v1 * v2,
h1 * h2,
data=train_data,
visible_activation_function=act.Identity(),
hidden_activation_function=act.Sigmoid(),
cost_function=cost.SquaredError(),
initial_weights=0.01,
initial_visible_bias=0.0,
initial_hidden_bias=-2.0,
# Set initially the units to be inactive, speeds up learning a little bit
initial_visible_offsets=0.0,
initial_hidden_offsets=0.02,
dtype=numx.float64)
# Initialized gradient descent trainer
trainer = aeTrainer.GDTrainer(ae)
# Train model
print 'Training'
print 'Epoch\tRE train\t\tRE test\t\t\tSparsness train\t\tSparsness test '
for epoch in range(0, max_epochs + 1, 1):
# Shuffle data
train_data = numx.random.permutation(train_data)
# Print reconstruction errors and sparseness for Training and test data
print epoch, ' \t\t', numx.mean(ae.reconstruction_error(train_data)), \
' \t', numx.mean(ae.reconstruction_error(test_data)),\
' \t', numx.mean(ae.encode(train_data)), \
' \t', numx.mean(ae.encode(test_data))
for b in range(0, train_data.shape[0], batch_size):
def perform_training(self,
ae,
data,
epsilon,
momentum,
update_visible_offsets,
update_hidden_offsets,
corruptor,
reg_L1Norm,
reg_L2Norm,
reg_sparseness,
desired_sparseness,
reg_contractive,
reg_slowness,
data_next,
restrict_gradient,
restriction_norm,
num_epochs=1000):
numx.random.seed(42)
tr = TRAINER.GDTrainer(ae)
tr.train(data=data,
num_epochs=num_epochs,
epsilon=epsilon,
momentum=momentum,
update_visible_offsets=update_visible_offsets,
update_hidden_offsets=update_hidden_offsets,
corruptor=corruptor,
reg_L1Norm=reg_L1Norm,
reg_L2Norm=reg_L2Norm,
reg_sparseness=reg_sparseness,
desired_sparseness=desired_sparseness,
reg_contractive=reg_contractive,
reg_slowness=reg_slowness,
data_next=data_next,
restrict_gradient=restrict_gradient,
restriction_norm=restriction_norm)
rec1 = numx.mean(
ae.energy(x=data,
contractive_penalty=reg_contractive,
sparse_penalty=reg_sparseness,
desired_sparseness=desired_sparseness,
x_next=data_next,
slowness_penalty=reg_slowness))
tr.train(data=data,
num_epochs=10,
epsilon=epsilon,
momentum=momentum,
update_visible_offsets=update_visible_offsets,
update_hidden_offsets=update_hidden_offsets,
corruptor=corruptor,
reg_L1Norm=reg_L1Norm,
reg_L2Norm=reg_L2Norm,
reg_sparseness=reg_sparseness,
desired_sparseness=desired_sparseness,
reg_contractive=reg_contractive,
reg_slowness=reg_slowness,
data_next=data_next,
restrict_gradient=restrict_gradient,
restriction_norm=restriction_norm)
rec2 = numx.mean(
ae.energy(x=data,
contractive_penalty=reg_contractive,
sparse_penalty=reg_sparseness,
desired_sparseness=desired_sparseness,
x_next=data_next,
slowness_penalty=reg_slowness))
assert numx.all(rec1 - rec2 >= 0.0)
