def main():
images, labels = dataset.load_test_images()
num_scatter = len(images)
_images, _, label_id = dataset.sample_labeled_data(images, labels,
num_scatter)
with tf.device(config.device):
t = build_graph(is_test=True)
with tf.Session(config=tf.ConfigProto(allow_soft_placement=True,
log_device_placement=True)) as sess:
saver = tf.train.Saver()
saver.restore(sess, tf.train.latest_checkpoint(config.ckpt_dir))
z, _x = sess.run([t.z_r, t.x_r], feed_dict={t.x: _images})
plot.scatter_labeled_z(z, label_id, dir=config.ckpt_dir)
plot.tile_images(_x[:100], dir=config.ckpt_dir)
plot.plot_loss_tendency(config.ckpt_dir)
hist_value, hist_head = plot.load_pickle_to_data(config.ckpt_dir)
for loss_name in ['reconstruction']:
plot.plot_loss_trace(hist_value[loss_name], loss_name, config.ckpt_dir)
plot.plot_adversarial_trace(hist_value['discriminator'],
hist_value['generator'], 'z', config.ckpt_dir)
plot.plot_adversarial_trace(hist_value['discriminator_z'],
hist_value['generator_z'], 'z',
config.ckpt_dir)
plot.plot_adversarial_trace(hist_value['discriminator_img'],
hist_value['generator_img'], 'z',
config.ckpt_dir)
def main():
# load MNIST images
images, labels = dataset.load_test_images()
# Settings
num_scatter = len(images)
_images, _, label_id = dataset.sample_labeled_data(images, labels,
num_scatter)
with tf.device(config.device):
t = build_graph(is_test=True)
with tf.Session(config=tf.ConfigProto(allow_soft_placement=True,
log_device_placement=True)) as sess:
saver = tf.train.Saver()
saver.restore(sess, tf.train.latest_checkpoint(config.ckpt_dir))
representation, x_reconstruction = sess.run([t.yz, t.x_r],
feed_dict={t.x: _images})
plot.scatter_labeled_z(representation, label_id, dir=config.ckpt_dir)
plot.tile_images(x_reconstruction[:100], dir=config.ckpt_dir)
# z distributed plot
num_segments = 20
limit = (-5, 5)
x_values = np.linspace(limit[0], limit[1], num_segments)
y_values = np.linspace(limit[0], limit[1], num_segments)
vacant = np.zeros((28 * num_segments, 28 * num_segments))
for i, x_element in enumerate(x_values):
for j, y_element in enumerate(y_values):
x_reconstruction = sess.run(
t.x_r,
feed_dict={
t.yz: np.reshape([x_element, y_element], [1, 2])
})
vacant[(num_segments - 1 - i) * 28:(num_segments - i) * 28,
j * 28:(j + 1) * 28] = x_reconstruction.reshape(28, 28)
vacant = (vacant + 1) / 2
pylab.figure(figsize=(10, 10), dpi=400, facecolor='white')
pylab.imshow(vacant, cmap='gray', origin='upper')
pylab.tight_layout()
pylab.axis('off')
pylab.savefig("{}/clusters.png".format(config.ckpt_dir))
# loss part
hist_value, hist_head = plot.load_pickle_to_data(config.ckpt_dir)
for loss_name in ['reconstruction', 'supervised']:
plot.plot_loss_trace(hist_value[loss_name], loss_name,
config.ckpt_dir)
plot.plot_adversarial_trace(hist_value['discriminator_y'],
hist_value['generator_y'], 'y',
config.ckpt_dir)
plot.plot_adversarial_trace(hist_value['discriminator_z'],
hist_value['generator_z'], 'z',
config.ckpt_dir)
plot.plot_adversarial_trace(hist_value['validation_accuracy'],
hist_value['transform'],
'validation_accuracy', config.ckpt_dir)
def main():
# load MNIST images
images, labels = dataset.load_test_images()
# Settings
num_anologies = 10
pylab.gray()
# generate style vector z
x = dataset.sample_unlabeled_data(images, num_anologies)
x = (x + 1) / 2
with tf.device(config.device):
x_input, img_y, img_z, reconstruction = build_graph(is_test=True)
with tf.Session(config=tf.ConfigProto(allow_soft_placement=True,
log_device_placement=True)) as sess:
saver = tf.train.Saver()
saver.restore(sess, tf.train.latest_checkpoint(config.ckpt_dir))
z = sess.run(img_z, feed_dict={x_input: x})
for m in range(num_anologies):
pylab.subplot(num_anologies, config.ndim_y + 2, m * 12 + 1)
pylab.imshow(x[m].reshape((28, 28)), interpolation='none')
pylab.axis('off')
all_y = np.identity(config.ndim_y, dtype=np.float32)
for m in range(num_anologies):
fixed_z = np.repeat(z[m].reshape(1, -1), config.ndim_y, axis=0)
gen_x = sess.run(reconstruction,
feed_dict={
img_z: fixed_z,
img_y: all_y
})
gen_x = (gen_x + 1) / 2
for n in range(config.ndim_y):
pylab.subplot(num_anologies, config.ndim_y + 2, m * 12 + 3 + n)
pylab.imshow(gen_x[n].reshape((28, 28)), interpolation='none')
pylab.axis('off')
fig = pylab.gcf()
fig.set_size_inches(num_anologies, config.ndim_y)
pylab.savefig('{}/analogy.png'.format(config.ckpt_dir))
hist_value, hist_head = plot.load_pickle_to_data(config.ckpt_dir)
for loss_name in [
'reconstruction', 'validation_accuracy', 'supervised'
]:
plot.plot_loss_trace(hist_value[loss_name], loss_name,
config.ckpt_dir)
plot.plot_adversarial_trace(hist_value['discriminator_y'],
hist_value['generator_y'], 'y',
config.ckpt_dir)
plot.plot_adversarial_trace(hist_value['discriminator_z'],
hist_value['generator_z'], 'z',
config.ckpt_dir)
def main(): images, labels = dataset.load_test_images() num_scatter = len(images) y_distribution, z = aae.encode_x_yz(images, apply_softmax=False, test=True) y = aae.argmax_onehot_from_unnormalized_distribution(y_distribution) representation = aae.to_numpy(aae.encode_yz_representation(y, z, test=True)) plot.scatter_labeled_z(representation, labels, dir=args.plot_dir)
def main(): # load MNIST images images, labels = dataset.load_test_images() # config config = aae.config num_scatter = len(images) x, _, label_ids = dataset.sample_labeled_data(images, labels, num_scatter, config.ndim_x, config.ndim_y) z = aae.to_numpy(aae.encode_x_z(x, test=True)) visualizer.plot_labeled_z(z, label_ids, dir=args.plot_dir)
def main():
# load MNIST images
images, labels = dataset.load_test_images()
# config
config = aae.config
num_scatter = len(images)
x, _, labels = dataset.sample_labeled_data(images, labels, num_scatter,
config.ndim_x, config.ndim_y)
y_distribution, z = aae.encode_x_yz(x, apply_softmax=False, test=True)
y = aae.argmax_onehot_from_unnormalized_distribution(y_distribution)
representation = aae.to_numpy(aae.encode_yz_representation(y, z,
test=True))
visualizer.plot_labeled_z(representation, labels, dir=args.plot_dir)
def main():
images, labels = dataset.load_test_images()
num_scatter = len(images)
_images, _, label_id = dataset.sample_labeled_data(images, labels,
num_scatter)
with tf.device(config.device):
x, z_respresentation, x_construction = build_graph(is_test=True)
with tf.Session(config=tf.ConfigProto(allow_soft_placement=True,
log_device_placement=True)) as sess:
saver = tf.train.Saver()
saver.restore(sess, tf.train.latest_checkpoint(config.ckpt_dir))
z, _x = sess.run([z_respresentation, x_construction],
feed_dict={x: _images})
scatter_labeled_z(z, label_id, dir=config.ckpt_dir)
tile_images(_x[:100], dir=config.ckpt_dir)
plot_loss_tendency(config.ckpt_dir)
def main():
# load MNIST images
images, labels = dataset.load_test_images()
# config
config = aae.config
# settings
num_analogies = 10
pylab.gray()
# generate style vector z
x = dataset.sample_unlabeled_data(images,
num_analogies,
config.ndim_x,
binarize=False)
_, z = aae.encode_x_yz(x, apply_softmax=True)
z = aae.to_numpy(z)
# plot original image on the left
for m in xrange(num_analogies):
pylab.subplot(num_analogies, config.ndim_y + 2, m * 12 + 1)
pylab.imshow(x[m].reshape((28, 28)), interpolation="none")
pylab.axis("off")
all_y = np.identity(config.ndim_y, dtype=np.float32)
for m in xrange(num_analogies):
# copy z as many as the number of classes
fixed_z = np.repeat(z[m].reshape(1, -1), config.ndim_y, axis=0)
gen_x = aae.to_numpy(aae.decode_yz_x(all_y, fixed_z))
# plot images generated from each label
for n in xrange(config.ndim_y):
pylab.subplot(num_analogies, config.ndim_y + 2, m * 12 + 3 + n)
pylab.imshow(gen_x[n].reshape((28, 28)), interpolation="none")
pylab.axis("off")
fig = pylab.gcf()
fig.set_size_inches(num_analogies, config.ndim_y)
pylab.savefig("{}/analogy.png".format(args.plot_dir))
def main():
# load MNIST images
images, labels = dataset.load_test_images()
# Settings
num_scatter = len(images)
_images, _, label_id = dataset.sample_labeled_data(images, labels,
num_scatter)
with tf.device(config.device):
t = build_graph(is_test=True)
with tf.Session(config=tf.ConfigProto(allow_soft_placement=True,
log_device_placement=True)) as sess:
saver = tf.train.Saver()
saver.restore(sess, tf.train.latest_checkpoint(config.ckpt_dir))
# z distributed plot
num_segments = 20
limit = (-5, 5)
x_values = np.linspace(limit[0], limit[1], num_segments)
y_values = np.linspace(limit[0], limit[1], num_segments)
vacant = np.zeros((28 * num_segments, 28 * num_segments))
for i, x_element in enumerate(x_values):
for j, y_element in enumerate(y_values):
x_reconstruction = sess.run(
t.x_r,
feed_dict={
t.yz: np.reshape([x_element, y_element], [1, 2])
})
vacant[(num_segments - 1 - i) * 28:(num_segments - i) * 28,
j * 28:(j + 1) * 28] = x_reconstruction.reshape(28, 28)
vacant = (vacant + 1) / 2
pylab.figure(figsize=(10, 10), dpi=400, facecolor='white')
pylab.imshow(vacant, cmap='gray', origin='upper')
pylab.tight_layout()
pylab.axis('off')
pylab.savefig("{}/clusters.png".format(config.ckpt_dir))
# -*- coding: utf-8 -*- import os, sys, time, pylab import numpy as np import matplotlib.patches as mpatches sys.path.append(os.path.abspath(os.path.join(os.getcwd(), "../../"))) import dataset from args import args from model import aae try: os.mkdir(args.plot_dir) except: pass images, labels = dataset.load_test_images() config = aae.config num_clusters = config.ndim_y num_plots_per_cluster = 11 image_width = 28 image_height = 28 ndim_x = image_width * image_height pylab.gray() # plot cluster head head_y = np.identity(config.ndim_y, dtype=np.float32) zero_z = np.zeros((config.ndim_y, config.ndim_z), dtype=np.float32) head_x = aae.to_numpy(aae.decode_yz_x(head_y, zero_z, test=True)) head_x = (head_x + 1.0) / 2.0 for n in xrange(num_clusters): pylab.subplot(num_clusters, num_plots_per_cluster + 2, n * (num_plots_per_cluster + 2) + 1) pylab.imshow(head_x[n].reshape((image_width, image_height)), interpolation="none")
import os, sys, time, pylab import numpy as np from chainer import cuda, Variable import matplotlib.patches as mpatches sys.path.append(os.path.split(os.getcwd())[0]) import dataset from model import adgm from args import args try: os.mkdir(args.plot_dir) except: pass # load test images images, labels = dataset.load_test_images() # config config = adgm.config num_analogies = 10 xp = np if args.gpu_device != -1: xp = cuda.cupy # sample data x = dataset.sample_unlabeled_data(images, num_analogies, config.ndim_x, binarize=False) z = adgm.encode_x_z(x, argmax_y=True, test=True) # plot fig = pylab.gcf() fig.set_size_inches(16.0, 16.0)
def main(run_load_from_file=False):
# load MNIST images
images, labels = dataset.load_test_images()
# config
opt = Operation()
opt.check_dir(config.ckpt_dir, is_restart=False)
opt.check_dir(config.log_dir, is_restart=True)
max_epoch = 510
num_trains_per_epoch = 500
batch_size_u = 100
# training
with tf.device(config.device):
h = build_graph()
sess_config = tf.ConfigProto(allow_soft_placement=True,
log_device_placement=True)
sess_config.gpu_options.allow_growth = True
sess_config.gpu_options.per_process_gpu_memory_fraction = 0.9
saver = tf.train.Saver(max_to_keep=2)
with tf.Session(config=sess_config) as sess:
'''
Load from checkpoint or start a new session
'''
if run_load_from_file:
saver.restore(sess, tf.train.latest_checkpoint(config.ckpt_dir))
training_epoch_loss, _ = pickle.load(
open(config.ckpt_dir + '/pickle.pkl', 'rb'))
else:
sess.run(tf.global_variables_initializer())
training_epoch_loss = []
# Recording loss per epoch
process = Process()
for epoch in range(max_epoch):
process.start_epoch(epoch, max_epoch)
'''
Learning rate generator
'''
learning_rate = 0.0001
# Recording loss per iteration
sum_loss_reconstruction = 0
sum_loss_discrminator_z = 0
sum_loss_discrminator_img = 0
sum_loss_generator_z = 0
sum_loss_generator_img = 0
process_iteration = Process()
for i in range(num_trains_per_epoch):
process_iteration.start_epoch(i, num_trains_per_epoch)
# Inputs
'''
_l -> labeled
_u -> unlabeled
'''
images_u = dataset.sample_unlabeled_data(images, batch_size_u)
if config.distribution_sampler == 'swiss_roll':
z_true_u = sampler.swiss_roll(batch_size_u, config.ndim_z,
config.num_types_of_label)
elif config.distribution_sampler == 'gaussian_mixture':
z_true_u = sampler.gaussian_mixture(
batch_size_u, config.ndim_z, config.num_types_of_label)
elif config.distribution_sampler == 'uniform_desk':
z_true_u = sampler.uniform_desk(batch_size_u,
config.ndim_z,
radius=2)
elif config.distribution_sampler == 'gaussian':
z_true_u = sampler.gaussian(batch_size_u,
config.ndim_z,
var=1)
elif config.distribution_sampler == 'uniform':
z_true_u = sampler.uniform(batch_size_u,
config.ndim_z,
minv=-1,
maxv=1)
# reconstruction_phase
_, loss_reconstruction = sess.run([h.opt_r, h.loss_r],
feed_dict={
h.x: images_u,
h.lr: learning_rate
})
# adversarial phase for discriminator_z
images_u_s = dataset.sample_unlabeled_data(
images, batch_size_u)
_, loss_discriminator_z = sess.run([h.opt_dz, h.loss_dz],
feed_dict={
h.x: images_u,
h.z: z_true_u,
h.lr: learning_rate
})
_, loss_discriminator_img = sess.run([h.opt_dimg, h.loss_dimg],
feed_dict={
h.x: images_u,
h.x_s: images_u_s,
h.lr: learning_rate
})
# adversarial phase for generator
_, loss_generator_z = sess.run([h.opt_e, h.loss_e],
feed_dict={
h.x: images_u,
h.lr: learning_rate
})
_, loss_generator_img = sess.run([h.opt_d, h.loss_d],
feed_dict={
h.x: images_u,
h.lr: learning_rate
})
sum_loss_reconstruction += loss_reconstruction
sum_loss_discrminator_z += loss_discriminator_z
sum_loss_discrminator_img += loss_discriminator_img
sum_loss_generator_z += loss_generator_z
sum_loss_generator_img += loss_generator_img
if i % 1000 == 0:
process_iteration.show_table_2d(
i, num_trains_per_epoch, {
'reconstruction':
sum_loss_reconstruction / (i + 1),
'discriminator_z':
sum_loss_discrminator_z / (i + 1),
'discriminator_img':
sum_loss_discrminator_img / (i + 1),
'generator_z':
sum_loss_generator_z / (i + 1),
'generator_img':
sum_loss_generator_img / (i + 1),
})
average_loss_per_epoch = [
sum_loss_reconstruction / num_trains_per_epoch,
sum_loss_discrminator_z / num_trains_per_epoch,
sum_loss_discrminator_img / num_trains_per_epoch,
sum_loss_generator_z / num_trains_per_epoch,
sum_loss_generator_img / num_trains_per_epoch,
(sum_loss_discrminator_z + sum_loss_discrminator_img) /
num_trains_per_epoch,
(sum_loss_generator_z + sum_loss_generator_img) /
num_trains_per_epoch
]
training_epoch_loss.append(average_loss_per_epoch)
training_loss_name = [
'reconstruction', 'discriminator_z', 'discriminator_img',
'generator_z', 'generator_img', 'discriminator', 'generator'
]
if epoch % 1 == 0:
process.show_bar(
epoch, max_epoch, {
'loss_r': average_loss_per_epoch[0],
'loss_d': average_loss_per_epoch[5],
'loss_g': average_loss_per_epoch[6]
})
plt.scatter_labeled_z(
sess.run(h.z_r, feed_dict={h.x: images[:1000]}),
[int(var) for var in labels[:1000]],
dir=config.log_dir,
filename='z_representation-{}'.format(epoch))
if epoch % 10 == 0:
saver.save(sess,
os.path.join(config.ckpt_dir, 'model_ckptpoint'),
global_step=epoch)
pickle.dump((training_epoch_loss, training_loss_name),
open(config.ckpt_dir + '/pickle.pkl', 'wb'))
def main(): images, labels = dataset.load_test_images() num_scatter = len(images) x, _, label_ids = dataset.sample_labeled_data(images, labels, num_scatter) z = aae.to_numpy(aae.encode_x_z(x, test=True)) plot.scatter_labeled_z(z, label_ids, dir=args.plot_dir)
