def main():
os.environ['CHAINER_SEED'] = str(args.seed)
logging.info('chainer seed = ' + os.environ['CHAINER_SEED'])
_mkdir(args.snapshot_directory)
_mkdir(args.log_directory)
meter_train = Meter()
meter_train.load(args.snapshot_directory)
#==============================================================================
# Dataset
#==============================================================================
def read_npy_files(directory):
filenames = []
files = os.listdir(os.path.join(directory, "images"))
for filename in files:
if filename.endswith(".npy"):
filenames.append(filename)
filenames.sort()
dataset_images = []
dataset_viewpoints = []
for i in range(len(filenames)):
images_npy_path = os.path.join(directory, "images", filenames[i])
viewpoints_npy_path = os.path.join(directory, "viewpoints", filenames[i])
tmp_images = np.load(images_npy_path)
tmp_viewpoints = np.load(viewpoints_npy_path)
assert tmp_images.shape[0] == tmp_viewpoints.shape[0]
dataset_images.extend(tmp_images)
dataset_viewpoints.extend(tmp_viewpoints)
dataset_images = np.array(dataset_images)
dataset_viewpoints = np.array(dataset_viewpoints)
dataset = list()
for i in range(len(dataset_images)):
item = {'image':dataset_images[i],'viewpoint':dataset_viewpoints[i]}
dataset.append(item)
return dataset
def read_files(directory):
filenames = []
files = os.listdir(directory)
for filename in files:
if filename.endswith(".h5"):
filenames.append(filename)
filenames.sort()
dataset_images = []
dataset_viewpoints = []
for i in range(len(filenames)):
F = h5py.File(os.path.join(directory,filenames[i]))
tmp_images = list(F["images"])
tmp_viewpoints = list(F["viewpoints"])
dataset_images.extend(tmp_images)
dataset_viewpoints.extend(tmp_viewpoints)
dataset_images = np.array(dataset_images)
dataset_viewpoints = np.array(dataset_viewpoints)
dataset = list()
for i in range(len(dataset_images)):
item = {'image':dataset_images[i],'viewpoint':dataset_viewpoints[i]}
dataset.append(item)
return dataset
dataset_train = read_files(args.train_dataset_directory)
# ipdb.set_trace()
if args.test_dataset_directory is not None:
dataset_test = read_files(args.test_dataset_directory)
# ipdb.set_trace()
#==============================================================================
# Hyperparameters
#==============================================================================
hyperparams = HyperParameters()
hyperparams.num_layers = args.generation_steps
hyperparams.generator_share_core = args.generator_share_core
hyperparams.inference_share_core = args.inference_share_core
hyperparams.h_channels = args.h_channels
hyperparams.z_channels = args.z_channels
hyperparams.u_channels = args.u_channels
hyperparams.r_channels = args.r_channels
hyperparams.image_size = (args.image_size, args.image_size)
hyperparams.representation_architecture = args.representation_architecture
hyperparams.pixel_sigma_annealing_steps = args.pixel_sigma_annealing_steps
hyperparams.initial_pixel_sigma = args.initial_pixel_sigma
hyperparams.final_pixel_sigma = args.final_pixel_sigma
hyperparams.save(args.snapshot_directory)
print(hyperparams, "\n")
#==============================================================================
# Model
#==============================================================================
model = Model(hyperparams)
model.load(args.snapshot_directory, meter_train.epoch)
#==============================================================================
# Pixel-variance annealing
#==============================================================================
variance_scheduler = PixelVarianceScheduler(
sigma_start=args.initial_pixel_sigma,
sigma_end=args.final_pixel_sigma,
final_num_updates=args.pixel_sigma_annealing_steps)
variance_scheduler.load(args.snapshot_directory)
print(variance_scheduler, "\n")
pixel_log_sigma = np.full(
(args.batch_size, 3) + hyperparams.image_size,
math.log(variance_scheduler.standard_deviation),
dtype="float32")
#==============================================================================
# Selecting the GPU
#==============================================================================
# xp = np
# gpu_device = args.gpu_device
# using_gpu = gpu_device >= 0
# if using_gpu:
# cuda.get_device(gpu_device).use()
# xp = cp
# devices = tuple([chainer.get_device(f"@cupy:{gpu}") for gpu in args.gpu_devices])
# if any(device.xp is chainerx for device in devices):
# sys.stderr.write("Cannot support ChainerX devices.")
# sys.exit(1)
ngpu = args.ngpu
using_gpu = ngpu > 0
xp=cp
if ngpu == 1:
gpu_id = 0
# Make a specified GPU current
chainer.cuda.get_device_from_id(gpu_id).use()
model.to_gpu() # Copy the model to the GPU
logging.info('single gpu calculation.')
elif ngpu > 1:
gpu_id = 0
devices = {'main': gpu_id}
for gid in six.moves.xrange(1, ngpu):
devices['sub_%d' % gid] = gid
logging.info('multi gpu calculation (#gpus = %d).' % ngpu)
logging.info('batch size is automatically increased (%d -> %d)' % (
args.batch_size, args.batch_size * args.ngpu))
else:
gpu_id = -1
logging.info('cpu calculation')
#==============================================================================
# Logging
#==============================================================================
csv = DataFrame()
csv.load(args.log_directory)
#==============================================================================
# Optimizer
#==============================================================================
initial_training_step=0
# lr = compute_lr_at_step(initial_training_step) # function in GQN AdamOptimizer
optimizer = chainer.optimizers.Adam(beta1=0.9, beta2=0.99, eps=1e-8) #lr is needed originally
optimizer.setup(model)
# optimizer = AdamOptimizer(
# model.parameters,
# initial_lr=args.initial_lr,
# final_lr=args.final_lr,
# initial_training_step=variance_scheduler.training_step)
# )
print(optimizer, "\n")
#==============================================================================
# Training iterations
#==============================================================================
if ngpu>1:
train_iters = [
chainer.iterators.MultiprocessIterator(dataset_train, args.batch_size, n_processes=args.number_processes, order_sampler=chainer.iterators.ShuffleOrderSampler()) for i in chainer.datasets.split_dataset_n_random(dataset_train, len(devices))
]
updater = CustomParallelUpdater(train_iters, optimizer, devices, converter=chainer.dataset.concat_examples, pixel_log_sigma=pixel_log_sigma)
elif ngpu==1:
train_iters = chainer.iterators.SerialIterator(dataset_train,args.batch_size,shuffle=True)
updater = CustomUpdater(train_iters, optimizer, device=0, converter=chainer.dataset.concat_examples, pixel_log_sigma=pixel_log_sigma)
else:
raise NotImplementedError('Implement for single gpu or cpu')
trainer = chainer.training.Trainer(updater,(args.epochs,'epoch'),args.snapshot_directory)
trainer.extend(AnnealLearningRate(
initial_lr=args.initial_lr,
final_lr=args.final_lr,
annealing_steps=args.pixel_sigma_annealing_steps,
optimizer=optimizer),
trigger=(1,'iteration'))
# add information per epoch with report?
# add learning rate annealing, snapshot saver, evaluator
trainer.extend(extensions.LogReport())
trainer.extend(extensions.snapshot(filename='snapshot_epoch_{.updater.epoch}',
savefun=chainer.serializers.save_hdf5,
target=optimizer.target),
trigger=(args.report_interval_iters,'epoch'))
trainer.extend(extensions.ProgressBar())
reports = ['epoch', 'main/loss', 'main/bits_per_pixel', 'main/NLL', 'main/MSE']
#Validation
if args.test_dataset_directory is not None:
test_iters = chainer.iterators.SerialIterator(
dataset_test,args.batch_size*6, repeat=False, shuffle=False)
trainer.extend(Validation(
test_iters, chainer.dataset.concat_examples, optimizer.target, variance_scheduler,device=0))
reports.append('validation/main/bits_per_pixel')
reports.append('validation/main/NLL')
reports.append('validation/main/MSE')
reports.append('elapsed_time')
trainer.extend(
extensions.PrintReport(reports), trigger=(args.report_interval_iters, 'iteration'))
# np.random.seed(args.seed)
# cp.random.seed(args.seed)
trainer.run()
def main():
try:
os.mkdir(args.snapshot_directory)
except:
pass
np.random.seed(0)
xp = np
device_gpu = args.gpu_device
device_cpu = -1
using_gpu = device_gpu >= 0
if using_gpu:
cuda.get_device(args.gpu_device).use()
xp = cupy
dataset = gqn.data.Dataset(args.dataset_directory)
hyperparams = HyperParameters()
hyperparams.generator_share_core = args.generator_share_core
hyperparams.generator_share_prior = args.generator_share_prior
hyperparams.generator_generation_steps = args.generation_steps
hyperparams.generator_share_upsampler = args.generator_share_upsampler
hyperparams.inference_share_core = args.inference_share_core
hyperparams.inference_share_posterior = args.inference_share_posterior
hyperparams.h_channels = args.h_channels
hyperparams.z_channels = args.z_channels
hyperparams.u_channels = args.u_channels
hyperparams.image_size = (args.image_size, args.image_size)
hyperparams.representation_channels = args.representation_channels
hyperparams.representation_architecture = args.representation_architecture
hyperparams.pixel_n = args.pixel_n
hyperparams.pixel_sigma_i = args.initial_pixel_variance
hyperparams.pixel_sigma_f = args.final_pixel_variance
hyperparams.save(args.snapshot_directory)
print(hyperparams)
model = Model(hyperparams,
snapshot_directory=args.snapshot_directory,
optimized=args.optimized)
if using_gpu:
model.to_gpu()
scheduler = Scheduler(sigma_start=args.initial_pixel_variance,
sigma_end=args.final_pixel_variance,
final_num_updates=args.pixel_n,
snapshot_directory=args.snapshot_directory)
print(scheduler)
optimizer = AdamOptimizer(model.parameters,
mu_i=args.initial_lr,
mu_f=args.final_lr,
initial_training_step=scheduler.num_updates)
print(optimizer)
pixel_var = xp.full((args.batch_size, 3) + hyperparams.image_size,
scheduler.pixel_variance**2,
dtype="float32")
pixel_ln_var = xp.full((args.batch_size, 3) + hyperparams.image_size,
math.log(scheduler.pixel_variance**2),
dtype="float32")
representation_shape = (args.batch_size,
hyperparams.representation_channels,
args.image_size // 4, args.image_size // 4)
fig = plt.figure(figsize=(9, 3))
axis_data = fig.add_subplot(1, 3, 1)
axis_data.set_title("Data")
axis_data.axis("off")
axis_reconstruction = fig.add_subplot(1, 3, 2)
axis_reconstruction.set_title("Reconstruction")
axis_reconstruction.axis("off")
axis_generation = fig.add_subplot(1, 3, 3)
axis_generation.set_title("Generation")
axis_generation.axis("off")
current_training_step = 0
for iteration in range(args.training_iterations):
mean_kld = 0
mean_nll = 0
mean_mse = 0
mean_elbo = 0
total_num_batch = 0
start_time = time.time()
for subset_index, subset in enumerate(dataset):
iterator = gqn.data.Iterator(subset, batch_size=args.batch_size)
for batch_index, data_indices in enumerate(iterator):
# shape: (batch, views, height, width, channels)
# range: [-1, 1]
images, viewpoints = subset[data_indices]
# (batch, views, height, width, channels) -> (batch, views, channels, height, width)
images = images.transpose((0, 1, 4, 2, 3)).astype(np.float32)
total_views = images.shape[1]
# Sample number of views
num_views = random.choice(range(1, total_views + 1))
observation_view_indices = list(range(total_views))
random.shuffle(observation_view_indices)
observation_view_indices = observation_view_indices[:num_views]
query_index = random.choice(range(total_views))
if num_views > 0:
observation_images = preprocess_images(
images[:, observation_view_indices])
observation_query = viewpoints[:, observation_view_indices]
representation = model.compute_observation_representation(
observation_images, observation_query)
else:
representation = xp.zeros(representation_shape,
dtype="float32")
representation = chainer.Variable(representation)
# Sample query
query_index = random.choice(range(total_views))
query_images = preprocess_images(images[:, query_index])
query_viewpoints = viewpoints[:, query_index]
# Transfer to gpu if necessary
query_images = to_device(query_images, device_gpu)
query_viewpoints = to_device(query_viewpoints, device_gpu)
z_t_param_array, mean_x = model.sample_z_and_x_params_from_posterior(
query_images, query_viewpoints, representation)
# Compute loss
## KL Divergence
loss_kld = 0
for params in z_t_param_array:
mean_z_q, ln_var_z_q, mean_z_p, ln_var_z_p = params
kld = gqn.functions.gaussian_kl_divergence(
mean_z_q, ln_var_z_q, mean_z_p, ln_var_z_p)
loss_kld += cf.sum(kld)
## Negative log-likelihood of generated image
loss_nll = cf.sum(
gqn.functions.gaussian_negative_log_likelihood(
query_images, mean_x, pixel_var, pixel_ln_var))
# Calculate the average loss value
loss_nll = loss_nll / args.batch_size
loss_kld = loss_kld / args.batch_size
loss = loss_nll / scheduler.pixel_variance + loss_kld
model.cleargrads()
loss.backward()
optimizer.update(current_training_step)
loss_nll = float(loss_nll.data) + math.log(256.0)
loss_kld = float(loss_kld.data)
elbo = -(loss_nll + loss_kld)
loss_mse = float(
cf.mean_squared_error(query_images, mean_x).data)
printr(
"Iteration {}: Subset {} / {}: Batch {} / {} - elbo: {:.2f} - loss: nll: {:.2f} mse: {:.6e} kld: {:.5f} - lr: {:.4e} - pixel_variance: {:.5f} - step: {} "
.format(iteration + 1,
subset_index + 1, len(dataset), batch_index + 1,
len(iterator), elbo, loss_nll, loss_mse, loss_kld,
optimizer.learning_rate, scheduler.pixel_variance,
current_training_step))
scheduler.step(iteration, current_training_step)
pixel_var[...] = scheduler.pixel_variance**2
pixel_ln_var[...] = math.log(scheduler.pixel_variance**2)
total_num_batch += 1
current_training_step += 1
mean_kld += loss_kld
mean_nll += loss_nll
mean_mse += loss_mse
mean_elbo += elbo
model.serialize(args.snapshot_directory)
# Visualize
if args.with_visualization:
axis_data.imshow(make_uint8(query_images[0]),
interpolation="none")
axis_reconstruction.imshow(make_uint8(mean_x.data[0]),
interpolation="none")
with chainer.no_backprop_mode():
generated_x = model.generate_image(
query_viewpoints[None, 0], representation[None, 0])
axis_generation.imshow(make_uint8(generated_x[0]),
interpolation="none")
plt.pause(1e-8)
elapsed_time = time.time() - start_time
print(
"\033[2KIteration {} - elbo: {:.2f} - loss: nll: {:.2f} mse: {:.6e} kld: {:.5f} - lr: {:.4e} - pixel_variance: {:.5f} - step: {} - time: {:.3f} min"
.format(iteration + 1, mean_elbo / total_num_batch,
mean_nll / total_num_batch, mean_mse / total_num_batch,
mean_kld / total_num_batch, optimizer.learning_rate,
scheduler.pixel_variance, current_training_step,
elapsed_time / 60))
model.serialize(args.snapshot_directory)
def main():
_mkdir(args.snapshot_directory)
_mkdir(args.log_directory)
meter_train = Meter()
meter_train.load(args.snapshot_directory)
#==============================================================================
# Workaround to fix OpenMPI bug
#==============================================================================
multiprocessing.set_start_method("forkserver")
p = multiprocessing.Process(target=print, args=("", ))
p.start()
p.join()
#==============================================================================
# Selecting the GPU
#==============================================================================
comm = chainermn.create_communicator()
device = comm.intra_rank
cuda.get_device(device).use()
def _print(*args):
if comm.rank == 0:
print(*args)
_print("Using {} GPUs".format(comm.size))
#==============================================================================
# Dataset
#==============================================================================
dataset_train = Dataset(args.train_dataset_directory)
dataset_test = None
if args.test_dataset_directory is not None:
dataset_test = Dataset(args.test_dataset_directory)
#==============================================================================
# Hyperparameters
#==============================================================================
hyperparams = HyperParameters()
hyperparams.num_layers = args.generation_steps
hyperparams.generator_share_core = args.generator_share_core
hyperparams.inference_share_core = args.inference_share_core
hyperparams.h_channels = args.h_channels
hyperparams.z_channels = args.z_channels
hyperparams.u_channels = args.u_channels
hyperparams.r_channels = args.r_channels
hyperparams.image_size = (args.image_size, args.image_size)
hyperparams.representation_architecture = args.representation_architecture
hyperparams.pixel_sigma_annealing_steps = args.pixel_sigma_annealing_steps
hyperparams.initial_pixel_sigma = args.initial_pixel_sigma
hyperparams.final_pixel_sigma = args.final_pixel_sigma
_print(hyperparams, "\n")
if comm.rank == 0:
hyperparams.save(args.snapshot_directory)
#==============================================================================
# Model
#==============================================================================
model = Model(hyperparams)
model.load(args.snapshot_directory, meter_train.epoch)
model.to_gpu()
#==============================================================================
# Pixel-variance annealing
#==============================================================================
variance_scheduler = PixelVarianceScheduler(
sigma_start=args.initial_pixel_sigma,
sigma_end=args.final_pixel_sigma,
final_num_updates=args.pixel_sigma_annealing_steps)
variance_scheduler.load(args.snapshot_directory)
_print(variance_scheduler, "\n")
pixel_log_sigma = cp.full(
(args.batch_size, 3) + hyperparams.image_size,
math.log(variance_scheduler.standard_deviation),
dtype="float32")
#==============================================================================
# Logging
#==============================================================================
csv = DataFrame()
csv.load(args.log_directory)
#==============================================================================
# Optimizer
#==============================================================================
optimizer = AdamOptimizer(
model.parameters,
initial_lr=args.initial_lr,
final_lr=args.final_lr,
initial_training_step=variance_scheduler.training_step)
_print(optimizer, "\n")
#==============================================================================
# Algorithms
#==============================================================================
def encode_scene(images, viewpoints):
# (batch, views, height, width, channels) -> (batch, views, channels, height, width)
images = images.transpose((0, 1, 4, 2, 3)).astype(np.float32)
# Sample number of views
total_views = images.shape[1]
num_views = random.choice(range(1, total_views + 1))
# Sample views
observation_view_indices = list(range(total_views))
random.shuffle(observation_view_indices)
observation_view_indices = observation_view_indices[:num_views]
observation_images = preprocess_images(
images[:, observation_view_indices])
observation_query = viewpoints[:, observation_view_indices]
representation = model.compute_observation_representation(
observation_images, observation_query)
# Sample query view
query_index = random.choice(range(total_views))
query_images = preprocess_images(images[:, query_index])
query_viewpoints = viewpoints[:, query_index]
# Transfer to gpu if necessary
query_images = cuda.to_gpu(query_images)
query_viewpoints = cuda.to_gpu(query_viewpoints)
return representation, query_images, query_viewpoints
def estimate_ELBO(query_images, z_t_param_array, pixel_mean,
pixel_log_sigma):
# KL Diverge, pixel_ln_varnce
kl_divergence = 0
for params_t in z_t_param_array:
mean_z_q, ln_var_z_q, mean_z_p, ln_var_z_p = params_t
normal_q = chainer.distributions.Normal(
mean_z_q, log_scale=ln_var_z_q)
normal_p = chainer.distributions.Normal(
mean_z_p, log_scale=ln_var_z_p)
kld_t = chainer.kl_divergence(normal_q, normal_p)
kl_divergence += cf.sum(kld_t)
kl_divergence = kl_divergence / args.batch_size
# Negative log-likelihood of generated image
batch_size = query_images.shape[0]
num_pixels_per_batch = np.prod(query_images.shape[1:])
normal = chainer.distributions.Normal(
query_images, log_scale=pixel_log_sigma)
log_px = cf.sum(normal.log_prob(pixel_mean)) / batch_size
negative_log_likelihood = -log_px
# Empirical ELBO
ELBO = log_px - kl_divergence
# https://arxiv.org/abs/1604.08772 Section.2
# https://www.reddit.com/r/MachineLearning/comments/56m5o2/discussion_calculation_of_bitsdims/
bits_per_pixel = -(ELBO / num_pixels_per_batch - np.log(256)) / np.log(
2)
return ELBO, bits_per_pixel, negative_log_likelihood, kl_divergence
#==============================================================================
# Training iterations
#==============================================================================
dataset_size = len(dataset_train)
random.seed(0)
np.random.seed(0)
cp.random.seed(0)
for epoch in range(args.epochs):
_print("Epoch {}/{}:".format(
epoch + 1,
args.epochs,
))
meter_train.next_epoch()
subset_indices = list(range(len(dataset_train.subset_filenames)))
subset_size_per_gpu = len(subset_indices) // comm.size
if len(subset_indices) % comm.size != 0:
subset_size_per_gpu += 1
for subset_loop in range(subset_size_per_gpu):
random.shuffle(subset_indices)
subset_index = subset_indices[comm.rank]
subset = dataset_train.read(subset_index)
iterator = gqn.data.Iterator(subset, batch_size=args.batch_size)
for batch_index, data_indices in enumerate(iterator):
#------------------------------------------------------------------------------
# Scene encoder
#------------------------------------------------------------------------------
# images.shape: (batch, views, height, width, channels)
images, viewpoints = subset[data_indices]
representation, query_images, query_viewpoints = encode_scene(
images, viewpoints)
#------------------------------------------------------------------------------
# Compute empirical ELBO
#------------------------------------------------------------------------------
# Compute distribution parameterws
(z_t_param_array,
pixel_mean) = model.sample_z_and_x_params_from_posterior(
query_images, query_viewpoints, representation)
# Compute ELBO
(ELBO, bits_per_pixel, negative_log_likelihood,
kl_divergence) = estimate_ELBO(query_images, z_t_param_array,
pixel_mean, pixel_log_sigma)
#------------------------------------------------------------------------------
# Update parameters
#------------------------------------------------------------------------------
loss = -ELBO
model.cleargrads()
loss.backward()
optimizer.update(meter_train.num_updates)
#------------------------------------------------------------------------------
# Logging
#------------------------------------------------------------------------------
with chainer.no_backprop_mode():
mean_squared_error = cf.mean_squared_error(
query_images, pixel_mean)
meter_train.update(
ELBO=float(ELBO.data),
bits_per_pixel=float(bits_per_pixel.data),
negative_log_likelihood=float(
negative_log_likelihood.data),
kl_divergence=float(kl_divergence.data),
mean_squared_error=float(mean_squared_error.data))
#------------------------------------------------------------------------------
# Annealing
#------------------------------------------------------------------------------
variance_scheduler.update(meter_train.num_updates)
pixel_log_sigma[...] = math.log(
variance_scheduler.standard_deviation)
if subset_loop % 100 == 0:
_print(" Subset {}/{}:".format(
subset_loop + 1,
subset_size_per_gpu,
dataset_size,
))
_print(" {}".format(meter_train))
_print(" lr: {} - sigma: {}".format(
optimizer.learning_rate,
variance_scheduler.standard_deviation))
#------------------------------------------------------------------------------
# Validation
#------------------------------------------------------------------------------
meter_test = None
if dataset_test is not None:
meter_test = Meter()
batch_size_test = args.batch_size * 6
subset_indices_test = list(
range(len(dataset_test.subset_filenames)))
pixel_log_sigma_test = cp.full(
(batch_size_test, 3) + hyperparams.image_size,
math.log(variance_scheduler.standard_deviation),
dtype="float32")
subset_size_per_gpu = len(subset_indices_test) // comm.size
with chainer.no_backprop_mode():
for subset_loop in range(subset_size_per_gpu):
subset_index = subset_indices_test[subset_loop * comm.size
+ comm.rank]
subset = dataset_train.read(subset_index)
iterator = gqn.data.Iterator(
subset, batch_size=batch_size_test)
for data_indices in iterator:
images, viewpoints = subset[data_indices]
# Scene encoder
representation, query_images, query_viewpoints = encode_scene(
images, viewpoints)
# Compute empirical ELBO
(z_t_param_array, pixel_mean
) = model.sample_z_and_x_params_from_posterior(
query_images, query_viewpoints, representation)
(ELBO, bits_per_pixel, negative_log_likelihood,
kl_divergence) = estimate_ELBO(
query_images, z_t_param_array, pixel_mean,
pixel_log_sigma_test)
mean_squared_error = cf.mean_squared_error(
query_images, pixel_mean)
# Logging
meter_test.update(
ELBO=float(ELBO.data),
bits_per_pixel=float(bits_per_pixel.data),
negative_log_likelihood=float(
negative_log_likelihood.data),
kl_divergence=float(kl_divergence.data),
mean_squared_error=float(mean_squared_error.data))
meter = meter_test.allreduce(comm)
_print(" Test:")
_print(" {} - done in {:.3f} min".format(
meter,
meter.elapsed_time,
))
model.save(args.snapshot_directory, meter_train.epoch)
variance_scheduler.save(args.snapshot_directory)
meter_train.save(args.snapshot_directory)
csv.save(args.log_directory)
_print("Epoch {} done in {:.3f} min".format(
epoch + 1,
meter_train.epoch_elapsed_time,
))
_print(" {}".format(meter_train))
_print(" lr: {} - sigma: {} - training_steps: {}".format(
optimizer.learning_rate,
variance_scheduler.standard_deviation,
meter_train.num_updates,
))
_print(" Time elapsed: {:.3f} min".format(
meter_train.elapsed_time))
def main():
try:
os.mkdir(args.snapshot_path)
except:
pass
comm = chainermn.create_communicator()
device = comm.intra_rank
print("device", device, "/", comm.size)
cuda.get_device(device).use()
xp = cupy
dataset = gqn.data.Dataset(args.dataset_path)
hyperparams = HyperParameters()
hyperparams.generator_share_core = args.generator_share_core
hyperparams.generator_share_prior = args.generator_share_prior
hyperparams.generator_generation_steps = args.generation_steps
hyperparams.inference_share_core = args.inference_share_core
hyperparams.inference_share_posterior = args.inference_share_posterior
hyperparams.channels_chz = args.channels_chz
hyperparams.generator_channels_u = args.channels_u
hyperparams.inference_channels_map_x = args.channels_map_x
hyperparams.pixel_n = args.pixel_n
hyperparams.pixel_sigma_i = args.initial_pixel_sigma
hyperparams.pixel_sigma_f = args.final_pixel_sigma
if comm.rank == 0:
hyperparams.save(args.snapshot_path)
hyperparams.print()
model = Model(hyperparams, snapshot_directory=args.snapshot_path)
model.to_gpu()
optimizer = Optimizer(
model.parameters,
communicator=comm,
mu_i=args.initial_lr,
mu_f=args.final_lr)
if comm.rank == 0:
optimizer.print()
dataset_mean, dataset_std = dataset.load_mean_and_std()
if comm.rank == 0:
np.save(os.path.join(args.snapshot_path, "mean.npy"), dataset_mean)
np.save(os.path.join(args.snapshot_path, "std.npy"), dataset_std)
# avoid division by zero
dataset_std += 1e-12
sigma_t = hyperparams.pixel_sigma_i
pixel_var = xp.full(
(args.batch_size, 3) + hyperparams.image_size,
sigma_t**2,
dtype="float32")
pixel_ln_var = xp.full(
(args.batch_size, 3) + hyperparams.image_size,
math.log(sigma_t**2),
dtype="float32")
random.seed(0)
subset_indices = list(range(len(dataset.subset_filenames)))
current_training_step = 0
for iteration in range(args.training_iterations):
mean_kld = 0
mean_nll = 0
total_batch = 0
subset_size_per_gpu = len(subset_indices) // comm.size
start_time = time.time()
for subset_loop in range(subset_size_per_gpu):
random.shuffle(subset_indices)
subset_index = subset_indices[comm.rank]
subset = dataset.read(subset_index)
iterator = gqn.data.Iterator(subset, batch_size=args.batch_size)
for batch_index, data_indices in enumerate(iterator):
# shape: (batch, views, height, width, channels)
# range: [-1, 1]
images, viewpoints = subset[data_indices]
# preprocessing
images = (images - dataset_mean) / dataset_std
# (batch, views, height, width, channels) -> (batch, views, channels, height, width)
images = images.transpose((0, 1, 4, 2, 3))
total_views = images.shape[1]
# sample number of views
num_views = random.choice(range(total_views))
query_index = random.choice(range(total_views))
if current_training_step == 0 and num_views == 0:
num_views = 1 # avoid OpenMPI error
if num_views > 0:
r = model.compute_observation_representation(
images[:, :num_views], viewpoints[:, :num_views])
else:
r = xp.zeros(
(args.batch_size, hyperparams.channels_r) +
hyperparams.chrz_size,
dtype="float32")
r = chainer.Variable(r)
query_images = images[:, query_index]
query_viewpoints = viewpoints[:, query_index]
# transfer to gpu
query_images = to_gpu(query_images)
query_viewpoints = to_gpu(query_viewpoints)
h0_gen, c0_gen, u_0, h0_enc, c0_enc = model.generate_initial_state(
args.batch_size, xp)
loss_kld = 0
hl_enc = h0_enc
cl_enc = c0_enc
hl_gen = h0_gen
cl_gen = c0_gen
ul_enc = u_0
xq = model.inference_downsampler.downsample(query_images)
for l in range(model.generation_steps):
inference_core = model.get_inference_core(l)
inference_posterior = model.get_inference_posterior(l)
generation_core = model.get_generation_core(l)
generation_piror = model.get_generation_prior(l)
h_next_enc, c_next_enc = inference_core.forward_onestep(
hl_gen, hl_enc, cl_enc, xq, query_viewpoints, r)
mean_z_q = inference_posterior.compute_mean_z(hl_enc)
ln_var_z_q = inference_posterior.compute_ln_var_z(hl_enc)
ze_l = cf.gaussian(mean_z_q, ln_var_z_q)
mean_z_p = generation_piror.compute_mean_z(hl_gen)
ln_var_z_p = generation_piror.compute_ln_var_z(hl_gen)
h_next_gen, c_next_gen, u_next_enc = generation_core.forward_onestep(
hl_gen, cl_gen, ul_enc, ze_l, query_viewpoints, r)
kld = gqn.nn.chainer.functions.gaussian_kl_divergence(
mean_z_q, ln_var_z_q, mean_z_p, ln_var_z_p)
loss_kld += cf.sum(kld)
hl_gen = h_next_gen
cl_gen = c_next_gen
ul_enc = u_next_enc
hl_enc = h_next_enc
cl_enc = c_next_enc
mean_x = model.generation_observation.compute_mean_x(ul_enc)
negative_log_likelihood = gqn.nn.chainer.functions.gaussian_negative_log_likelihood(
query_images, mean_x, pixel_var, pixel_ln_var)
loss_nll = cf.sum(negative_log_likelihood)
loss_nll /= args.batch_size
loss_kld /= args.batch_size
loss = loss_nll + loss_kld
model.cleargrads()
loss.backward()
optimizer.update(current_training_step)
if comm.rank == 0:
printr(
"Iteration {}: Subset {} / {}: Batch {} / {} - loss: nll: {:.3f} kld: {:.3f} - lr: {:.4e} - sigma_t: {:.6f}".
format(iteration + 1, subset_loop * comm.size + 1,
len(dataset), batch_index + 1,
len(subset) // args.batch_size,
float(loss_nll.data), float(loss_kld.data),
optimizer.learning_rate, sigma_t))
sf = hyperparams.pixel_sigma_f
si = hyperparams.pixel_sigma_i
sigma_t = max(
sf + (si - sf) *
(1.0 - current_training_step / hyperparams.pixel_n), sf)
pixel_var[...] = sigma_t**2
pixel_ln_var[...] = math.log(sigma_t**2)
total_batch += 1
current_training_step += comm.size
# current_training_step += 1
mean_kld += float(loss_kld.data)
mean_nll += float(loss_nll.data)
if comm.rank == 0:
model.serialize(args.snapshot_path)
if comm.rank == 0:
elapsed_time = time.time() - start_time
print(
"\033[2KIteration {} - loss: nll: {:.3f} kld: {:.3f} - lr: {:.4e} - sigma_t: {:.6f} - step: {} - elapsed_time: {:.3f} min".
format(iteration + 1, mean_nll / total_batch,
mean_kld / total_batch, optimizer.learning_rate,
sigma_t, current_training_step, elapsed_time / 60))
model.serialize(args.snapshot_path)
def main():
try:
os.mkdir(args.snapshot_directory)
except:
pass
images = []
files = os.listdir(args.dataset_path)
for filename in files:
image = np.load(os.path.join(args.dataset_path, filename))
image = image / 255 * 2.0 - 1.0
images.append(image)
images = np.vstack(images)
images = images.transpose((0, 3, 1, 2)).astype(np.float32)
train_dev_split = 0.9
num_images = images.shape[0]
num_train_images = int(num_images * train_dev_split)
num_dev_images = num_images - num_train_images
images_train = images[:args.batch_size]
images_dev = images[args.batch_size:]
xp = np
using_gpu = args.gpu_device >= 0
if using_gpu:
cuda.get_device(args.gpu_device).use()
xp = cp
hyperparams = HyperParameters()
hyperparams.generator_share_core = args.generator_share_core
hyperparams.generator_share_prior = args.generator_share_prior
hyperparams.generator_generation_steps = args.generation_steps
hyperparams.inference_share_core = args.inference_share_core
hyperparams.inference_share_posterior = args.inference_share_posterior
hyperparams.layer_normalization_enabled = args.layer_normalization
hyperparams.pixel_n = args.pixel_n
hyperparams.chz_channels = args.chz_channels
hyperparams.inference_channels_downsampler_x = args.channels_downsampler_x
hyperparams.pixel_sigma_i = args.initial_pixel_sigma
hyperparams.pixel_sigma_f = args.final_pixel_sigma
hyperparams.chrz_size = (32, 32)
hyperparams.save(args.snapshot_directory)
hyperparams.print()
model = Model(hyperparams, snapshot_directory=args.snapshot_directory)
if using_gpu:
model.to_gpu()
optimizer = AdamOptimizer(model.parameters,
lr_i=args.initial_lr,
lr_f=args.final_lr)
optimizer.print()
sigma_t = hyperparams.pixel_sigma_i
pixel_var = xp.full((args.batch_size, 3) + hyperparams.image_size,
sigma_t**2,
dtype="float32")
pixel_ln_var = xp.full((args.batch_size, 3) + hyperparams.image_size,
math.log(sigma_t**2),
dtype="float32")
num_pixels = images.shape[1] * images.shape[2] * images.shape[3]
figure = plt.figure(figsize=(20, 4))
axis_1 = figure.add_subplot(1, 5, 1)
axis_2 = figure.add_subplot(1, 5, 2)
axis_3 = figure.add_subplot(1, 5, 3)
axis_4 = figure.add_subplot(1, 5, 4)
axis_5 = figure.add_subplot(1, 5, 5)
for iteration in range(args.training_steps):
x = to_gpu(images_train)
loss_kld = 0
z_t_params_array, r_final = model.generate_z_params_and_x_from_posterior(
x)
for params in z_t_params_array:
mean_z_q, ln_var_z_q, mean_z_p, ln_var_z_p = params
kld = draw.nn.functions.gaussian_kl_divergence(
mean_z_q, ln_var_z_q, mean_z_p, ln_var_z_p)
loss_kld += cf.sum(kld)
mean_x_enc = r_final
negative_log_likelihood = draw.nn.functions.gaussian_negative_log_likelihood(
x, mean_x_enc, pixel_var, pixel_ln_var)
loss_nll = cf.sum(negative_log_likelihood)
loss_mse = cf.mean_squared_error(mean_x_enc, x)
loss_nll /= args.batch_size
loss_kld /= args.batch_size
loss = loss_nll + loss_kld
loss = loss_nll
model.cleargrads()
loss.backward()
optimizer.update(iteration)
sf = hyperparams.pixel_sigma_f
si = hyperparams.pixel_sigma_i
sigma_t = max(sf + (si - sf) * (1.0 - iteration / hyperparams.pixel_n),
sf)
pixel_var[...] = sigma_t**2
pixel_ln_var[...] = math.log(sigma_t**2)
model.serialize(args.snapshot_directory)
print(
"\033[2KIteration {} - loss: nll_per_pixel: {:.6f} - mse: {:.6f} - kld: {:.6f} - lr: {:.4e} - sigma_t: {:.6f}"
.format(iteration + 1,
float(loss_nll.data) / num_pixels, float(loss_mse.data),
float(loss_kld.data), optimizer.learning_rate, sigma_t))
if iteration % 10 == 0:
axis_1.imshow(make_uint8(x[0]))
axis_2.imshow(make_uint8(mean_x_enc.data[0]))
x_dev = images_dev[random.choice(range(num_dev_images))]
axis_3.imshow(make_uint8(x_dev))
with chainer.using_config("train", False), chainer.using_config(
"enable_backprop", False):
x_dev = to_gpu(x_dev)[None, ...]
_, r_final = model.generate_z_params_and_x_from_posterior(
x_dev)
mean_x_enc = r_final
axis_4.imshow(make_uint8(mean_x_enc.data[0]))
mean_x_d = model.generate_image(batch_size=1, xp=xp)
axis_5.imshow(make_uint8(mean_x_d[0]))
plt.pause(0.01)
def main():
##############################################
# To avoid OpenMPI bug
multiprocessing.set_start_method("forkserver")
p = multiprocessing.Process(target=print, args=("", ))
p.start()
p.join()
##############################################
try:
os.mkdir(args.snapshot_directory)
except:
pass
comm = chainermn.create_communicator()
device = comm.intra_rank
print("device", device, "/", comm.size)
cuda.get_device(device).use()
xp = cupy
dataset = gqn.data.Dataset(args.dataset_directory)
hyperparams = HyperParameters()
hyperparams.generator_share_core = args.generator_share_core
hyperparams.generator_share_prior = args.generator_share_prior
hyperparams.generator_generation_steps = args.generation_steps
hyperparams.generator_u_channels = args.u_channels
hyperparams.generator_share_upsampler = args.generator_share_upsampler
hyperparams.generator_subpixel_convolution_enabled = args.generator_subpixel_convolution_enabled
hyperparams.inference_share_core = args.inference_share_core
hyperparams.inference_share_posterior = args.inference_share_posterior
hyperparams.inference_downsampler_channels = args.inference_downsampler_channels
hyperparams.h_channels = args.h_channels
hyperparams.z_channels = args.z_channels
hyperparams.representation_channels = args.representation_channels
hyperparams.pixel_n = args.pixel_n
hyperparams.pixel_sigma_i = args.initial_pixel_variance
hyperparams.pixel_sigma_f = args.final_pixel_variance
if comm.rank == 0:
hyperparams.save(args.snapshot_directory)
print(hyperparams)
model = Model(hyperparams, snapshot_directory=args.snapshot_directory)
model.to_gpu()
optimizer = AdamOptimizer(model.parameters,
communicator=comm,
mu_i=args.initial_lr,
mu_f=args.final_lr)
if comm.rank == 0:
print(optimizer)
scheduler = Scheduler(sigma_start=args.initial_pixel_variance,
sigma_end=args.final_pixel_variance,
final_num_updates=args.pixel_n)
if comm.rank == 0:
print(scheduler)
pixel_var = xp.full((args.batch_size, 3) + hyperparams.image_size,
scheduler.pixel_variance**2,
dtype="float32")
pixel_ln_var = xp.full((args.batch_size, 3) + hyperparams.image_size,
math.log(scheduler.pixel_variance**2),
dtype="float32")
num_pixels = hyperparams.image_size[0] * hyperparams.image_size[1] * 3
random.seed(0)
subset_indices = list(range(len(dataset.subset_filenames)))
representation_shape = (
args.batch_size,
hyperparams.representation_channels) + hyperparams.chrz_size
current_training_step = 0
for iteration in range(args.training_iterations):
mean_kld = 0
mean_nll = 0
mean_mse = 0
mean_elbo = 0
total_num_batch = 0
subset_size_per_gpu = len(subset_indices) // comm.size
if len(subset_indices) % comm.size != 0:
subset_size_per_gpu += 1
start_time = time.time()
for subset_loop in range(subset_size_per_gpu):
random.shuffle(subset_indices)
subset_index = subset_indices[comm.rank]
subset = dataset.read(subset_index)
iterator = gqn.data.Iterator(subset, batch_size=args.batch_size)
for batch_index, data_indices in enumerate(iterator):
# shape: (batch, views, height, width, channels)
# range: [-1, 1]
images, viewpoints = subset[data_indices]
# (batch, views, height, width, channels) -> (batch, views, channels, height, width)
images = images.transpose((0, 1, 4, 2, 3)).astype(np.float32)
images = images / 255.0
images += np.random.uniform(
0, 1.0 / 256.0, size=images.shape).astype(np.float32)
total_views = images.shape[1]
# Sample observations
num_views = random.choice(range(total_views + 1))
if current_training_step == 0 and num_views == 0:
num_views = 1 # avoid OpenMPI error
observation_view_indices = list(range(total_views))
random.shuffle(observation_view_indices)
observation_view_indices = observation_view_indices[:num_views]
if num_views > 0:
representation = model.compute_observation_representation(
images[:, observation_view_indices],
viewpoints[:, observation_view_indices])
else:
representation = xp.zeros(representation_shape,
dtype="float32")
representation = chainer.Variable(representation)
# Sample query
query_index = random.choice(range(total_views))
query_images = images[:, query_index]
query_viewpoints = viewpoints[:, query_index]
# Transfer to gpu
query_images = to_gpu(query_images)
query_viewpoints = to_gpu(query_viewpoints)
z_t_param_array, mean_x = model.sample_z_and_x_params_from_posterior(
query_images, query_viewpoints, representation)
# Compute loss
## KL Divergence
loss_kld = chainer.Variable(xp.zeros((), dtype=xp.float32))
for params in z_t_param_array:
mean_z_q, ln_var_z_q, mean_z_p, ln_var_z_p = params
kld = gqn.functions.gaussian_kl_divergence(
mean_z_q, ln_var_z_q, mean_z_p, ln_var_z_p)
loss_kld += cf.sum(kld)
##Negative log-likelihood of generated image
loss_nll = cf.sum(
gqn.functions.gaussian_negative_log_likelihood(
query_images, mean_x, pixel_var, pixel_ln_var))
# Calculate the average loss value
loss_nll = loss_nll / args.batch_size
loss_kld = loss_kld / args.batch_size
loss = (loss_nll / scheduler.pixel_variance) + loss_kld
model.cleargrads()
loss.backward()
optimizer.update(current_training_step)
loss_nll = float(loss_nll.data) + math.log(256.0)
loss_kld = float(loss_kld.data)
elbo = -(loss_nll + loss_kld)
loss_mse = float(
cf.mean_squared_error(query_images, mean_x).data)
if comm.rank == 0:
printr(
"Iteration {}: Subset {} / {}: Batch {} / {} - elbo: {:.2f} - loss: nll: {:.2f} mse: {:.5f} kld: {:.5f} - lr: {:.4e} - pixel_variance: {:.5f} - step: {} "
.format(iteration + 1, subset_loop + 1,
subset_size_per_gpu, batch_index + 1,
len(iterator), elbo, loss_nll, loss_mse,
loss_kld, optimizer.learning_rate,
scheduler.pixel_variance,
current_training_step))
total_num_batch += 1
current_training_step += comm.size
mean_kld += loss_kld
mean_nll += loss_nll
mean_mse += loss_mse
mean_elbo += elbo
scheduler.step(current_training_step)
pixel_var[...] = scheduler.pixel_variance**2
pixel_ln_var[...] = math.log(scheduler.pixel_variance**2)
if comm.rank == 0:
model.serialize(args.snapshot_directory)
if comm.rank == 0:
elapsed_time = time.time() - start_time
mean_elbo /= total_num_batch
mean_nll /= total_num_batch
mean_mse /= total_num_batch
mean_kld /= total_num_batch
print(
"\033[2KIteration {} - elbo: {:.2f} - loss: nll: {:.2f} mse: {:.5f} kld: {:.5f} - lr: {:.4e} - pixel_variance: {:.5f} - step: {} - time: {:.3f} min"
.format(iteration + 1, mean_elbo, mean_nll, mean_mse, mean_kld,
optimizer.learning_rate, scheduler.pixel_variance,
current_training_step, elapsed_time / 60))
model.serialize(args.snapshot_directory)
def main():
try:
os.mkdir(args.snapshot_directory)
except:
pass
comm = chainermn.create_communicator()
device = comm.intra_rank
cuda.get_device(device).use()
xp = cp
images = []
files = os.listdir(args.dataset_path)
files.sort()
subset_size = int(math.ceil(len(files) / comm.size))
files = deque(files)
files.rotate(-subset_size * comm.rank)
files = list(files)[:subset_size]
for filename in files:
image = np.load(os.path.join(args.dataset_path, filename))
image = image / 256
images.append(image)
print(comm.rank, files)
images = np.vstack(images)
images = images.transpose((0, 3, 1, 2)).astype(np.float32)
train_dev_split = 0.9
num_images = images.shape[0]
num_train_images = int(num_images * train_dev_split)
num_dev_images = num_images - num_train_images
images_train = images[:num_train_images]
# To avoid OpenMPI bug
# multiprocessing.set_start_method("forkserver")
# p = multiprocessing.Process(target=print, args=("", ))
# p.start()
# p.join()
hyperparams = HyperParameters()
hyperparams.chz_channels = args.chz_channels
hyperparams.generator_generation_steps = args.generation_steps
hyperparams.generator_share_core = args.generator_share_core
hyperparams.generator_share_prior = args.generator_share_prior
hyperparams.generator_share_upsampler = args.generator_share_upsampler
hyperparams.generator_downsampler_channels = args.generator_downsampler_channels
hyperparams.inference_share_core = args.inference_share_core
hyperparams.inference_share_posterior = args.inference_share_posterior
hyperparams.inference_downsampler_channels = args.inference_downsampler_channels
hyperparams.batch_normalization_enabled = args.enable_batch_normalization
hyperparams.use_gru = args.use_gru
hyperparams.no_backprop_diff_xr = args.no_backprop_diff_xr
if comm.rank == 0:
hyperparams.save(args.snapshot_directory)
hyperparams.print()
if args.use_gru:
model = GRUModel(hyperparams,
snapshot_directory=args.snapshot_directory)
else:
model = LSTMModel(hyperparams,
snapshot_directory=args.snapshot_directory)
model.to_gpu()
optimizer = AdamOptimizer(model.parameters,
lr_i=args.initial_lr,
lr_f=args.final_lr,
beta_1=args.adam_beta1,
communicator=comm)
if comm.rank == 0:
optimizer.print()
num_pixels = images.shape[1] * images.shape[2] * images.shape[3]
dataset = draw.data.Dataset(images_train)
iterator = draw.data.Iterator(dataset, batch_size=args.batch_size)
num_updates = 0
for iteration in range(args.training_steps):
mean_kld = 0
mean_nll = 0
mean_mse = 0
start_time = time.time()
for batch_index, data_indices in enumerate(iterator):
x = dataset[data_indices]
x += np.random.uniform(0, 1 / 256, size=x.shape)
x = to_gpu(x)
z_t_param_array, x_param, r_t_array = model.sample_z_and_x_params_from_posterior(
x)
loss_kld = 0
for params in z_t_param_array:
mean_z_q, ln_var_z_q, mean_z_p, ln_var_z_p = params
kld = draw.nn.functions.gaussian_kl_divergence(
mean_z_q, ln_var_z_q, mean_z_p, ln_var_z_p)
loss_kld += cf.sum(kld)
loss_sse = 0
for r_t in r_t_array:
loss_sse += cf.sum(cf.squared_error(r_t, x))
mu_x, ln_var_x = x_param
loss_nll = cf.gaussian_nll(x, mu_x, ln_var_x)
loss_nll /= args.batch_size
loss_kld /= args.batch_size
loss_sse /= args.batch_size
loss = args.loss_beta * loss_nll + loss_kld + args.loss_alpha * loss_sse
model.cleargrads()
loss.backward(loss_scale=optimizer.loss_scale())
optimizer.update(num_updates, loss_value=float(loss.array))
num_updates += 1
mean_kld += float(loss_kld.data)
mean_nll += float(loss_nll.data)
mean_mse += float(loss_sse.data) / num_pixels / (
hyperparams.generator_generation_steps - 1)
printr(
"Iteration {}: Batch {} / {} - loss: nll_per_pixel: {:.6f} - mse: {:.6f} - kld: {:.6f} - lr: {:.4e}"
.format(
iteration + 1, batch_index + 1, len(iterator),
float(loss_nll.data) / num_pixels + math.log(256.0),
float(loss_sse.data) / num_pixels /
(hyperparams.generator_generation_steps - 1),
float(loss_kld.data), optimizer.learning_rate))
if comm.rank == 0 and batch_index > 0 and batch_index % 100 == 0:
model.serialize(args.snapshot_directory)
if comm.rank == 0:
model.serialize(args.snapshot_directory)
if comm.rank == 0:
elapsed_time = time.time() - start_time
print(
"\r\033[2KIteration {} - loss: nll_per_pixel: {:.6f} - mse: {:.6f} - kld: {:.6f} - lr: {:.4e} - elapsed_time: {:.3f} min"
.format(
iteration + 1,
mean_nll / len(iterator) / num_pixels + math.log(256.0),
mean_mse / len(iterator), mean_kld / len(iterator),
optimizer.learning_rate, elapsed_time / 60))
def main():
try:
os.mkdir(args.snapshot_path)
except:
pass
xp = np
using_gpu = args.gpu_device >= 0
if using_gpu:
cuda.get_device(args.gpu_device).use()
xp = cupy
dataset = gqn.data.Dataset(args.dataset_path)
hyperparams = HyperParameters()
hyperparams.generator_share_core = args.generator_share_core
hyperparams.generator_share_prior = args.generator_share_prior
hyperparams.generator_generation_steps = args.generation_steps
hyperparams.inference_share_core = args.inference_share_core
hyperparams.inference_share_posterior = args.inference_share_posterior
hyperparams.pixel_n = args.pixel_n
hyperparams.pixel_sigma_i = args.initial_pixel_sigma
hyperparams.pixel_sigma_f = args.final_pixel_sigma
hyperparams.save(args.snapshot_path)
hyperparams.print()
model = Model(hyperparams, snapshot_directory=args.snapshot_path)
if using_gpu:
model.to_gpu()
optimizer = Optimizer(model.parameters,
mu_i=args.initial_lr,
mu_f=args.final_lr)
optimizer.print()
if args.with_visualization:
figure = gqn.imgplot.figure()
axis1 = gqn.imgplot.image()
axis2 = gqn.imgplot.image()
axis3 = gqn.imgplot.image()
figure.add(axis1, 0, 0, 1 / 3, 1)
figure.add(axis2, 1 / 3, 0, 1 / 3, 1)
figure.add(axis3, 2 / 3, 0, 1 / 3, 1)
plot = gqn.imgplot.window(
figure, (500 * 3, 500),
"Query image / Reconstructed image / Generated image")
plot.show()
sigma_t = hyperparams.pixel_sigma_i
pixel_var = xp.full((args.batch_size, 3) + hyperparams.image_size,
sigma_t**2,
dtype="float32")
pixel_ln_var = xp.full((args.batch_size, 3) + hyperparams.image_size,
math.log(sigma_t**2),
dtype="float32")
dataset_mean, dataset_std = dataset.load_mean_and_std()
np.save(os.path.join(args.snapshot_path, "mean.npy"), dataset_mean)
np.save(os.path.join(args.snapshot_path, "std.npy"), dataset_std)
# avoid division by zero
dataset_std += 1e-12
current_training_step = 0
for iteration in range(args.training_iterations):
mean_kld = 0
mean_nll = 0
total_batch = 0
for subset_index, subset in enumerate(dataset):
iterator = gqn.data.Iterator(subset, batch_size=args.batch_size)
for batch_index, data_indices in enumerate(iterator):
# shape: (batch, views, height, width, channels)
# range: [-1, 1]
images, viewpoints = subset[data_indices]
# preprocessing
images = (images - dataset_mean) / dataset_std
# (batch, views, height, width, channels) -> (batch, views, channels, height, width)
images = images.transpose((0, 1, 4, 2, 3))
total_views = images.shape[1]
# sample number of views
num_views = random.choice(range(total_views))
query_index = random.choice(range(total_views))
if num_views > 0:
r = model.compute_observation_representation(
images[:, :num_views], viewpoints[:, :num_views])
else:
r = xp.zeros((args.batch_size, hyperparams.channels_r) +
hyperparams.chrz_size,
dtype="float32")
r = chainer.Variable(r)
query_images = images[:, query_index]
query_viewpoints = viewpoints[:, query_index]
# transfer to gpu
query_images = to_gpu(query_images)
query_viewpoints = to_gpu(query_viewpoints)
h0_gen, c0_gen, u_0, h0_enc, c0_enc = model.generate_initial_state(
args.batch_size, xp)
loss_kld = 0
hl_enc = h0_enc
cl_enc = c0_enc
hl_gen = h0_gen
cl_gen = c0_gen
ul_enc = u_0
xq = model.inference_downsampler.downsample(query_images)
for l in range(model.generation_steps):
inference_core = model.get_inference_core(l)
inference_posterior = model.get_inference_posterior(l)
generation_core = model.get_generation_core(l)
generation_piror = model.get_generation_prior(l)
h_next_enc, c_next_enc = inference_core.forward_onestep(
hl_gen, hl_enc, cl_enc, xq, query_viewpoints, r)
mean_z_q = inference_posterior.compute_mean_z(hl_enc)
ln_var_z_q = inference_posterior.compute_ln_var_z(hl_enc)
ze_l = cf.gaussian(mean_z_q, ln_var_z_q)
mean_z_p = generation_piror.compute_mean_z(hl_gen)
ln_var_z_p = generation_piror.compute_ln_var_z(hl_gen)
h_next_gen, c_next_gen, u_next_enc = generation_core.forward_onestep(
hl_gen, cl_gen, ul_enc, ze_l, query_viewpoints, r)
kld = gqn.nn.chainer.functions.gaussian_kl_divergence(
mean_z_q, ln_var_z_q, mean_z_p, ln_var_z_p)
loss_kld += cf.sum(kld)
hl_gen = h_next_gen
cl_gen = c_next_gen
ul_enc = u_next_enc
hl_enc = h_next_enc
cl_enc = c_next_enc
mean_x = model.generation_observation.compute_mean_x(ul_enc)
negative_log_likelihood = gqn.nn.chainer.functions.gaussian_negative_log_likelihood(
query_images, mean_x, pixel_var, pixel_ln_var)
loss_nll = cf.sum(negative_log_likelihood)
loss_nll /= args.batch_size
loss_kld /= args.batch_size
loss = loss_nll + loss_kld
model.cleargrads()
loss.backward()
optimizer.update(current_training_step)
if args.with_visualization and plot.closed() is False:
axis1.update(
make_uint8(query_images[0], dataset_mean, dataset_std))
axis2.update(
make_uint8(mean_x.data[0], dataset_mean, dataset_std))
with chainer.no_backprop_mode():
generated_x = model.generate_image(
query_viewpoints[None, 0], r[None, 0], xp)
axis3.update(
make_uint8(generated_x[0], dataset_mean,
dataset_std))
printr(
"Iteration {}: Subset {} / {}: Batch {} / {} - loss: nll: {:.3f} kld: {:.3f} - lr: {:.4e} - sigma_t: {:.6f}"
.format(iteration + 1,
subset_index + 1, len(dataset), batch_index + 1,
len(iterator), float(loss_nll.data),
float(loss_kld.data), optimizer.learning_rate,
sigma_t))
sf = hyperparams.pixel_sigma_f
si = hyperparams.pixel_sigma_i
sigma_t = max(
sf + (si - sf) *
(1.0 - current_training_step / hyperparams.pixel_n), sf)
pixel_var[...] = sigma_t**2
pixel_ln_var[...] = math.log(sigma_t**2)
total_batch += 1
current_training_step += 1
mean_kld += float(loss_kld.data)
mean_nll += float(loss_nll.data)
model.serialize(args.snapshot_path)
print(
"\033[2KIteration {} - loss: nll: {:.3f} kld: {:.3f} - lr: {:.4e} - sigma_t: {:.6f} - step: {}"
.format(iteration + 1, mean_nll / total_batch,
mean_kld / total_batch, optimizer.learning_rate, sigma_t,
current_training_step))
def main():
_mkdir(args.snapshot_directory)
_mkdir(args.log_directory)
meter_train = Meter()
meter_train.load(args.snapshot_directory)
#==============================================================================
# Selecting the GPU
#==============================================================================
xp = np
gpu_device = args.gpu_device
using_gpu = gpu_device >= 0
if using_gpu:
cuda.get_device(gpu_device).use()
xp = cp
#==============================================================================
# Dataset
#==============================================================================
dataset_train = Dataset(args.train_dataset_directory)
dataset_test = None
if args.test_dataset_directory is not None:
dataset_test = Dataset(args.test_dataset_directory)
#==============================================================================
# Hyperparameters
#==============================================================================
hyperparams = HyperParameters()
hyperparams.num_layers = args.generation_steps
hyperparams.generator_share_core = args.generator_share_core
hyperparams.inference_share_core = args.inference_share_core
hyperparams.h_channels = args.h_channels
hyperparams.z_channels = args.z_channels
hyperparams.u_channels = args.u_channels
hyperparams.r_channels = args.r_channels
hyperparams.image_size = (args.image_size, args.image_size)
hyperparams.representation_architecture = args.representation_architecture
hyperparams.pixel_sigma_annealing_steps = args.pixel_sigma_annealing_steps
hyperparams.initial_pixel_sigma = args.initial_pixel_sigma
hyperparams.final_pixel_sigma = args.final_pixel_sigma
hyperparams.save(args.snapshot_directory)
print(hyperparams, "\n")
#==============================================================================
# Model
#==============================================================================
model = Model(hyperparams)
model.load(args.snapshot_directory, meter_train.epoch)
if using_gpu:
model.to_gpu()
#==============================================================================
# Pixel-variance annealing
#==============================================================================
variance_scheduler = PixelVarianceScheduler(
sigma_start=args.initial_pixel_sigma,
sigma_end=args.final_pixel_sigma,
final_num_updates=args.pixel_sigma_annealing_steps)
variance_scheduler.load(args.snapshot_directory)
print(variance_scheduler, "\n")
pixel_log_sigma = xp.full(
(args.batch_size, 3) + hyperparams.image_size,
math.log(variance_scheduler.standard_deviation),
dtype="float32")
#==============================================================================
# Logging
#==============================================================================
csv = DataFrame()
csv.load(args.log_directory)
#==============================================================================
# Optimizer
#==============================================================================
optimizer = AdamOptimizer(
model.parameters,
initial_lr=args.initial_lr,
final_lr=args.final_lr,
initial_training_step=variance_scheduler.training_step)
print(optimizer, "\n")
#==============================================================================
# Visualization
#==============================================================================
fig = plt.figure(figsize=(9, 6))
axes_train = [
fig.add_subplot(2, 3, 1),
fig.add_subplot(2, 3, 2),
fig.add_subplot(2, 3, 3),
]
axes_train[0].set_title("Training Data")
axes_train[0].axis("off")
axes_train[1].set_title("Reconstruction")
axes_train[1].axis("off")
axes_train[2].set_title("Generation")
axes_train[2].axis("off")
axes_test = [
fig.add_subplot(2, 3, 4),
fig.add_subplot(2, 3, 5),
fig.add_subplot(2, 3, 6),
]
axes_test[0].set_title("Validation Data")
axes_test[0].axis("off")
axes_test[1].set_title("Reconstruction")
axes_test[1].axis("off")
axes_test[2].set_title("Generation")
axes_test[2].axis("off")
#==============================================================================
# Algorithms
#==============================================================================
def encode_scene(images, viewpoints):
# (batch, views, height, width, channels) -> (batch, views, channels, height, width)
images = images.transpose((0, 1, 4, 2, 3)).astype(np.float32)
# Sample number of views
total_views = images.shape[1]
num_views = random.choice(range(1, total_views + 1))
# Sample views
observation_view_indices = list(range(total_views))
random.shuffle(observation_view_indices)
observation_view_indices = observation_view_indices[:num_views]
observation_images = preprocess_images(
images[:, observation_view_indices])
observation_query = viewpoints[:, observation_view_indices]
representation = model.compute_observation_representation(
observation_images, observation_query)
# Sample query view
query_index = random.choice(range(total_views))
query_images = preprocess_images(images[:, query_index])
query_viewpoints = viewpoints[:, query_index]
# Transfer to gpu if necessary
query_images = to_device(query_images, gpu_device)
query_viewpoints = to_device(query_viewpoints, gpu_device)
return representation, query_images, query_viewpoints
def estimate_ELBO(query_images, z_t_param_array, pixel_mean,
pixel_log_sigma):
# KL Diverge, pixel_ln_varnce
kl_divergence = 0
for params_t in z_t_param_array:
mean_z_q, ln_var_z_q, mean_z_p, ln_var_z_p = params_t
normal_q = chainer.distributions.Normal(
mean_z_q, log_scale=ln_var_z_q)
normal_p = chainer.distributions.Normal(
mean_z_p, log_scale=ln_var_z_p)
kld_t = chainer.kl_divergence(normal_q, normal_p)
kl_divergence += cf.sum(kld_t)
kl_divergence = kl_divergence / args.batch_size
# Negative log-likelihood of generated image
batch_size = query_images.shape[0]
num_pixels_per_batch = np.prod(query_images.shape[1:])
normal = chainer.distributions.Normal(
query_images, log_scale=pixel_log_sigma)
log_px = cf.sum(normal.log_prob(pixel_mean)) / batch_size
negative_log_likelihood = -log_px
# Empirical ELBO
ELBO = log_px - kl_divergence
# https://arxiv.org/abs/1604.08772 Section.2
# https://www.reddit.com/r/MachineLearning/comments/56m5o2/discussion_calculation_of_bitsdims/
bits_per_pixel = -(ELBO / num_pixels_per_batch - np.log(256)) / np.log(
2)
return ELBO, bits_per_pixel, negative_log_likelihood, kl_divergence
#==============================================================================
# Training iterations
#==============================================================================
dataset_size = len(dataset_train)
np.random.seed(0)
cp.random.seed(0)
start_training = True
for epoch in range(meter_train.epoch, args.epochs):
print("Epoch {}/{}:".format(
epoch + 1,
args.epochs,
))
meter_train.next_epoch()
for subset_index, subset in enumerate(dataset_train):
iterator = Iterator(subset, batch_size=args.batch_size)
for batch_index, data_indices in enumerate(iterator):
#------------------------------------------------------------------------------
# Scene encoder
#------------------------------------------------------------------------------
# images.shape: (batch, views, height, width, channels)
images, viewpoints = subset[data_indices]
representation, query_images, query_viewpoints = encode_scene(
images, viewpoints)
#------------------------------------------------------------------------------
# Compute empirical ELBO
#------------------------------------------------------------------------------
# Compute distribution parameterws
(z_t_param_array
) = model.sample_z_and_x_params_from_posterior(
query_images, query_viewpoints, representation)
# # Compute ELBO
# (ELBO, bits_per_pixel, negative_log_likelihood,
# kl_divergence) = estimate_ELBO(query_images, z_t_param_array,
# pixel_mean, pixel_log_sigma)
#------------------------------------------------------------------------------
# Update parameters
#------------------------------------------------------------------------------
loss = -ELBO
model.cleargrads()
loss.backward()
# if start_training:
# g = chainer.computational_graph.build_computational_graph(pixel_mean)
# with open(os.path.join(args.snapshot_directory,'cg.dot'), 'w') as o:
# o.write(g.dump())
# start_training = False
# exit()
optimizer.update(meter_train.num_updates)
#------------------------------------------------------------------------------
# Logging
#------------------------------------------------------------------------------
with chainer.no_backprop_mode():
mean_squared_error = cf.mean_squared_error(
query_images, pixel_mean)
meter_train.update(
ELBO=float(ELBO.data),
bits_per_pixel=float(bits_per_pixel.data),
negative_log_likelihood=float(
negative_log_likelihood.data),
kl_divergence=float(kl_divergence.data),
mean_squared_error=float(mean_squared_error.data))
#------------------------------------------------------------------------------
# Annealing
#------------------------------------------------------------------------------
variance_scheduler.update(meter_train.num_updates)
pixel_log_sigma[...] = math.log(
variance_scheduler.standard_deviation)
if subset_index % 100 == 0:
print(" Subset {}/{}:".format(
subset_index + 1,
dataset_size,
))
print(" {}".format(meter_train))
print(" lr: {} - sigma: {}".format(
optimizer.learning_rate,
variance_scheduler.standard_deviation))
#------------------------------------------------------------------------------
# Visualization
#------------------------------------------------------------------------------
if args.visualize:
axes_train[0].imshow(
make_uint8(query_images[0]), interpolation="none")
axes_train[1].imshow(
make_uint8(pixel_mean.data[0]), interpolation="none")
with chainer.no_backprop_mode():
generated_x = model.generate_image(query_viewpoints[None, 0],
representation[None, 0])
axes_train[2].imshow(
make_uint8(generated_x[0]), interpolation="none")
#------------------------------------------------------------------------------
# Validation
#------------------------------------------------------------------------------
meter_test = None
if dataset_test is not None:
meter_test = Meter()
batch_size_test = args.batch_size * 6
pixel_log_sigma_test = xp.full(
(batch_size_test, 3) + hyperparams.image_size,
math.log(variance_scheduler.standard_deviation),
dtype="float32")
with chainer.no_backprop_mode():
for subset in dataset_test:
iterator = Iterator(subset, batch_size=batch_size_test)
for data_indices in iterator:
images, viewpoints = subset[data_indices]
# Scene encoder
representation, query_images, query_viewpoints = encode_scene(
images, viewpoints)
# Compute empirical ELBO
(z_t_param_array, pixel_mean
) = model.sample_z_and_x_params_from_posterior(
query_images, query_viewpoints, representation)
(ELBO, bits_per_pixel, negative_log_likelihood,
kl_divergence) = estimate_ELBO(
query_images, z_t_param_array, pixel_mean,
pixel_log_sigma_test)
mean_squared_error = cf.mean_squared_error(
query_images, pixel_mean)
# Logging
meter_test.update(
ELBO=float(ELBO.data),
bits_per_pixel=float(bits_per_pixel.data),
negative_log_likelihood=float(
negative_log_likelihood.data),
kl_divergence=float(kl_divergence.data),
mean_squared_error=float(mean_squared_error.data))
print(" Test:")
print(" {} - done in {:.3f} min".format(
meter_test,
meter_test.elapsed_time,
))
if args.visualize:
axes_test[0].imshow(
make_uint8(query_images[0]), interpolation="none")
axes_test[1].imshow(
make_uint8(pixel_mean.data[0]), interpolation="none")
with chainer.no_backprop_mode():
generated_x = model.generate_image(
query_viewpoints[None, 0], representation[None, 0])
axes_test[2].imshow(
make_uint8(generated_x[0]), interpolation="none")
if args.visualize:
plt.pause(1e-10)
csv.append(epoch, meter_train, meter_test)
#------------------------------------------------------------------------------
# Snapshot
#------------------------------------------------------------------------------
model.save(args.snapshot_directory, epoch)
variance_scheduler.save(args.snapshot_directory)
meter_train.save(args.snapshot_directory)
csv.save(args.log_directory)
print("Epoch {} done in {:.3f} min".format(
epoch + 1,
meter_train.epoch_elapsed_time,
))
print(" {}".format(meter_train))
print(" lr: {} - sigma: {} - training_steps: {}".format(
optimizer.learning_rate,
variance_scheduler.standard_deviation,
meter_train.num_updates,
))
print(" Time elapsed: {:.3f} min".format(meter_train.elapsed_time))
