GaussianDenoiserSolver.py 🚧

Source: LION/optimizers/GaussianDenoiserSolver.py

Warning

This file has not yet received a complete narrative and docstring audit. Its public source-level API is listed automatically below.

class GaussianDenoiserSolver(model: LIONmodel, optimizer: Optimizer, loss_fn, geometry: Geometry = None, verbose: bool = False, model_regularization=None, device: torch.device = None, save_folder: str = None, noise_level: np.ndarray = np.array([0.05]))

No docstring is available.

GaussianDenoiserSolver.mini_batch_step(self, sino, target)

This function isresponsible for performing a single mini-batch step of the optimization. returns the loss of the mini-batch

GaussianDenoiserSolver.default_parameters() → SolverParams

No docstring is available.

GaussianDenoiserSolver.set_patch_strategy(self, n_patches, patch_size)

Set the patch strategy for the solver. :param n_patches: Number of patches to extract. :param patch_size: Size of each patch.

GaussianDenoiserSolver.validate(self)

This function is responsible for performing a single validation set of the optimization. returns the average loss of the validation set this epoch.