# -- import packages: ---------------------------------------------------------
import neural_diffeqs
import torch
# -- import local dependencies: -----------------------------------------------
from . import base, mix_ins
# -- set type hints: ----------------------------------------------------------
from typing import List, Optional, Union
# -- lightning model: ---------------------------------------------------------
[docs]
class LightningODE(
mix_ins.BaseForwardMixIn,
base.BaseLightningDiffEq,
):
"""LightningODE"""
[docs]
def __init__(
self,
# -- ode params: ------------------------------------------------------
latent_dim: int = 50,
name: Optional[str] = None,
mu_hidden: Union[List[int], int] = [2000, 2000],
mu_activation: Union[str, List[str]] = "LeakyReLU",
mu_dropout: Union[float, List[float]] = 0.2,
mu_bias: bool = True,
mu_output_bias: bool = True,
mu_n_augment: int = 0,
sde_type: str = "ito",
noise_type: str = "general",
backend: str = "auto",
# -- general params: --------------------------------------------------
train_lr: float = 1e-4,
train_optimizer: torch.optim.Optimizer = torch.optim.RMSprop,
train_scheduler: torch.optim.lr_scheduler._LRScheduler = torch.optim.lr_scheduler.StepLR,
train_step_size: int = 10,
dt: float = 0.1,
adjoint: bool=False,
loading_existing: bool = False,
*args,
**kwargs,
) -> None:
"""
LightningODE
Extended description.
Parameters
----------
latent_dim : int, optional
Description. Default is 50.
name : str, optional
Description. Default is None.
mu_hidden : Union[List[int], int], optional
Description. Default is [2000, 2000].
mu_activation : Union[str, List[str]], optional
Description. Default is 'LeakyReLU'.
mu_dropout : Union[float, List[float]], optional
Description. Default is 0.2.
mu_bias : bool, optional
Description. Default is True.
mu_output_bias : bool, optional
Description. Default is True.
mu_n_augment : int, optional
Description. Default is 0.
sde_type : str, optional
Description. Default is 'ito'.
noise_type : str, optional
Description. Default is 'general'.
backend : str, optional
Description. Default is "auto".
train_lr : float, optional
Description. Default is 1e-4.
train_optimizer : torch.optim.Optimizer, optional
Description. Default is torch.optim.RMSprop.
train_scheduler : torch.optim.lr_scheduler._LRScheduler, optional
Description. Default is torch.optim.lr_scheduler.StepLR.
train_step_size : int, optional
Description. Default is 10.
dt : float, optional
Description. Default is 0.1.
adjoint : bool, optional
Description. Default is False.
version : str, optional
Description. Default is __version__.
loading_existing : bool, optional
Whether to load an existing model. Default is False.
Returns
-------
None
"""
super().__init__()
name = self._configure_name(name, loading_existing=loading_existing)
self.save_hyperparameters()
# -- torch modules: ---------------------------------------------------
self._configure_torch_modules(func=neural_diffeqs.NeuralODE, kwargs=locals())
self._configure_lightning_model(kwargs=locals())
def __repr__(self) -> str:
return "LightningODE"
