ConfigurableBaseEstimator¶

class lightkit.ConfigurableBaseEstimator(*, default_params=None, user_params=None, overwrite_params=None)[source]¶

Extension of the base estimator which allows to manage a single model that uses the lightkit.nn.Configurable mixin.

Parameters

default_params (dict[str, Any] | None) -- Estimator-specific parameters that provide defaults for configuring the PyTorch Lightning trainer. An example might be setting max_epochs. Overwrites the default parameters established by the base estimator.
user_params (dict[str, Any] | None) -- User-specific parameters that configure the PyTorch Lightning trainer. This dictionary should be passed through from a trainer_params init argument in subclasses. Overwrites any of the default parameters.
overwrite_params (dict[str, Any] | None) -- PyTorch Lightning trainer flags that need to be ensured independently of user-provided parameters. For example, max_epochs could be fixed to a certain value.

Methods

`load_attributes`	Loads the fitted attributes that are stored at the fitted path.
`save_attributes`	Saves the fitted attributes of this estimator.

Inherited Methods

`clone`	Clones the estimator without copying any fitted attributes.
`get_params`	Returns the estimator's parameters as passed to the initializer.
`load`	Loads the estimator and (if available) the fitted model.
`load_parameters`	Initializes this estimator by loading its parameters.
`save`	Saves the estimator to the provided directory.
`save_parameters`	Saves the parameters of this estimator.
`set_params`	Sets the provided values on the estimator.
`trainer`	Returns the trainer as configured by the estimator.

Attributes

`persistent_attributes`	Returns the list of fitted attributes that ought to be saved and loaded.
`model_`