ConnectivityMeasure¶
Estimate the specified connectivity measure from a previously computed MVAR model.
This is a generic node that can compute various connectivity measures that are also separately available in the connectivity category. Like the other nodes computing dynamical measures, this node accepts as input a multi-variate auto-regressive ( MVAR) model, or, more commonly, a time series of such models. Such models are typically estimated using a model-fitting node such as the Group LASSO MVAR node, the output of which can be directly used by this node (see documentation of that node for proper usage). The output of this node is a square matrix that quantifies connectivity between pairs of brain sources for each frequency bin of interest, and possibly for each time point if the input data were a time series (then yielding a 4-way tensor as output). The desired frequencies can be selected via a parameter. In addition to dDTF, this node can also compute the short-time dDTF (sdDTF) using the normalization parameter. More Info... Version 0.8.0
Ports/Properties¶
data¶
Data to process.
measure¶
The connectivity measure to compute. Measures include the directed transfer function (DTF) and its variants the direct directed transfer function (dDTF), the normalized DTF (nDTF), the short-time DTF (sdDTF) and full-frequency DTF (ffDTF), coherence (Coh), the partial coherence (pCoh), the imaginary coherence (iCoh), the modified coherence (mCoh), partial directed coherence (PDC) and its variants the normalized PDC (nPDC), the generalized PDC (GPDC), Granger-Geweke causality (GGC), the bivariate GGC (bGGC), and the spectral density (Spec).
frequencies¶
Frequencies over which to compute the measure, in Hz. This is a list of frequencies (e.g., [1,2,3,4]), or a range expression such as 1...4, or a half-open Python-style range such as 1:5. The node will output one connectivity matrix for each frequency, formatted as a tensor.
absolute_value_squared¶
Compute squared magnitude of complex values. If disabled the raw complex values will be returned, i.e., including some phase information.
include_feature_axis¶
Include a feature axis. If enabled, a one-element (dummy) feature axis will be appended that has the name of this measure set as the name of the feature. This way, the output of multiple dynamical-measure nodes can be concatenated along the feature axis, and the resulting tensor will retain the names of the measures in its resulting (multi-element) feature axis.
remove_auto_connections¶
Set auto-connectivity estimates to zero.
denominator_epsilon¶
Small value to add to denominators to avoid division by zero.
regularization¶
Regularization parameter to add to matrices that need to be inverted.
time_reversed¶
Compute time-reversed connectivity. This is a simple analytical reversal unless two MVAR solutions (one forward, one backward are provided).
measure_in_streamname¶
Include the measure name in the output stream name.
set_breakpoint¶
Set a breakpoint on this node. If this is enabled, your debugger (if one is attached) will trigger a breakpoint.
metadata¶
User-definable meta-data associated with the node. Usually reserved for technical purposes.