Deep learning for NeuroImaging in Python.

Note

This page is a reference documentation. It only explains the class signature, and not how to use it. Please refer to the gallery for the big picture.

class nidl.estimators.base.BaseEstimator(random_state: int | None = None, ignore: Sequence[str] | None = None, **kwargs)[source]¶

Base class for all estimators in nidl.

Basicaly, this class is a LightningModule with embeded Trainer parameters. It defines:

a fit method.
a transform or predict method if the child class inherit from a valid Mixin class.

This class also provides a way to connect a custom DataLoader to a predefined estimator. Using the set_batch_connector method allows you to pass a function that reorganizes your batch of data according to the estimator’s specifications.

Init class.

Parameters:

random_state : int, default=None

When shuffling is used, random_state affects the ordering of the indices, which controls the randomness of each batch. Pass an int for reproducible output across multiple function calls.

ignore : list of str, default=None

Ignore attribute of instance nn.Module.

kwargs : dict

Trainer parameters.