * n_vars + n_clauses
if n_nodes > args.max_nodes_per_batch:
continue
batch_ready = False
if (args.one and len(problems) > 0):
batch_ready = True
elif (prev_n_vars and n_vars != prev_n_vars):
batch_ready = True
elif (not args.one
) and n_nodes_in_batch + n_nodes > args.max_nodes_per_batch:
batch_ready = True
if batch_ready:
batches.append(mk_batch_problem(problems))
print("batch %d done (%d vars, %d problems)..." %
(len(batches), prev_n_vars, len(problems)),
file=f)
del problems[:]
n_nodes_in_batch = 0
prev_n_vars = n_vars
is_sat, stats = solve_sat(n_vars, iclauses)
problems.append(("sr_n=%.4d_pk2=%.2f_pg=%.2f_t=%d_sat=0" %
(n_vars, args.p_k_2, args.p_geo, problems_idx),
n_vars, iclauses, is_sat))
n_nodes_in_batch += n_nodes
if len(problems) > 0:
batches.append(mk_batch_problem(problems))
print("batch %d done (%d vars, %d problems)..." %
(len(batches), n_vars, len(problems)),
file=f)
del problems[:]
return batches
n_nodes = 2 * n_vars + n_clauses
if n_nodes > opts.max_nodes_per_batch:
continue
batch_ready = False
if (opts.one and len(problems) > 0):
batch_ready = True
elif (prev_n_vars and n_vars != prev_n_vars):
batch_ready = True
elif (not opts.one
) and n_nodes_in_batch + n_nodes > opts.max_nodes_per_batch:
batch_ready = True
if batch_ready:
batches.append(mk_batch_problem(problems))
print("batch %d done (%d vars, %d problems)...\n" %
(len(batches), prev_n_vars, len(problems)))
del problems[:]
n_nodes_in_batch = 0
prev_n_vars = n_vars
print('Number of Variables =', n_vars, '\nClauses =', iclauses)
is_sat, is_time, stats = solve_sat(n_vars, iclauses)
print('Time =', is_time, 'secs')
avg_time.append(is_time)
print('Satisfiability =', is_sat, '\n')
problems.append((filename, n_vars, iclauses, is_time))
n_nodes_in_batch += n_nodes
def fun_generate(num=1, dir=None, dirout=None):
opts = Package()
problems = []
batches = []
n_nodes_in_batch = 0
if dir != None:
opts.dimacs_dir = dir
if dirout != None:
opts.out_dir = dirout
filenames = os.listdir(opts.dimacs_dir)
filenames = sorted(filenames)
prev_n_vars = None
for filename in filenames:
""" n_vars:变元数量，iclauses:[[1,2,-3],[?],[?]]
n_clauses:子句数量， n_cells:文字和，n_nodes：2 * n_vars + n_clauses
"""
print(filename)
n_vars, iclauses, answerpro = parse_dimacs("%s/%s" %
(opts.dimacs_dir, filename))
print(n_vars)
n_clauses = len(iclauses)
n_cells = sum([len(iclause) for iclause in iclauses])
n_nodes = 2 * n_vars + n_clauses
if n_nodes > opts.max_nodes_per_batch:
continue
batch_ready = False
opts.one = True # 一个公式一个的包装
if (opts.one and len(problems) > 0):
batch_ready = True
elif (prev_n_vars and n_vars != prev_n_vars):
# bianyuan shuliang bu tong
batch_ready = True
elif (not opts.one
) and n_nodes_in_batch + n_nodes > opts.max_nodes_per_batch:
# chao guo dan yuan shu
batch_ready = True
if batch_ready:
batches.append(mk_batch_problem(problems))
print("batch %d done (%d vars, %d problems)...\n" %
(len(batches), prev_n_vars, len(problems)))
del problems[:]
n_nodes_in_batch = 0
prev_n_vars = n_vars
# is_sat, stats = solve_sat(n_vars, iclauses)
# 默认为可满足
# print("----"+answerpro)
problems.append((filename, n_vars, iclauses, True))
n_nodes_in_batch += n_nodes
if len(problems) > 0:
batches.append(mk_batch_problem(problems))
print("batch %d done (%d vars, %d problems)...\n" %
(len(batches), n_vars, len(problems)))
del problems[:]
# create directory
if not os.path.exists(opts.out_dir):
os.mkdir(opts.out_dir)
dataset_filename = mk_dataset_filename(opts, len(batches))
print("Writing %d batches to %s...\n" % (len(batches), dataset_filename))
with open(dataset_filename, 'wb') as f_dump:
pickle.dump(batches, f_dump)