call this function when @denoise_fn module is hooked(a module hook that saved intermediate feature map) Args: x (

The noise predicition model function that is used for DPM-Solver.

Multistep solver DPM-Solver-2 from time `t_prev_list[-1]` to time `t`. Args: x: A pytorch tensor. The initial value at t

Multistep solver DPM-Solver-3 from time `t_prev_list[-1]` to time `t`. Args: x: A pytorch tensor. The initial value at t

norm input to [-1, 1] Args: x (torch.Tensor): input Returns: torch.Tensor: output with data ranging in [-1, 1]

KL divergence between normal distributions parameterized by mean and log-variance.

calculate PSNR for one image :param img_gt: ground truth image, numpy array, shape [H, W, C] :param img_test: test or inference image, nu

Singlestep DPM-Solver with the order `order` from time `s` to time `t`. Args: x: A pytorch tensor. The initial value at

Create a list of timesteps to use from an original diffusion process, given the number of timesteps we want to take from equally-size

unnormalized input to data range [0, 1] Args: x (Tensor): input data ranging in [-1, 1] Returns: Tensor: output data ranging

tensorboard logger Args: place (str, optional): place to save tb logging. Defaults to './tb_runs/'. file_logger_name

:param d: h5py.File or dict :param aug_prob: augmentation probability :param hp: high pass for ms and pan. x = x - cv2.boxFi

Create a wrapper class for the forward SDE (VP type). *** Update: We support discrete-time diffusion models by implementing a picewis

Construct a DPM-Solver. We support both DPM-Solver (`algorithm_type="dpmsolver"`) and DPM-Solver++ (`algorithm_type="dpmsolver++"`).