def generator_factory(type):
if type == 'naive':
return gan_cls.generator()
elif type == 'gan':
return gan_cls.generator()
elif type == 'wgan':
return wgan_cls.generator()
elif type == 'vanilla_gan':
return gan.generator()
elif type == 'vanilla_wgan':
return wgan.generator()
def generator_factory(type):
if type == 'gan':
return gan_cls.generator()
elif type == 'wgan':
return wgan_cls.generator()
elif type == 'vanilla_gan':
return gan.generator()
elif type == 'vanilla_wgan':
return wgan.generator()
elif type == 'inverse_gan':
return gan_inverse.generator_inverse()
elif type == 'stage2_gan':
return gan_cls.generator2()
def generator_factory(type, dataset, b_size, h, scale_size, num_channels):
if type == 'gan':
return gan_cls.generator()
elif type == 'wgan':
return wgan_cls.generator(dataset=dataset)
elif type == 'vanilla_gan':
return gan.generator()
elif type == 'vanilla_wgan':
return wgan.generator()
elif type == 'began':
return began.Decoder(b_size, h, scale_size, num_channels)
elif type == 'acgan':
return segan.generator()
#return acgan.generator()
elif type == 'segan':
return segan.generator()
def generator_factory(type):
if type == 'gan':
return gan_cls.generator()
elif type == 'stage2_gan':
return gan_cls.generator2()