def _get_arguments(argv):
parser = argparse.ArgumentParser()
config.add_arguments(parser)
models.add_arguments(parser)
solver.add_arguments(parser)
loss_metrics.add_arguments(parser)
input_pipeline.add_arguments(parser)
custom_evaluator.add_arguments(parser)
args = parser.parse_args(argv[1:])
config.check_args(args, parser)
config.fill_default_args(args)
return args
def main():
parser = argparse.ArgumentParser()
config.add_arguments(parser)
args = parser.parse_args()
config.fill_default(args)
tsp_spec = data_kits.load(args.tsp_file, args.sln_file)
# data_kits.show_path(tsp_spec, tsp_spec.solutions)
# return
args.ants = config.maybe_fill(args.ants, tsp_spec.dimension)
print(args)
num_ants = args.ants
num_iters = args.iters
num_rep = args.repeat
g = graph.TSP(tsp_spec, args=args)
best_path = None
best_cost = sys.maxsize
best_cost_iters = []
# Animation
player = data_kits.DynamicShow(tsp_spec)
player.launch(args)
# Main loop
for _ in tqdm.tqdm(range(num_rep), ascii=True):
ant_col = colony.AntColony(g, num_ants, num_iters, args)
ant_col.begin(player)
if ant_col.best_path_cost < best_cost:
best_path = ant_col.best_path
best_cost = ant_col.best_path_cost
best_cost_iters.append(ant_col.iter_costs)
best_path = data_kits.adj_path(best_path)
print("Best path:", best_path)
print("Best cost:", best_cost)
# plt.subplot(121)
# data_kits.show_path(tsp_spec, [best_path])
# plt.subplot(122)
# data_kits.show_iters(best_cost_iters)
#
plt.ioff()
plt.show()
hparams = config.create_hparams(FLAGS)
# Source vocab
src_vocab_file = "%s.%s" % (hparams.vocab_prefix, hparams.src)
tgt_vocab_file = "%s.%s" % (hparams.vocab_prefix, hparams.tgt)
print (src_vocab_file, tgt_vocab_file)
src_vocab_size, src_vocab_file = vocab_table_util.check_vocab(
src_vocab_file,
hparams.out_dir,
sos=hparams.sos,
eos=hparams.eos,
unk=vocab_table_util.UNK)
# Target vocab
tgt_vocab_size, tgt_vocab_file = vocab_table_util.check_vocab(
tgt_vocab_file,
hparams.out_dir,
sos=hparams.sos,
eos=hparams.eos,
unk=vocab_table_util.UNK)
hparams.add_hparam("src_vocab_size", src_vocab_size)
hparams.add_hparam("tgt_vocab_size", tgt_vocab_size)
hparams.add_hparam("src_vocab_file", src_vocab_file)
hparams.add_hparam("tgt_vocab_file", tgt_vocab_file)
infer(hparams)
pass
if __name__ == "__main__":
nmt_parser = argparse.ArgumentParser()
config.add_arguments(nmt_parser)
FLAGS, unparsed = nmt_parser.parse_known_args()
tf.app.run(main=main, argv=[sys.argv[0]] + unparsed)
def get_args():
parser = argparse.ArgumentParser()
parser = add_arguments(parser)
return parser.parse_args()
decoder_reference_list,
decoder_prediction_list,
cls_logits,
cls_orig_logits,
cls_labels,
vocab,
sent_embs,
adv_sent_embs,
is_test=True,
orig_alphas=orig_alphas,
trans_alphas=trans_alphas,
cls_logits_def=cls_logits_def,
cls_origs_def=cls_origs_def)
def main(args):
if args.do_train:
train(args)
if args.do_test:
test(args)
if args.do_cond_test:
test_adv_pos_neg(args)
if __name__ == '__main__':
args = config.add_arguments()
main(args)