IMAGE_SIZE = 32
DATASET_LEN = 20000
EPOCHS = 30000
device = 'cuda'
gloss_cfg = LossConfig('BCE')
dloss_cfg = LossConfig('BCE')
sampler_cfg = SamplerConfig(
name='Onehot',
out_shape=DATASET_LEN,
latent_dim=LATENT_DIM
)
goptim_cfg = OptimConfig('Adam', lr=1e-3)
doptim_cfg = OptimConfig('Adam', lr=1e-3)
dataset_cfg = DatasetConfig('MNIST', train=True, stack=False, along_width=False, size=32)
loader_cfg = LoaderConfig('naive', batch_size=128, shuffle=True)
gen_cfg = ModelConfig('EDCG', input_size=LATENT_DIM, hidden_size=128, output_size=1, data_num=DATASET_LEN)
dis_cfg = ModelConfig('DCD', input_size=1, hidden_size=128, output_size=1)
train_data_cfg = DatasetConfig('MNIST', stack=False, train=True, along_width=False)
test_data_cfg = DatasetConfig('MNIST', stack=False, train=False, along_width=False)
train_loader = LoaderConfig('naive', batch_size=128, shuffle=True)
test_loader = LoaderConfig('naive', batch_size=128, shuffle=True)
util_cfg = UtilityModelConfig('NaiveClassifier', False, '/home/bourgan/gan_dev/checkpoints/mnist_naive.pth.tar',
15, train_loader, test_loader, train_data_cfg, test_data_cfg, 32, 'cuda')
gan_cfg = GanConfig(
name='StackMNISTLPG', gen_cfg=gen_cfg, dis_cfg=dis_cfg,
PRELOAD_LEN = 20000
EPOCHS = 50000
device = 'cuda'
gloss_cfg = LossConfig('BCE')
dloss_cfg = LossConfig('BCE')
sampler_cfg = SamplerConfig(
name='Onehot',
out_shape=PRELOAD_LEN,
latent_dim=LATENT_DIM,
random_sampling=False
)
goptim_cfg = OptimConfig('Adam', lr=0.0002)
doptim_cfg = OptimConfig('Adam', lr=0.0002)
dataset_cfg = DatasetConfig('CelebA', dataset_len=DATASET_LEN, image_size=IMAGE_SIZE, preload_len=PRELOAD_LEN)
loader_cfg = LoaderConfig('naive', batch_size=128, shuffle=True)
gen_cfg = ModelConfig(
'EDCG', input_size=LATENT_DIM,
hidden_size=128, output_size=3,
data_num=PRELOAD_LEN, out_64=IMAGE_SIZE == 64
)
dis_cfg = ModelConfig('DCD', input_size=3, hidden_size=64, output_size=1, out_64=IMAGE_SIZE == 64)
gan_cfg = GanConfig(
name='CelebAGAN', gen_cfg=gen_cfg, dis_cfg=dis_cfg,
gen_step=1, dis_step=1, gan_epoch=EPOCHS, loader_cfg=loader_cfg,
dataset_cfg=dataset_cfg, gloss_cfg=gloss_cfg, dloss_cfg=dloss_cfg,
goptim_cfg=goptim_cfg, doptim_cfg=doptim_cfg, label_smooth=False,
sampler_cfg=sampler_cfg, dist_loss=False, device=device
device = 'cuda'
gloss_cfg = LossConfig('BCE')
dloss_cfg = LossConfig('BCE')
sampler_cfg = SamplerConfig(
name='Onehot',
out_shape=dataset_len,
)
goptim_cfg = OptimConfig('Adam', lr=1e-3)
doptim_cfg = OptimConfig('Adam', lr=1e-3)
dataset_cfg = DatasetConfig('Spiral',
mode=mode,
sig=1,
num_per_mode=dataset_len // mode,
return_idx=True)
# dataset_cfg = DatasetConfig('GMM', mode=mode, sig=1, num_per_mode=dataset_len // mode, return_idx=True)
loader_cfg = LoaderConfig('naive', batch_size=128, shuffle=True)
gen_cfg = ModelConfig('MLG',
input_size=latent_dim,
hidden_size=32,
output_size=2)
dis_cfg = ModelConfig('MLD', input_size=2, hidden_size=32, output_size=1)
gan_cfg = GanConfig(name='LearningPrior2',
gen_cfg=gen_cfg,
dis_cfg=dis_cfg,
gen_step=1,
device = 'cuda'
gloss_cfg = LossConfig('BCE')
dloss_cfg = LossConfig('BCE')
sampler_cfg = SamplerConfig(
name='Onehot',
out_shape=DATASET_LEN,
)
goptim_cfg = OptimConfig('Adam', lr=1e-3)
doptim_cfg = OptimConfig('Adam', lr=1e-3)
dataset_cfg = DatasetConfig('Spiral',
mode=MODE,
sig=1,
num_per_mode=DATASET_LEN // MODE,
return_idx=True)
# dataset_cfg = DatasetConfig('GMM', mode=mode, sig=1, num_per_mode=dataset_len // mode, return_idx=True)
loader_cfg = LoaderConfig('naive', batch_size=128, shuffle=True)
gen_cfg = ModelConfig('EMLG',
input_size=LATENT_DIM,
hidden_size=32,
output_size=2,
data_num=DATASET_LEN)
dis_cfg = ModelConfig('MLD', input_size=2, hidden_size=32, output_size=1)
gan_cfg = GanConfig(name='Embedding',
gen_cfg=gen_cfg,
dis_cfg=dis_cfg,
BATCH_SIZE = 1
label_len = int(args.label_len)
EXP_NAME = 'refocus_final_' + str(label_len)
# logger = SummaryWriter('./log/' + EXP_NAME)
LR = 2e-4
gloss_cfg = LossConfig('MSE')
dloss_cfg = LossConfig('MSE')
goptim_cfg = OptimConfig('Adam', lr=LR, beta=(0.5, 0.999))
doptim_cfg = OptimConfig('Adam', lr=LR, beta=(0.5, 0.999))
dataset_cfg = DatasetConfig('FlowerFull',
size=IMG_SIZE,
num=label_len,
train=False)
loader_cfg = LoaderConfig('naive', batch_size=1, shuffle=False)
# gen_cfg = ModelConfig('ResNetGen', input_nc=3, output_nc=3, ngf=64, n_blocks=6)
gen_cfg = ModelConfig('UNetGen',
input_nc=3,
output_nc=3,
num_downs=6,
ngf=64,
use_resizeconv=True,
ex_label=True,
label_len=label_len)
dis_cfg = ModelConfig('PatchDis', input_nc=6)
gan_cfg = GanConfig(name='Pix2Pix',
mode = 20
latent_dim = 10
device = 'cuda'
gloss_cfg = LossConfig('BCE')
dloss_cfg = LossConfig('BCE')
sampler_cfg = SamplerConfig(
name='Gaussian',
out_shape=latent_dim,
)
goptim_cfg = OptimConfig('Adam', lr=1e-3)
doptim_cfg = OptimConfig('Adam', lr=1e-3)
dataset_cfg = DatasetConfig('Spiral', mode=mode, sig=1, num_per_mode=dataset_len // mode)
# dataset_cfg = DatasetConfig('GMM', mode=mode, sig=1, num_per_mode=dataset_len // mode)
loader_cfg = LoaderConfig('naive', batch_size=128, shuffle=True)
gen_cfg = ModelConfig('MLG', input_size=latent_dim, hidden_size=32, output_size=2)
dis_cfg = ModelConfig('MLD', input_size=2, hidden_size=32, output_size=1)
gan_cfg = GanConfig(
name='Vanilla', gen_cfg=gen_cfg, dis_cfg=dis_cfg,
gen_step=1, dis_step=1, gan_epoch=15000, loader_cfg=loader_cfg,
dataset_cfg=dataset_cfg, gloss_cfg=gloss_cfg, dloss_cfg=dloss_cfg,
goptim_cfg=goptim_cfg, doptim_cfg=doptim_cfg, label_smooth=False,
sampler_cfg=sampler_cfg, device=device
)
gan_schema = GanSchema()
gan_desc = gan_schema.dump(gan_cfg)
if __name__ == '__main__':
device = 'cuda'
GAN_EPOCHS = 10000
IMG_SIZE = 256
BATCH_SIZE = 2
# gloss_cfg = LossConfig('BCEWithLogits')
# dloss_cfg = LossConfig('BCEWithLogits')
gloss_cfg = LossConfig('MSE')
dloss_cfg = LossConfig('MSE')
goptim_cfg = OptimConfig('Adam', lr=1e-3)
doptim_cfg = OptimConfig('Adam', lr=1e-3)
dataset_cfg = DatasetConfig('Flower',
suffix=('_center', '_1'),
size=IMG_SIZE)
loader_cfg = LoaderConfig('naive', batch_size=BATCH_SIZE, shuffle=True)
# gen_cfg = ModelConfig('ResNetGen', input_nc=3, output_nc=3, ngf=64, n_blocks=6)
gen_cfg = ModelConfig('UNetGen',
input_nc=3,
output_nc=3,
num_downs=8,
ngf=64)
dis_cfg = ModelConfig('PatchDis', input_nc=6)
gan_cfg = GanConfig(name='Pix2Pix',
gen_cfg=gen_cfg,
dis_cfg=dis_cfg,
gen_step=1,