def main(_):
pp.pprint(flags.FLAGS.__flags)
if FLAGS.output_width is None:
FLAGS.output_width = FLAGS.output_height
assert (os.path.exists(FLAGS.checkpoint_dir))
run_config = tf.ConfigProto()
run_config.gpu_options.allow_growth = True
with tf.Session(config=run_config) as sess:
dcgan = DCGAN(sess,
output_width=FLAGS.output_width,
output_height=FLAGS.output_height,
batch_size=FLAGS.batch_size,
dataset_name=FLAGS.dataset,
checkpoint_dir=FLAGS.checkpoint_dir,
lam=FLAGS.lam)
#dcgan.load(FLAGS.checkpoint_dir):
dcgan.complete(FLAGS)
show_all_variables()
# to_json("./web/js/layers.js", [dcgan.h0_w, dcgan.h0_b, dcgan.g_bn0],
# [dcgan.h1_w, dcgan.h1_b, dcgan.g_bn1],
# [dcgan.h2_w, dcgan.h2_b, dcgan.g_bn2],
# [dcgan.h3_w, dcgan.h3_b, dcgan.g_bn3],
# [dcgan.h4_w, dcgan.h4_b, None])
# Below is codes for visualization
OPTION = 1
visualize(sess, dcgan, FLAGS, OPTION)
def main(_):
pp.pprint(flags.FLAGS.__flags)
if FLAGS.input_width is None:
FLAGS.input_width = FLAGS.input_height
if FLAGS.output_width is None:
FLAGS.output_width = FLAGS.output_height
if not os.path.exists(FLAGS.checkpoint_dir):
os.makedirs(FLAGS.checkpoint_dir)
if not os.path.exists(FLAGS.sample_dir):
os.makedirs(FLAGS.sample_dir)
#gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=0.333)
run_config = tf.ConfigProto()
run_config.gpu_options.allow_growth=True
with tf.Session(config=run_config) as sess:
dcgan = DCGAN(
sess,
input_width=FLAGS.input_width,
input_height=FLAGS.input_height,
output_width=FLAGS.output_width,
output_height=FLAGS.output_height,
batch_size=FLAGS.batch_size,
sample_num=FLAGS.batch_size,
dataset_name=FLAGS.dataset,
input_fname_pattern=FLAGS.input_fname_pattern,
crop=FLAGS.crop,
checkpoint_dir=FLAGS.checkpoint_dir,
sample_dir=FLAGS.sample_dir)
print('Will try to complete images...')
dcgan.complete()
print('Done!')
parser.add_argument('--lr', type=float, default=0.01)
parser.add_argument('--beta1', type=float, default=0.9)
parser.add_argument('--beta2', type=float, default=0.999)
parser.add_argument('--eps', type=float, default=1e-8)
parser.add_argument('--hmcBeta', type=float, default=0.2)
parser.add_argument('--hmcEps', type=float, default=0.001)
parser.add_argument('--hmcL', type=int, default=100)
parser.add_argument('--nIter', type=int, default=1000)
parser.add_argument('--imgSize', type=int, default=64)
parser.add_argument('--lam', type=float, default=0.1)
parser.add_argument('--checkpointDir', type=str, default='checkpoint')
parser.add_argument('--outDir', type=str, default='completions')
parser.add_argument('--outInterval', type=int, default=50)
parser.add_argument('--maskType', type=str,
choices=['random', 'center', 'left', 'full', 'grid', 'lowres'],
default='center')
parser.add_argument('--centerScale', type=float, default=0.25)
parser.add_argument('imgs', type=str, nargs='+')
args = parser.parse_args()
assert(os.path.exists(args.checkpointDir))
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
with tf.Session(config=config) as sess:
dcgan = DCGAN(sess, image_size=args.imgSize,
batch_size=(1 if args.approach == 'hmc' else 64),
checkpoint_dir=args.checkpointDir, lam=args.lam)
dcgan.complete(args)
parser.add_argument('--beta1', type=float, default=0.9)
parser.add_argument('--beta2', type=float, default=0.999)
parser.add_argument('--eps', type=float, default=1e-8)
parser.add_argument('--hmcBeta', type=float, default=0.2)
parser.add_argument('--hmcEps', type=float, default=0.001)
parser.add_argument('--hmcL', type=int, default=100)
parser.add_argument('--hmcAnneal', type=float, default=1)
parser.add_argument('--nIter', type=int, default=1000)
parser.add_argument('--imgSize', type=int, default=64)
parser.add_argument('--lam', type=float, default=0.1)
parser.add_argument('--checkpointDir', type=str, default='checkpoint')
parser.add_argument('--outDir', type=str, default='completions')
parser.add_argument('--outInterval', type=int, default=50)
parser.add_argument('--maskType', type=str,
choices=['random', 'center', 'left', 'full', 'grid', 'lowres'],
default='center')
parser.add_argument('--centerScale', type=float, default=0.25)
parser.add_argument('imgs', type=str, nargs='+')
args = parser.parse_args()
assert(os.path.exists(args.checkpointDir))
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
with tf.Session(config=config) as sess:
dcgan = DCGAN(sess, image_size=args.imgSize,
batch_size=min(64, len(args.imgs)),
checkpoint_dir=args.checkpointDir, lam=args.lam)
dcgan.complete(args)
def main(_):
pp.pprint(flags.FLAGS.__flags)
if not os.path.exists(FLAGS.checkpoint_dir):
os.makedirs(FLAGS.checkpoint_dir)
if not os.path.exists(FLAGS.sample_dir):
os.makedirs(FLAGS.sample_dir)
# Do not take all memory
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=0.80)
# sess = tf.Session(config=tf.ConfigProto(gpu_options=gpu_options))
with tf.Session(config=tf.ConfigProto(gpu_options=gpu_options)) as sess:
# w/ y label
if FLAGS.dataset == 'mnist':
dcgan = DCGAN(sess,
image_size=FLAGS.image_size,
batch_size=FLAGS.batch_size,
y_dim=10,
output_size=28,
c_dim=1,
dataset_name=FLAGS.dataset,
is_crop=FLAGS.is_crop,
checkpoint_dir=FLAGS.checkpoint_dir)
# w/o y label
else:
if FLAGS.dataset == 'cityscapes':
print 'Select CITYSCAPES'
mask_dir = CITYSCAPES_mask_dir
FLAGS.output_size_h, FLAGS.output_size_w, FLAGS.is_crop = 192, 512, False
FLAGS.dataset_dir = CITYSCAPES_dir
elif FLAGS.dataset == 'inria':
print 'Select INRIAPerson'
FLAGS.output_size_h, FLAGS.output_size_w, FLAGS.is_crop = 160, 96, False
FLAGS.dataset_dir = INRIA_dir
elif FLAGS.dataset == 'indoor':
print 'Select indoor'
FLAGS.output_size_h, FLAGS.output_size_w, FLAGS.is_crop = 256, 256, False
FLAGS.dataset_dir = indoor_dir
elif FLAGS.dataset == 'indoor_bedroom':
print 'Select indoor bedroom'
FLAGS.output_size_h, FLAGS.output_size_w, FLAGS.is_crop = 256, 256, False
FLAGS.dataset_dir = indoor_bedroom_dir
elif FLAGS.dataset == 'indoor_dining':
print 'Select indoor dining'
FLAGS.output_size_h, FLAGS.output_size_w, FLAGS.is_crop = 256, 256, False
FLAGS.dataset_dir = indoor_bedroom_dir
dcgan = DCGAN(sess,
batch_size=FLAGS.batch_size,
output_size_h=FLAGS.output_size_h,
output_size_w=FLAGS.output_size_w,
c_dim=FLAGS.c_dim,
dataset_name=FLAGS.dataset,
is_crop=FLAGS.is_crop,
checkpoint_dir=FLAGS.checkpoint_dir,
dataset_dir=FLAGS.dataset_dir)
if FLAGS.mode == 'test':
print('Testing!')
dcgan.test(FLAGS)
elif FLAGS.mode == 'train':
print('Train!')
dcgan.train(FLAGS)
elif FLAGS.mode == 'complete':
print('Complete!')
dcgan.complete(FLAGS, mask_dir)
'random', 'center', 'left', 'full', 'grid',
'lowres', 'parameters', 'wfc3'
],
default='parameters')
parser.add_argument('--input_spectrum',
type=str,
default='./input_spectrum.dat')
parser.add_argument('--centerScale', type=float, default=0.25)
parser.add_argument('--make_corner', type=bool, default=False)
parser.add_argument('--spectra_int_norm', type=bool, default=False)
parser.add_argument('--spectra_norm', type=bool, default=False)
parser.add_argument('--spectra_real_norm', type=bool, default=True)
args = parser.parse_args()
assert (os.path.exists(args.checkpointDir))
tf.reset_default_graph()
config = tf.ConfigProto(log_device_placement=True)
config.gpu_options.allow_growth = True
sess = tf.Session(config=config)
spectrum = args.input_spectrum
dcgan = DCGAN(sess,
image_size=args.imgSize,
z_dim=100,
batch_size=64,
checkpoint_dir=args.checkpointDir,
c_dim=1,
lam=args.lam)
dcgan.complete(args, spectrum)
parser.add_argument('--nIter', type=int, default=1001)
parser.add_argument('--imgSize', type=int, default=33)
parser.add_argument('--lam', type=float, default=0.1)
parser.add_argument('--checkpointDir', type=str, default='checkpoint_test')
parser.add_argument('--outDir', type=str, default='exogan_output')
parser.add_argument('--outInterval', type=int, default=50)
parser.add_argument('--maskType', type=str,
choices=['random', 'center', 'left', 'full',
'grid', 'lowres', 'parameters', 'wfc3'],
default='parameters')
parser.add_argument('--centerScale', type=float, default=0.25)
parser.add_argument('--make_corner', type=bool, default=False)
parser.add_argument('--spectra_int_norm', type=bool, default=False)
parser.add_argument('--spectra_norm', type=bool, default=False)
parser.add_argument('--spectra_real_norm', type=bool, default=True)
args = parser.parse_args()
assert(os.path.exists(args.checkpointDir))
tf.reset_default_graph()
config = tf.ConfigProto(log_device_placement=True)
config.gpu_options.allow_growth = True
sess = tf.Session(config=config)
dcgan = DCGAN(sess,
image_size=args.imgSize,
z_dim=100,
batch_size=64,
checkpoint_dir=args.checkpointDir,
c_dim=1,
lam=args.lam)
dcgan.complete(args, spectrum[0], sigma=0.0)
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
with tf.Session(config=config) as sess:
dcgan = DCGAN(sess,
image_size=args.imgSize,
batch_size=min(64, len(args.imgs)),
checkpoint_dir=args.checkpointDir,
lam=args.lam)
img1 = cv2.imread("C:/Users/engab/Desktop/PROJECT/CRW_4901_JFRtamp37.jpg",
cv2.IMREAD_COLOR)
points, positions = sift.all_experiments(img1)
while True:
dcgan.complete(args, points, positions, 110,
175) #put the starting coordinates of the object
output = cv2.imread(
"C:/Users/engab/Desktop/PROJECT/tensorflow dcgan inpainting/DSC_1535tamp1.jpg",
cv2.IMREAD_COLOR) #inpainted image path
img1[175:239, 110:174] = output
points, positions = sift.all_experiments(img1)
#your code here,put image number 950 in the loop folder
parser.set_defaults(
imgs=
"C:/Users\\engab\\Desktop\\PROJECT\\tensorflow dcgan inpainting\\data\\loop"
) #inpainted image path
args = parser.parse_args()
if points < 4:
break
