mne_rsa.rdm_array#

class mne_rsa.rdm_array(X, patches=None, dist_metric='correlation', dist_params={}, y=None, labels=None, n_folds=1, n_jobs=1)[source]#

Generate RDMs from an array of data, possibly in a searchlight pattern.

First use searchlight to compute the searchlight patches. Then you can use this function to compute RDMs for each searchlight patch.

Parameters:

Xndarray, shape (n_items, n_series, n_times): An array containing the data.
patchesgenerator of tuples | None: Searchlight patches as generated by searchlight. If None, no searchlight is used. Defaults to None.
dist_metricstr | function: The distance metric to use to compute the RDMs. Can be any metric supported by scipy.spatial.distance.pdist(). When a function is specified, it needs to take in two vectors and output a single number. See also the dist_params parameter to specify and additional parameter for the distance function. Defaults to ‘correlation’.
dist_paramsdict: Extra arguments for the distance metric used to compute the RDMs. Refer to scipy.spatial.distance for a list of all other metrics and their arguments. Defaults to an empty dictionary.
yndarray of int, shape (n_items,) | None: Deprecated, use labels instead. For each item, a number indicating the class to which the item belongs. Defaults to None, in which case labels is used.
labelslist | None: For each element in X (=the first dimension), a label that identifies the item to which it corresponds. Multiple elements in X may correspond to the same item, in which case they should have the same label and will either be averaged when computing the data RDM (n_folds=1) or used for cross-validation (n_folds>1). Labels may be of any python type that can be compared with == (int, float, string, tuple, etc). By default (None), the integers 0:len(X) are used as labels.

Added in version 0.10.
n_foldsint | sklearn.model_selection.BaseCrollValidator | None: Number of cross-validation folds to use when computing the distance metric. Folds are created based on the y parameter. Specify None to use the maximum number of folds possible, given the data. Alternatively, you can pass a Scikit-Learn cross validator object (e.g. sklearn.model_selection.KFold) to assert fine-grained control over how folds are created. Defaults to 1 (no cross-validation).

Attributes:

shapetuple of int

Multidimensional shape of the generated RDMs.

This is useful for re-shaping the result obtained after consuming the this generator.

For a spatio-temporal searchlight:: Three elements: the number of time-series, number of time samples and length of a consensed RDM.
For a spatial searchlight:: Two element: the number of time-series and length of a condensed RDM.
For a temporal searchlight:: Two elements: the number of time-samples and length of a condensed RDM.
For no searchlight:: One element: the length of a condensed RDM.

Yields:

rdmndarray, shape (n_patches, n_items * (n_items - 1)): A RDM (in condensed form) for each searchlight patch.