def on_epoch_end (self, epoch, logs=None):
"""Calls model_weights_to_images and writes the images to
self.dir\<layer_name>_<epoch number>.png"""
images = WeightWriter.model_weights_to_images (self.model)
# write the images
for name, image in images:
save_image (join (self.dir, ".".join ([name, str (epoch).zfill (4), "png"])), image)
def display_current_results(self, visuals, size, epoch, iteration):
if self.tboard:
img_summaries = []
for label, image_numpy in visuals.items():
self.writer.add_image("size-{}-epoch-{}-iter-{}-{}.jpg".format(
size, epoch, iteration, label),
image_numpy,
dataformats='HWC')
if self.use_html:
for label, image_numpy in visuals.items():
if isinstance(image_numpy, list):
for i in range(len(image_numpy)):
img_path = os.path.join(
self.img_dir,
"size-{}-epoch-{}-iter-{}-{}_{}.jpg".format(
size, epoch, iteration, label, i))
misc.save_image(image_numpy[i], img_path)
else:
img_path = os.path.join(
self.img_dir, "size-{}-epoch-{}-iter-{}-{}.jpg".format(
size, epoch, iteration, label))
misc.save_image(image_numpy, img_path)
# update webpage
webpage = html.HTML(self.web_dir,
"Experiment name = {}".format(self.name),
refresh=30)
images = sorted(os.listdir(self.img_dir), reverse=True)
webpage.add_header("Training process visulization")
ims = []
txts = []
links = []
for image in images:
ims.append(image)
links.append(image)
label = image.split('.')[0].split('_')[0].split('-')[-1]
label += ": " + image.split('.')[0].replace('_', '@')
txts.append(label)
if len(ims) < 10:
webpage.add_images(ims, txts, links, width=self.win_size)
else:
num = int(round(len(ims) / 2.0))
webpage.add_images(ims[:num],
txts[:num],
links[:num],
width=self.win_size)
webpage.add_images(ims[num:],
txts[num:],
links[num:],
width=self.win_size)
webpage.save()
def encode_patchwork(inp, out, msg):
"""Hides a text message in pixels of two key streams."""
assert os.path.isfile(inp), '%s is not a file.' % inp
img = read_img(inp)
A = get_random_pos(len(msg), img)
B = get_random_pos(len(msg), img, A)
print "A: %s" % json.dumps(A)
print "B: %s" % json.dumps(B)
#while writing an jpg image, compression destroys the steganographic message
ext = os.path.splitext(out)[1].lower()
assert ext != '.jpeg' and ext != '.jpg', 'jpg/jpeg is currently not a valid extension for output images.'
save_image(out, encode_msg_with_patchwork(img, msg, A, B))
def encode_patchwork(inp, out, msg):
"""Hides a text message in pixels of two key streams."""
assert os.path.isfile(inp), '%s is not a file.' % inp
img = read_img(inp)
A = get_random_pos(len(msg), img)
B = get_random_pos(len(msg), img, A)
print "A: %s" %json.dumps(A)
print "B: %s" %json.dumps(B)
#while writing an jpg image, compression destroys the steganographic message
ext = os.path.splitext(out)[1].lower()
assert ext != '.jpeg' and ext != '.jpg', 'jpg/jpeg is currently not a valid extension for output images.'
save_image(out, encode_msg_with_patchwork(img, msg, A, B))
def on_epoch_end(self, epoch, logs=None):
"""Run prediction over n items,
store inputs and outputs in <dir>/<epoch_number>,
store scores in <dir>/<epoch_number>/scores.csv"""
epoch_path = join(self.dir, str(epoch).zfill(4))
try:
makedirs(epoch_path)
except IOError:
pass
for i in range(0, self.count):
X, _ = next(self.data)
for j in range(0, X.shape[0]):
data_path = join(epoch_path, "%04i_%04i_in.png" % (i, j))
save_image(data_path, X[j, :, :, 0])
Y = self.model.predict(X, batch_size=X.shape[0])
# for multiple outputs, assume the first output is an image...
if type(Y) is type([]):
for idx, y in enumerate(Y):
for j in range(0, y.shape[0]):
pred_path = join(
epoch_path,
"%04i_%04i_pred_%02i.png" % (i, j, idx))
save_image(pred_path, y[j, :, :, 0])
else:
for j in range(0, Y.shape[0]):
pred_path = join(epoch_path, "%04i_%04i_pred.png" % (i, j))
save_image(pred_path, Y[j, :, :, 0])
def write_img(path, img_array):
image = img_array.reshape(img_array.shape)
save_image(path, image)
def generate_fake_images(run_id,
snapshot=None,
grid_size=[1, 1],
batch_size=8,
num_pngs=1,
image_shrink=1,
png_prefix=None,
random_seed=1000,
minibatch_size=8):
network_pkl = misc.locate_network_pkl(run_id, snapshot)
if png_prefix is None:
png_prefix = misc.get_id_string_for_network_pkl(network_pkl) + '-'
random_state = np.random.RandomState(random_seed)
print('Loading network from "%s"...' % network_pkl)
G, D, Gs = misc.load_network_pkl(run_id, snapshot)
lsfm_model = m3io.import_lsfm_model(
'/home/baris/Projects/faceganhd/models/all_all_all.mat')
lsfm_tcoords = \
mio.import_pickle('/home/baris/Projects/team members/stelios/UV_spaces_V2/UV_dicts/full_face/512_UV_dict.pkl')[
'tcoords']
lsfm_params = []
result_subdir = misc.create_result_subdir(config.result_dir, config.desc)
for png_idx in range(int(num_pngs / batch_size)):
start = time.time()
print('Generating png %d-%d / %d... in ' %
(png_idx * batch_size, (png_idx + 1) * batch_size, num_pngs),
end='')
latents = misc.random_latents(np.prod(grid_size) * batch_size,
Gs,
random_state=random_state)
labels = np.zeros([latents.shape[0], 0], np.float32)
images = Gs.run(latents,
labels,
minibatch_size=minibatch_size,
num_gpus=config.num_gpus,
out_shrink=image_shrink)
for i in range(batch_size):
if images.shape[1] == 3:
mio.export_pickle(
images[i],
os.path.join(
result_subdir,
'%s%06d.pkl' % (png_prefix, png_idx * batch_size + i)))
# misc.save_image(images[i], os.path.join(result_subdir, '%s%06d.png' % (png_prefix, png_idx*batch_size+i)), [0,255], grid_size)
elif images.shape[1] == 6:
mio.export_pickle(images[i][3:6],
os.path.join(
result_subdir, '%s%06d.pkl' %
(png_prefix, png_idx * batch_size + i)),
overwrite=True)
misc.save_image(
images[i][0:3],
os.path.join(
result_subdir,
'%s%06d.png' % (png_prefix, png_idx * batch_size + i)),
[-1, 1], grid_size)
elif images.shape[1] == 9:
texture = Image(np.clip(images[i, 0:3] / 2 + 0.5, 0, 1))
mesh_raw = from_UV_2_3D(Image(images[i, 3:6]))
normals = images[i, 6:9]
normals_norm = (normals - normals.min()) / (normals.max() -
normals.min())
mesh = lsfm_model.reconstruct(mesh_raw)
lsfm_params.append(lsfm_model.project(mesh_raw))
t_mesh = TexturedTriMesh(mesh.points, lsfm_tcoords.points,
texture, mesh.trilist)
m3io.export_textured_mesh(
t_mesh,
os.path.join(result_subdir,
'%06d.obj' % (png_idx * minibatch_size + i)),
texture_extension='.png')
mio.export_image(
Image(normals_norm),
os.path.join(
result_subdir,
'%06d_nor.png' % (png_idx * minibatch_size + i)))
shape = images[i, 3:6]
shape_norm = (shape - shape.min()) / (shape.max() -
shape.min())
mio.export_image(
Image(shape_norm),
os.path.join(
result_subdir,
'%06d_shp.png' % (png_idx * minibatch_size + i)))
mio.export_pickle(
t_mesh,
os.path.join(result_subdir,
'%06d.pkl' % (png_idx * minibatch_size + i)))
print('%0.2f seconds' % (time.time() - start))
open(os.path.join(result_subdir, '_done.txt'), 'wt').close()
def write_img(path, img_array):
image = img_array.reshape(img_array.shape)
save_image(path, image)
def generate_fake_images(run_id,
snapshot=None,
grid_size=[1, 1],
batch_size=8,
num_pngs=1,
image_shrink=1,
png_prefix=None,
random_seed=1000,
minibatch_size=8):
network_pkl = misc.locate_network_pkl(run_id, snapshot)
if png_prefix is None:
png_prefix = misc.get_id_string_for_network_pkl(network_pkl) + '-'
random_state = np.random.RandomState(random_seed)
print('Loading network from "%s"...' % network_pkl)
G, D, Gs = misc.load_network_pkl(run_id, snapshot)
result_subdir = misc.create_result_subdir(config_test.result_dir,
config_test.desc)
for png_idx in range(int(num_pngs / batch_size)):
start = time.time()
print('Generating png %d-%d / %d... in ' %
(png_idx * batch_size, (png_idx + 1) * batch_size, num_pngs),
end='')
latents = misc.random_latents(np.prod(grid_size) * batch_size,
Gs,
random_state=random_state)
labels = np.zeros([latents.shape[0], 7], np.float32)
images = Gs.run(latents,
labels,
minibatch_size=minibatch_size,
num_gpus=config_test.num_gpus,
out_shrink=image_shrink)
for i in range(batch_size):
if images.shape[1] == 3:
mio.export_pickle(
images[i],
os.path.join(
result_subdir,
'%s%06d.pkl' % (png_prefix, png_idx * batch_size + i)))
# misc.save_image(images[i], os.path.join(result_subdir, '%s%06d.png' % (png_prefix, png_idx*batch_size+i)), [0,255], grid_size)
elif images.shape[1] == 6:
mio.export_pickle(images[i][3:6],
os.path.join(
result_subdir, '%s%06d.pkl' %
(png_prefix, png_idx * batch_size + i)),
overwrite=True)
misc.save_image(
images[i][0:3],
os.path.join(
result_subdir,
'%s%06d.png' % (png_prefix, png_idx * batch_size + i)),
[-1, 1], grid_size)
elif images.shape[1] == 9:
mio.export_pickle(images[i][3:6],
os.path.join(
result_subdir, '%s%06d_shp.pkl' %
(png_prefix, png_idx * batch_size + i)),
overwrite=True)
mio.export_pickle(images[i][6:9],
os.path.join(
result_subdir, '%s%06d_nor.pkl' %
(png_prefix, png_idx * batch_size + i)),
overwrite=True)
misc.save_image(
images[i][0:3],
os.path.join(
result_subdir,
'%s%06d.png' % (png_prefix, png_idx * batch_size + i)),
[-1, 1], grid_size)
print('%0.2f seconds' % (time.time() - start))
open(os.path.join(result_subdir, '_done.txt'), 'wt').close()
