dtype=c.float_type))
config = tf.ConfigProto(gpu_options=tf.GPUOptions(allow_growth=True))
session = tf.Session(config=config)
kern = gplvm.SEKernel(session=session,
alpha=flags['kernel_alpha'],
gamma=flags['kernel_gamma'],
ARD=flags['kernel_ard'],
Q=flags['q'])
kern.restore(flags['kernel_ckpt'])
layer = gplvm.GPLVM(Y=Y,
Q=flags['q'],
kern=kern,
noise_variance=flags['noise_variance'],
x_test_var=x_test,
session=session,
name=flags['gplvm_name'])
layer.restore(flags['gplvm_ckpt'])
layer.build_model()
learning_rate = flags['test_generalisation_learning_rate']
optimizer = tf.train.GradientDescentOptimizer(
learning_rate=flags['test_generalisation_learning_rate'])
table_path = os.path.join(flags['test_generalisation_plots_dir'], 'table')
layer.test_generalisation(
test_data=test_data_batch,
Y = training_data.next_batch()
config = tf.ConfigProto(gpu_options=tf.GPUOptions(allow_growth=True))
session = tf.Session(config=config)
kern = gplvm.SEKernel(session=session,
alpha=flags['kernel_alpha'],
gamma=flags['kernel_gamma'],
ARD=flags['kernel_ard'],
Q=flags['q'])
layer = gplvm.GPLVM(Y=Y,
Q=flags['q'],
kern=kern,
noise_variance=flags['noise_variance'],
latent_point_plotter=lp_plotter,
latent_sample_plotter=ls_plotter,
session=session,
name=flags['gplvm_name'])
layer.build_model()
optimizer = tf.train.AdamOptimizer(learning_rate=flags['learning_rate'])
summary_writer = tf.summary.FileWriter(flags['summary_writer_file'],
session.graph)
layer.optimize(optimizer=optimizer,
num_iterations=flags['num_iterations'],
eval_interval=flags['eval_interval'],
ckpt_interval=flags['ckpt_interval'],
batch_size=flags['training_batch_size'],
log_epochs=flags['data_log_epochs'],
name='TrainingData')
Y = training_data.next_batch()
config = tf.ConfigProto(gpu_options=tf.GPUOptions(allow_growth=True))
session = tf.Session(config=config)
kern = gplvm.SEKernel(session=session,
alpha=flags['kernel_alpha'],
gamma=flags['kernel_gamma'],
ARD=flags['kernel_ard'],
Q=flags['q'])
kern.restore(flags['kernel_ckpt'])
layer = gplvm.GPLVM(Y=Y,
Q=flags['q'],
kern=kern,
noise_variance=flags['noise_variance'],
latent_space_explorer=latent_space_explorer,
session=session,
name=flags['gplvm_name'])
layer.restore(flags['gplvm_ckpt'])
layer.build_model()
layer.explore_2D_latent_space()
session.close()