def test_train():
with tempfile.TemporaryDirectory() as tmp_dir:
train_config = train.TrainConfig(
train_image_dir_path=(Path(__file__).parent /
'dammy_image_data').resolve(),
max_iter=2,
batch_size=5,
snapshot_iter_interval=2,
display_iter_interval=1,
n_discriminator_update=3,
evaluation_iter_interval=1)
model_config = ModelConfig(width=64,
height=64,
n_units_xyrz=10,
n_hidden_units=[10, 10],
z_size=2)
train.train(Path(tmp_dir), train_config, model_config)
if not argslist.use_lstm:
argslist.history_length = 1
os.environ['CUDA_VISIBLE_DEVICES'] = '1'
params = ["reward_type", "history_length", "shared_lstm", "batch_size", "num_task", "train_data_name",
"buffer_size", "max_episode_len", "save_rate", "gamma", "num_units"]
save_path = "policy"
dict_arg = vars(argslist)
for param in params:
save_path = save_path + "_" + param + "_" + str(dict_arg[param])
save_path += "_UAVnumber_" + str(FLAGS.num_uav) + "_size_map_" + str(FLAGS.size_map) + "_radius_" + str(FLAGS.radius)
argslist.save_dir = argslist.save_dir + save_path + "_debug/"
print(argslist.save_dir)
# train
if argslist.train:
train(argslist)
if argslist.transfer_train:
transfer_train(argslist, 300)
# train test
if argslist.train_test:
argslist.draw_picture_test = True
if argslist.mp:
train_multi_process_test(argslist)
else:
train_test(argslist, int(396/argslist.save_rate))
# transfer test
if argslist.transfer_test:
argslist.draw_picture_test = True
