BaseLightningDiffEq#
BaseLightningDiffEq: Base class for lightning modules with differential equation solving capabilities.
This class serves as the base for lightning modules that involve solving differential equations. It provides methods for configuring the model, integrating initial value problems (IVPs), computing Sinkhorn divergence loss, defining custom steps, and implementing LightningModule methods.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.None#
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.__init__()#
Initialize the BaseLightningDiffEq object.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq._update_lit_diffeq_hparams()#
Update LightningDifferentialEquation hyperparameters.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq._configure_lightning_model()#
Configure lightning model with optimizer, scheduler, and additional components.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq._configure_torch_modules()#
Configure Torch modules for the model.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.PRETRAIN()#
Property indicating if the model is in pre-training mode.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq._INTEGRATOR()#
Property returning the integrator function based on hyperparameters.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.integrate()#
Integrate initial value problems (IVPs) using the specified integrator.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.compute_sinkhorn_divergence()#
Compute Sinkhorn divergence loss.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.forward()#
Abstract method for performing forward pass.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.step()#
Abstract method for performing a step in optimization.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.training_step()#
LightningModule method for training step.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.validation_step()#
LightningModule method for validation step.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.test_step()#
LightningModule method for test step.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.predict_step()#
LightningModule method for prediction step.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.configure_optimizers()#
Configure optimizers and schedulers for training.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.__repr__()#
Return the representation of the class.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq._configure_name()#
Configure the name of the class.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.__eq__(self, value, /)#
Return self==value.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.__format__(self, format_spec, /)#
Default object formatter.
Return str(self) if format_spec is empty. Raise TypeError otherwise.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.__ge__(self, value, /)#
Return self>=value.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.__getattribute__(self, name, /)#
Return getattr(self, name).
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.__gt__(self, value, /)#
Return self>value.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.__hash__(self, /)#
Return hash(self).
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.__init_subclass__()#
This method is called when a class is subclassed.
The default implementation does nothing. It may be overridden to extend subclasses.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.__le__(self, value, /)#
Return self<=value.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.__lt__(self, value, /)#
Return self<value.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.__ne__(self, value, /)#
Return self!=value.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.__new__(*args, **kwargs)#
Create and return a new object. See help(type) for accurate signature.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.__reduce__(self, /)#
Helper for pickle.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.__reduce_ex__(self, protocol, /)#
Helper for pickle.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.__sizeof__(self, /)#
Size of object in memory, in bytes.
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.__str__(self, /)#
Return str(self).
- scdiffeq.core.lightning_models.base._base_lightning_diffeq.BaseLightningDiffEq.__subclasshook__()#
Abstract classes can override this to customize issubclass().
This is invoked early on by abc.ABCMeta.__subclasscheck__(). It should return True, False or NotImplemented. If it returns NotImplemented, the normal algorithm is used. Otherwise, it overrides the normal algorithm (and the outcome is cached).