def run_autoencoder(optimizer):
""" Runs the autoencoder model using the specified optimizer.
Parameters
----------
optimizer : RMSProp/Adam
Optimization algorithm to be used for parameter learning
"""
optimizer = Adam(learning_rate=0.03) if optimizer == 'adam' else RMSProp(
learning_rate=0.05)
train_matrix, val_matrix = get_training_and_val_data()
model = Autoencoder(input_dim=train_matrix.shape[1])
model.print_summary()
model.compile(optimizer)
errors = model.fit(train_matrix,
train_matrix,
num_epochs=60,
val_set=(val_matrix, val_matrix),
early_stopping=True)
plot_losses(errors['training'], errors['validation'])
neuron_num = model.model.layers[0].optimizer.reference_index
learning_rates = model.model.layers[0].optimizer.learning_rates
plot_learning_rates(learning_rates['weights'], learning_rates['bias'],
neuron_num)
def train(x_train, learning_rate, batch_size, epochs):
autoencoder = Autoencoder(input_shape=(28, 28, 1),
conv_filters=(32, 64, 64, 64),
conv_kernels=(3, 3, 3, 3),
conv_strides=(1, 2, 2, 1),
latent_space_dim=2)
autoencoder.summary()
autoencoder.compile(learning_rate)
autoencoder.train(x_train, batch_size, epochs)
return autoencoder
def train_grid(x_train,
learning_rate,
batch_size,
epochs,
latent_space_dim=32):
autoencoder = Autoencoder(input_shape=(23, 23, 5),
conv_filters=(16, 32, 32, 32),
conv_kernels=(3, 3, 3, 3),
conv_strides=(1, 1, 1, 1),
latent_space_dim=latent_space_dim,
name="Autoencoder_CNN_Grid_" +
str(latent_space_dim))
autoencoder.summary()
autoencoder.compile(learning_rate)
history = autoencoder.train(x_train, batch_size, epochs)
return autoencoder, history
RUN_FOLDER = 'run'
(x_train, y_train), (x_test, y_test) = mnist.load_data()
x_train = x_train[:, :, :, np.newaxis].astype('float32') / 255.
AE = Autoencoder(input_dim=(28, 28, 1),
encoder_conv_filters=[32, 64, 64, 64],
encoder_conv_kernel_size=[3, 3, 3, 3],
encoder_conv_strides=[1, 2, 2, 1],
decoder_conv_t_filters=[64, 64, 32, 1],
decoder_conv_t_kernel_size=[3, 3, 3, 3],
decoder_conv_t_strides=[1, 2, 2, 1],
z_dim=2)
AE.save(RUN_FOLDER)
AE.encoder.summary()
AE.decoder.summary()
LEARNING_RATE = 0.0005
BATCH_SIZE = 8
INITIAL_EPOCH = 0
AE.compile(LEARNING_RATE)
AE.train(x_train[:60000],
batch_size=BATCH_SIZE,
epochs=10,
run_folder=RUN_FOLDER,
initial_epoch=INITIAL_EPOCH)
AE.save_networks(RUN_FOLDER)
