backend.features

class backend.features.Feature

excluded_params = []

trainable = False

trained()

Returns whether this Feature has been trained Only necessary to implement if trainable is True

output_size(n_chan)

Return the width of this feature’s output vector The default is just the number of channels

get_editable_params()

Return dictionary of params that can be edited by the user

train(features, labels)

Train this feature - only called if trainable is True features: 3D np array with shape (number of timesteps in training set, buffer size, n_chan) labels: 1D np array with shape (n_dof)

process(data)

Compute this feature on the given data with shape: (buffer size, n_chan)

class backend.features.MAV

__init__()

Initialize self. See help(type(self)) for accurate signature.

process(data)

Compute this feature on the given data with shape: (buffer size, n_chan)

class backend.features.MFL

__init__()

Initialize self. See help(type(self)) for accurate signature.

process(data)

Compute this feature on the given data with shape: (buffer size, n_chan)

class backend.features.PosDevs

__init__()

Initialize self. See help(type(self)) for accurate signature.

process(data)

Compute this feature on the given data with shape: (buffer size, n_chan)

class backend.features.ZRC

__init__()

Initialize self. See help(type(self)) for accurate signature.

process(data)

Compute this feature on the given data with shape: (buffer size, n_chan)

class backend.features.WVL

__init__()

Initialize self. See help(type(self)) for accurate signature.

process(data)

Compute this feature on the given data with shape: (buffer size, n_chan)

class backend.features.SLC

__init__()

Initialize self. See help(type(self)) for accurate signature.

process(data)

Compute this feature on the given data with shape: (buffer size, n_chan)

class backend.features.Raw

__init__()

Initialize self. See help(type(self)) for accurate signature.

output_size(n_chan)

Return the width of this feature’s output vector The default is just the number of channels

process(data)

Compute this feature on the given data with shape: (buffer size, n_chan)