ica¶
Module: missions.nodes.spatial_filtering.ica
¶
Independent Component Analysis variants
Inheritance diagram for pySPACE.missions.nodes.spatial_filtering.ica
:
ICAWrapperNode
¶
-
class
pySPACE.missions.nodes.spatial_filtering.ica.
ICAWrapperNode
(retained_channels=None, load_path=None, **kwargs)[source]¶ Bases:
pySPACE.missions.nodes.spatial_filtering.spatial_filtering.SpatialFilteringNode
Wrapper around the Independent Component Analysis filtering of mdp
This Node implements the unsupervised independent component analysis algorithm for spatial filtering.
- Parameters
retained_channels: Determines how many of the ICA pseudo channels are retained. Default is None which means “all channels”. For ICA channels, there is no sorting, hence data has to be reduced in the internal whitening beforehand.
(optional, default: None)
load_path: An absolute path from which the ICA filter is loaded. If not specified, this matrix is learned from the training data.
(optional, default: None)
Exemplary Call
- node : ICA parameters: retained_channels : 42
POSSIBLE NODE NAMES: - ICAWrapper
- ICAWrapperNode
- ICA
POSSIBLE INPUT TYPES: - TimeSeries
Class Components Summary
_execute
(data)Execute learned transformation on data. _train
(data[, label])Uses data to learn a decomposition into independent components. get_filter
()input_types
is_supervised
()Returns whether this node requires supervised training. is_trainable
()Returns whether this node is trainable. store_state
(result_dir[, index])Stores this node in the given directory result_dir. -
_execute
(data)[source]¶ Execute learned transformation on data.
Changes the base of the space in which the data is located so that the dimensions correspond to independent components
-
input_types
= ['TimeSeries']¶