func_name='training.loss.D_wgan_gp') # Options for discriminator loss.
dataset = EasyDict() # Options for load_dataset().
sched = EasyDict() # Options for TrainingSchedule.
grid = EasyDict(
size='1080p',
layout='random') # Options for setup_snapshot_image_grid().
metrics = [metric_base.fid50k] # Options for MetricGroup.
submit_config = dnnlib.SubmitConfig() # Options for dnnlib.submit_run().
tf_config = {'rnd.np_random_seed': 1000} # Options for tflib.init_tf().
# Puzzle Modes: choose one
desc += '-2parts-faces-celebahq256'
dataset = EasyDict(tfrecord_dir='celebahq', resolution=256)
train.mirror_augment = True
G.mode = '2parts-faces'
G.latents_sizes = [128, 384]
G.firstblock_res = 8
#desc += '-5parts-faces-celebahq256'; dataset = EasyDict(tfrecord_dir='celebahq', resolution=256); train.mirror_augment = True; G.mode = '5parts-faces'; G.latents_sizes = [288, 64, 64, 32, 64]; G.firstblock_res = 8;
#desc += '-2parts-bedrooms-lsun128'; dataset = EasyDict(tfrecord_dir='lsun-bedroom-100k', resolution=128); train.mirror_augment = False; G.mode= '2parts-bedrooms'; G.latents_sizes = [256, 256]; G.firstblock_res = 8;
#desc += '-4parts-digits-mnist32'; dataset = EasyDict(tfrecord_dir='mnist', resolution=32); train.mirror_augment = True; G.mode= '4parts-digits'; G.latents_sizes = [128, 128, 128, 128]; G.firstblock_res = 16;
# Number of GPUs.
desc += '-1gpu'
submit_config.num_gpus = 1
sched.minibatch_base = 4
sched.minibatch_dict = {
4: 128,
8: 128,
16: 128,
32: 64,
64: 32,
