Spaces:

yslan
/

LN3Diff_I23D

Runtime error

LN3Diff_I23D / sgm /modules /diffusionmodules /loss_weighting.py

NIRVANALAN

init

11e6f7b about 2 years ago

855 Bytes

from abc import ABC, abstractmethod

import torch


class DiffusionLossWeighting(ABC):
    @abstractmethod
    def __call__(self, sigma: torch.Tensor) -> torch.Tensor:
        pass


class UnitWeighting(DiffusionLossWeighting):
    def __call__(self, sigma: torch.Tensor) -> torch.Tensor:
        return torch.ones_like(sigma, device=sigma.device)


class EDMWeighting(DiffusionLossWeighting):
    def __init__(self, sigma_data: float = 0.5):
        self.sigma_data = sigma_data

    def __call__(self, sigma: torch.Tensor) -> torch.Tensor:
        return (sigma**2 + self.sigma_data**2) / (sigma * self.sigma_data) ** 2


class VWeighting(EDMWeighting):
    def __init__(self):
        super().__init__(sigma_data=1.0)


class EpsWeighting(DiffusionLossWeighting):
    def __call__(self, sigma: torch.Tensor) -> torch.Tensor:
        return sigma**-2.0

	from abc import ABC, abstractmethod

	import torch


	class DiffusionLossWeighting(ABC):
	@abstractmethod
	def __call__(self, sigma: torch.Tensor) -> torch.Tensor:
	pass


	class UnitWeighting(DiffusionLossWeighting):
	def __call__(self, sigma: torch.Tensor) -> torch.Tensor:
	return torch.ones_like(sigma, device=sigma.device)


	class EDMWeighting(DiffusionLossWeighting):
	def __init__(self, sigma_data: float = 0.5):
	self.sigma_data = sigma_data

	def __call__(self, sigma: torch.Tensor) -> torch.Tensor:
	return (sigma2 + self.sigma_data2) / (sigma * self.sigma_data) ** 2


	class VWeighting(EDMWeighting):
	def __init__(self):
	super().__init__(sigma_data=1.0)


	class EpsWeighting(DiffusionLossWeighting):
	def __call__(self, sigma: torch.Tensor) -> torch.Tensor:
	return sigma**-2.0