LR = FLAGS.lr MOMENTUM = FLAGS.momentum ROOT_PATH = os.path.dirname(os.path.realpath(__file__)) LOG_PATH = os.path.join(ROOT_PATH, FLAGS.log_dir) if not os.path.exists(LOG_PATH): os.mkdir(LOG_PATH) acc_count = 0 while True: if os.path.exists(os.path.join(LOG_PATH, 'log_%02d.txt' % acc_count)): acc_count += 1 else: break LOG_FNAME = 'log_%02d.txt' % acc_count LOG_FOUT = open(os.path.join(LOG_PATH, LOG_FNAME), 'w') (train_images, train_labels, train_att), train_iters = data.data_train(REAL_PATH, TRAIN_LABEL, BATCH_SIZE) (fake_images, fake_labels, fake_att), fake_iters = data.data_train(FAKE_PATH, TRAIN_LABEL, BATCH_SIZE) (valid_images, valid_labels, valid_att), valid_iters = data.data_test(REAL_PATH, VALID_LABEL, BATCH_SIZE * 10) (test_images, test_labels, test_att), test_iters = data.data_test(REAL_PATH, TEST_LABEL, BATCH_SIZE) #################################################### def log_string(out_str): LOG_FOUT.write(out_str + '\n') LOG_FOUT.flush() print(out_str)
VALID_LABEL = FLAGS.valid_label BATCH_SIZE = FLAGS.batch_size N_SAMPLE = FLAGS.sample_size N_EPOCH = FLAGS.n_epoch N_ADV = FLAGS.n_adv N_CLASS = FLAGS.n_class LR = FLAGS.lr MOMENTUM = FLAGS.momentum ROOT_PATH = os.path.dirname(os.path.realpath(__file__)) LOG_PATH = os.path.join(ROOT_PATH, FLAGS.log_dir) OUT_PATH = os.path.join(ROOT_PATH, FLAGS.output_dir) if not os.path.exists(LOG_PATH): os.mkdir(LOG_PATH) if not os.path.exists(OUT_PATH): os.mkdir(OUT_PATH) (train_images, train_labels), train_iters = data.data_train(IMG_PATH, TRAIN_LABEL, BATCH_SIZE) (valid_images, valid_labels), valid_iters = data.data_test(IMG_PATH, VALID_LABEL, N_SAMPLE) Genc = model.Genc() Gdec = model.Gdec() D = model.D() Adv = model.Adv_cls() #################################################### def V_graph(sess, phv): real_labels = valid_labels * 2 - 1 fake_labels = -real_labels u = Genc.build(valid_images, phv['is_training_v'])
'Wavy_Hair': 33, 'Wearing_Earrings': 34, 'Wearing_Hat': 35, 'Wearing_Lipstick': 36, 'Wearing_Necklace': 37, 'Wearing_Necktie': 38, 'Young': 39 } os.environ["CUDA_DEVICE_ORDER"] = "PCI_BUS_ID" os.environ["CUDA_VISIBLE_DEVICES"] = FLAGS.gpu # tf.set_random_seed(0)# 0 for 512 tf.set_random_seed(100) (train_images, train_labels, train_att), train_iters = data.data_train(FLAGS.real_path, FLAGS.train_label, 64) (fake_images, fake_labels, fake_att), fake_iters = data.data_fake(FLAGS.fake_path, FLAGS.train_label, 64) (valid_images, valid_labels, valid_att), valid_iters = data.data_test(FLAGS.real_path, FLAGS.valid_label, FLAGS.batch_size) (test_images, test_labels, test_att), test_iters = data.data_test(FLAGS.real_path, FLAGS.test_label, FLAGS.batch_size) batch_images = tf.placeholder(tf.float32, [None, 128, 128, 3]) batch_labels = tf.placeholder(tf.int32, [ None, ]) is_training = tf.placeholder(tf.bool) lr_ph = tf.placeholder(tf.float32)
N_EPISODE = FLAGS.n_episode N_CLASS = FLAGS.n_class N_ACTION = FLAGS.n_action LR = FLAGS.lr MOMENTUM = FLAGS.momentum ROOT_PATH = os.path.dirname(os.path.realpath(__file__)) LOG_PATH = os.path.join(ROOT_PATH, FLAGS.log_dir) if not os.path.exists(LOG_PATH): os.mkdir(LOG_PATH) acc_count = 0 while True: if os.path.exists(os.path.join(LOG_PATH, 'log_%02d.txt' % acc_count)): acc_count += 1 else: break LOG_FNAME = 'log_%02d.txt' % acc_count LOG_FOUT = open(os.path.join(LOG_PATH, LOG_FNAME), 'w') (train_images, train_labels, train_att), train_iters = data.data_train(REAL_PATH, TRAIN_LABEL, BATCH_SIZE) (fake_images, fake_labels, fake_att), fake_iters = data.data_train(FAKE_PATH, TRAIN_LABEL, BATCH_SIZE) (valid_images, valid_labels, valid_att), valid_iters = data.data_test(REAL_PATH, VALID_LABEL, BATCH_SIZE) (test_images, test_labels, test_att), test_iters = data.data_test(REAL_PATH, TEST_LABEL, BATCH_SIZE) #################################################### def log_string(out_str): LOG_FOUT.write(out_str+'\n') LOG_FOUT.flush() print(out_str) def choose_action(prob_actions): actions = [] for i in range(prob_actions.shape[0]): action = np.random.choice(range(prob_actions.shape[1]), p=prob_actions[i])