obob_mne.decoding.GeneralizedTemporalArray¶

class obob_mne.decoding.GeneralizedTemporalArray(scores_raw, scoring_name, n_classes, c_factors_training, weights, info, times_training, tmin, times_testing=None, c_factors_testing=None, nave=1, nave_testing=None)[source]¶

Base class for Temporal Generalization Decoding.

Parameters:

scores_raw (numpy.ndarray shape (n_times, n_times) or (n_folds, n_times, n_times)) – Classifier accuracies.
scoring_name (str) – Name of the scoring function (i.e. Accuracy…)
n_classes (int) – Number of classes of the classification
c_factors_training (str) – Name of the factors over which was collapsed in the training set
weights (numpy.ndarray shape (n_channels, n_times)) – Classifier weights
info (dict) – Info dict
times_training (numpy.ndarray shape (n_times)) – Array of times in seconds of the training data
tmin (int) –
???
times_testing (None or numpy.ndarray shape (n_times)) – Array of times in seconds of the testing data. If None, it is copied from times_training.
c_factors_testing (None or str) – Name of the factors over which was collapsed in the testing set. If None, it is copied from c_factors_training.
nave (int) – Number of epochs in the training set.
nave_testing (int or None) – Number of epochs in the testing set. If None, it is copied from nave.

__init__(scores_raw, scoring_name, n_classes, c_factors_training, weights, info, times_training, tmin, times_testing=None, c_factors_testing=None, nave=1, nave_testing=None)[source]¶

Methods

`__init__`(scores_raw, scoring_name, …[, …])
`diagonal_as_temporal`()	Return the non-generalized results.
`drop_channels`(chs)	Drop the channels from the weights.
`get_temporal_from_training_interval`(tmin, tmax)	Average the scores of a training interval.
`plot_scores`([axes, show, cmap, colorbar, …])	Plot the scores as a Matrix.

Attributes

`chance_level`	The chance level of the classifier.
`info`	`Measurement info`
`nave`	Number of epochs in the training set.
`nave_testing`	Number of epochs in the testing set.
`scores`	The scores of the classification
`tmin`	tmin
`weights`	The classifier weights