Accumulates a consistently ordered list of all the thinkers in the dataset. It is this order that any automatic segmenting through :p

This creates a dataset using the config values. If tlen and tmin are specified in the config, creates epoched dataset, otherwise Raw.

Collects multiple recordings of the same person, intended to be of the same task, at different times or conditions. Paramete

Collect all the targets (i.e. labels) that this Thinker's data is annotated with. Returns ------- targets: np.ndarra

All the trainable parameters in the Trainable. This includes any architecture parameters and meta-parameters. Returns ------

Updates which ids are to be returned by the dataset. If any argument is `None` it preserves the previous value. Parameters -

Given the inputs to and outputs from underlying modules, calculate the loss. Returns ------- Loss : Sin

Creates a Raw instance from `create_basic_data` Returns: ------- raw : mne.io.Raw

This *generates* a "Leave-one-subject-out" (LOSO) split. Tests each person one-by-one, and validates on the previous (the first is va

This is an optional method that indicates the transformation modifies the sampling frequency of the underlying time-series.

A copy of this object to allow the repetition of recordings, thinkers, etc. that load data from the same memory/files but have their

This provides a standard way to copy models, weights and all.

Given a dataset, return the proportion of each target class and the counts of each class type Parameters ---------- dataset Ret

Generate and return any transform associated with this preprocessor. Should be used after applying this to a dataset, i.e. through :m

This *generates* a "Leave-multiple-subject-out" (LMSO) split. In other words X-fold cross-validation, with boundaries enforced at thi

Applies a preprocessor to the dataset Parameters ---------- preprocessor : Preprocessor A pre

Validate that all the channel sets are consistent, return false if not

This allow the addition of a learning rate schedule to the process. By default, a linear warmup with cosine decay will be used. Any s

Split the thinker's data into training, validation and testing sets. Parameters ---------- test_frac : float

Batch transforms are operations that are performed on trial tensors after being accumulated into batches via the :meth:`__call__` met

Checks if multiple outputs were provided, and selects

Any args that make more sense as a convenience function to be set

Given the inputs to and outputs from the underlying module. Return tracked metrics. Parameters ---------- inputs :

Calculate and return metrics for a dataset Parameters ---------- dataset: DN3ataset, DataLoader The

sklearn/keras-like convenience method to simply proceed with training across multiple epochs of the provided dataset Paramet

Given a batch of inputs, return the outputs produced by the trainable module. Parameters ---------- inputs :

In many cases, the features learned by a model in one domain can be applied to another case. This method freezes (or un-freezes) all

Maps channel names to the Deep1010 format, will automatically map EOG and extra channels if they have been named according to standard conven

Create a snapshot of what is being currently trained for re-laoding with the :py:meth:`load_best()` method. Returns -------

Parses dataset entries in DN3 config Parameters ---------- name : str The name of the dataset specifi

Generate negative samples to compare each sequence location against