imaginAIry/imaginairy/modules/sgm/diffusionmodules/loss.py
2023-11-22 17:20:08 -08:00

109 lines
3.6 KiB
Python

from typing import Dict, List, Optional, Tuple, Union
import torch
import torch.nn as nn
from imaginairy.modules.sgm.autoencoding.lpips.loss.lpips import LPIPS
from imaginairy.modules.sgm.encoders.modules import GeneralConditioner
from imaginairy.utils import instantiate_from_config
from imaginairy.vendored.k_diffusion.utils import append_dims
from .denoiser import Denoiser
class StandardDiffusionLoss(nn.Module):
def __init__(
self,
sigma_sampler_config: dict,
loss_weighting_config: dict,
loss_type: str = "l2",
offset_noise_level: float = 0.0,
batch2model_keys: Optional[Union[str, List[str]]] = None,
):
super().__init__()
assert loss_type in ["l2", "l1", "lpips"]
self.sigma_sampler = instantiate_from_config(sigma_sampler_config)
self.loss_weighting = instantiate_from_config(loss_weighting_config)
self.loss_type = loss_type
self.offset_noise_level = offset_noise_level
if loss_type == "lpips":
self.lpips = LPIPS().eval()
if not batch2model_keys:
batch2model_keys = []
if isinstance(batch2model_keys, str):
batch2model_keys = [batch2model_keys]
self.batch2model_keys = set(batch2model_keys)
def get_noised_input(
self, sigmas_bc: torch.Tensor, noise: torch.Tensor, input_tensor: torch.Tensor
) -> torch.Tensor:
noised_input = input_tensor + noise * sigmas_bc
return noised_input
def forward(
self,
network: nn.Module,
denoiser: Denoiser,
conditioner: GeneralConditioner,
input_tensor: torch.Tensor,
batch: Dict,
) -> torch.Tensor:
cond = conditioner(batch)
return self._forward(network, denoiser, cond, input_tensor, batch)
def _forward(
self,
network: nn.Module,
denoiser: Denoiser,
cond: Dict,
input_tensor: torch.Tensor,
batch: Dict,
) -> Tuple[torch.Tensor, Dict]:
additional_model_inputs = {
key: batch[key] for key in self.batch2model_keys.intersection(batch)
}
sigmas = self.sigma_sampler(input_tensor.shape[0]).to(input)
noise = torch.randn_like(input_tensor)
if self.offset_noise_level > 0.0:
offset_shape = (
(input_tensor.shape[0], 1, input.shape[2])
if self.n_frames is not None
else (input_tensor.shape[0], input.shape[1])
)
noise = noise + self.offset_noise_level * append_dims(
torch.randn(offset_shape, device=input_tensor.device),
input_tensor.ndim,
)
sigmas_bc = append_dims(sigmas, input_tensor.ndim)
noised_input = self.get_noised_input(sigmas_bc, noise, input_tensor)
model_output = denoiser(
network, noised_input, sigmas, cond, **additional_model_inputs
)
w = append_dims(self.loss_weighting(sigmas), input_tensor.ndim)
return self.get_loss(model_output, input_tensor, w)
def get_loss(self, model_output, target, w):
if self.loss_type == "l2":
return torch.mean(
(w * (model_output - target) ** 2).reshape(target.shape[0], -1), 1
)
elif self.loss_type == "l1":
return torch.mean(
(w * (model_output - target).abs()).reshape(target.shape[0], -1), 1
)
elif self.loss_type == "lpips":
loss = self.lpips(model_output, target).reshape(-1)
return loss
else:
msg = f"Unknown loss type {self.loss_type}"
raise NotImplementedError(msg)