def generate_default_image_optim_loop_processing_asset(
root, file="default_image_optim_loop_processing"
):
torch.manual_seed(0)
input_image = torch.rand(1, 3, 32, 32)
criterion = TotalVariationOperator()
def get_optimizer(image):
from torch.optim import Adam
return Adam([image.requires_grad_(True)], lr=1e-1)
num_steps = 5
preprocessor = CaffePreprocessing()
postprocessor = CaffePostprocessing()
_generate_default_image_optim_loop_asset(
path.join(root, file),
input_image,
criterion,
get_optimizer=get_optimizer,
num_steps=num_steps,
preprocessor=preprocessor,
postprocessor=postprocessor,
)
def generate_default_image_pyramid_optim_loop_asset(
root, file="pyramid_image_optimization"):
torch.manual_seed(0)
input_image = torch.rand(1, 3, 32, 32)
criterion = TotalVariationOperator()
pyramid = ImagePyramid((16, 32), 3)
def get_optimizer(image):
from torch.optim import Adam
return Adam([image.requires_grad_(True)], lr=1e-1)
_generate_default_image_pyramid_optim_loop_asset(
path.join(root, file),
input_image,
criterion,
pyramid,
get_optimizer=get_optimizer,
)
def generate_default_image_optim_loop_asset(root, file="image_optimization"):
torch.manual_seed(0)
input_image = torch.rand(1, 3, 32, 32)
criterion = TotalVariationOperator()
def get_optimizer(image):
from torch.optim import Adam
return Adam([image.requires_grad_(True)], lr=1e-1)
num_steps = 5
_generate_default_image_optim_loop_asset(
path.join(root, file),
input_image,
criterion,
get_optimizer=get_optimizer,
num_steps=num_steps,
)
def test_PerceptualLoss(self):
op = TotalVariationOperator()
required_components = {"content_loss", "style_loss"}
all_components = {*required_components, "regularization"}
for components in powerset(all_components):
if not set(components).intersection(required_components):
with self.assertRaises(RuntimeError):
loss.PerceptualLoss()
continue
perceptual_loss = loss.PerceptualLoss(
**{component: op
for component in components})
for component in components:
self.assertTrue(getattr(perceptual_loss, f"has_{component}"))
self.assertIs(getattr(perceptual_loss, component), op)
for component in all_components - set(components):
self.assertFalse(getattr(perceptual_loss, f"has_{component}"))