print "loading", discriminator_sequence_filename
with open(discriminator_sequence_filename, "r") as f:
try:
params = json.load(f)
except Exception as e:
raise Exception(
"could not load {}".format(discriminator_sequence_filename))
else:
config = DiscriminatorParams()
config.weight_init_std = 0.001
config.weight_initializer = "Normal"
config.use_weightnorm = False
config.nonlinearity = "elu"
config.optimizer = "Adam"
config.learning_rate = 0.0001
config.momentum = 0.5
config.gradient_clipping = 10
config.weight_decay = 0
config.use_feature_matching = False
config.use_minibatch_discrimination = False
discriminator = Sequential(weight_initializer=config.weight_initializer,
weight_init_std=config.weight_init_std)
discriminator.add(gaussian_noise(std=0.3))
discriminator.add(
Convolution2D(3,
32,
ksize=4,
stride=2,
pad=1,
use_weightnorm=config.use_weightnorm))
try:
discriminator_params = json.load(f)
except Exception as e:
raise Exception(
"could not load {}".format(discriminator_sequence_filename))
else:
config = DiscriminatorParams()
config.a = 0
config.b = 1
config.c = 1
config.weight_std = 0.001
config.weight_initializer = "Normal"
config.nonlinearity = "leaky_relu"
config.optimizer = "adam"
config.learning_rate = 0.0001
config.momentum = 0.1
config.gradient_clipping = 1
config.weight_decay = 0
discriminator = Sequential()
discriminator.add(Linear(None, 500))
# discriminator.add(gaussian_noise(std=0.5))
discriminator.add(Activation(config.nonlinearity))
# discriminator.add(BatchNormalization(500))
discriminator.add(Linear(None, 500))
discriminator.add(Activation(config.nonlinearity))
# discriminator.add(BatchNormalization(500))
discriminator.add(Linear(None, 1))
discriminator_params = {
"config": config.to_dict(),