bn4 = BatchNormLayer(net3, sess, beta=beta, gamma=gamma)
net4 = HiddenLayer(bn4,
1,
sess,
non_liniarity=non_lin,
bactivate=bactivate,
unsupervised_cost=.001,
noise_std=noise_std)
bn5 = BatchNormLayer(net4, sess, beta=beta, gamma=gamma)
outputNet = HiddenLayer(bn5,
10,
sess,
non_liniarity=tf.sigmoid,
bactivate=False,
supervised_cost=1.)
trainer = CategoricalTrainer(outputNet, 0.15)
trainPolicy = TrainPolicy(trainer,
data,
batch_size,
max_iterations=3000,
grow_after_turns_without_improvement=2,
start_grow_epoch=1,
learn_rate_decay=0.99,
learn_rate_boost=0.01,
back_loss_on_misclassified_only=True)
trainPolicy.run_full()
print trainer.accuracy(data.test.features, data.test.labels)
with tf.Session() as sess:
net, trainer = create_network(sess, hidden_layers)
# train minimal model on mnist/load checkpoints
if not os.path.exists(checkpoint_path):
os.mkdir(checkpoint_path)
saver = tf.train.Saver()
checkpoints = tf.train.get_checkpoint_state(checkpoint_path)
if checkpoints:
saver.restore(sess, checkpoints.model_checkpoint_path)
print("Loaded checkpoints %s" % checkpoints.model_checkpoint_path)
else:
print("retraining network")
tp = TrainPolicy(trainer,
data,
batch_size,
learn_rate_decay=learn_rate_decay)
tp.train_till_convergence()
if SAVE:
saver.save(sess, checkpoint_path + "/network")
# get train error
print("train error ",
trainer.accuracy(data.validation.features, data.validation.labels))
# get reconstruction errors
print trainer.back_losses_per_layer(data.train.features)
# get error just on miss-classifications
print trainer.back_losses_per_layer(data.train.features,
inputs = tf.placeholder(tf.float32, shape=(None, 784))
bactivate = True
noise_std = 0.3
beta = 0.5
gamma = 0.5
non_lin = tf.nn.sigmoid
input_layer = InputLayer(inputs)
bn1 = BatchNormLayer(input_layer, sess, beta=beta, gamma=gamma)
net1 = Layer(bn1, 1, sess, non_liniarity=non_lin, bactivate=bactivate, unsupervised_cost=.001, noise_std=noise_std)
bn2 = BatchNormLayer(net1, sess, beta=beta, gamma=gamma)
net2 = Layer(bn2, 1, sess, non_liniarity=non_lin, bactivate=bactivate, unsupervised_cost=.001, noise_std=noise_std)
bn3 = BatchNormLayer(net2, sess, beta=beta, gamma=gamma)
net3 = Layer(bn3, 1, sess, non_liniarity=non_lin, bactivate=bactivate, unsupervised_cost=.001, noise_std=noise_std)
bn4 = BatchNormLayer(net3, sess, beta=beta, gamma=gamma)
net4 = Layer(bn4, 1, sess, non_liniarity=non_lin, bactivate=bactivate, unsupervised_cost=.001, noise_std=noise_std)
bn5 = BatchNormLayer(net4, sess, beta=beta, gamma=gamma)
outputNet = Layer(bn5, 10, sess, non_liniarity=tf.sigmoid, bactivate=False, supervised_cost=1.)
trainer = CategoricalTrainer(outputNet, 0.15)
trainPolicy = TrainPolicy(trainer, data, batch_size, max_iterations=3000,
grow_after_turns_without_improvement=2,
start_grow_epoch=1,
learn_rate_decay=0.99,
learn_rate_boost=0.01,
back_loss_on_misclassified_only=True)
trainPolicy.run_full()
print trainer.accuracy(data.test.images, data.test.labels)
with tf.Session() as sess:
net, trainer = create_network(sess, hidden_layers)
# train minimal model on mnist/load checkpoints
if not os.path.exists(checkpoint_path):
os.mkdir(checkpoint_path)
saver = tf.train.Saver()
checkpoints = tf.train.get_checkpoint_state(checkpoint_path)
if checkpoints:
saver.restore(sess, checkpoints.model_checkpoint_path)
print ("Loaded checkpoints %s" % checkpoints.model_checkpoint_path)
else:
print ("retraining network")
tp = TrainPolicy(trainer, data, batch_size, learn_rate_decay=learn_rate_decay)
tp.train_till_convergence()
if SAVE:
saver.save(sess, checkpoint_path + "/network")
# get train error
print ("train error ", trainer.accuracy(data.validation.images, data.validation.labels))
# get reconstruction errors
print trainer.back_losses_per_layer(data.train.images)
# get error just on miss-classifications
print trainer.back_losses_per_layer(data.train.images, misclassification_only=True, labels=data.train.labels)
results = {}